Meiobenthology The Microscopic Motile Fauna of Aquatic Sediments
Olav Giere
Meiobenthology The Microscopic Motile Fauna of Aquatic Sediments 2nd revised and extended edition with 125 Figures, 20 Tables, and 41 Information Boxes
Prof. Dr. Olav Giere Universität Hamburg Zoologisches Institut und Zoologisches Museum Martin-Luther-King-Platz 3 20146 Hamburg Germany
[email protected]
ISBN: 978-3-540-68657-6
e-ISBN: 978-3-540-68661-3
Library of Congress Control Number: 2008927365 © 2009 Springer-Verlag Berlin Heidelberg This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Cover design: WMX Design GmbH, Heidelberg Cover illustration: Two interstitial Amblyosyllis (Annelida: Syllidae); courtesy of Nathan W. Riser, Institute of Marine Science, Nahant, Mass., USA. Printed on acid-free paper 9 8 7 6 5 4 3 2 springer.com
To Gaby, to whom I owe it all.
Preface to the Second Edition
Also bestimmt die Gestalt die Lebensweise des Thieres, und die Weise zu leben sie wirkt auf alle Gestalten mächtig zurück. So the shape of an animal patterns its manner of living, likewise their manner of living exerts on the animals’ shape massive effects. goethe 1806: Metamorphose der Thiere
Encouraged by the friendly acceptance of the first edition and stimulated by numerous requests and comments from the community of meiobenthologists, this second edition updates my monograph on meiobenthology. The revised text emphasizes new discoveries and developments of relevance; it has been extended by adding chapters on meiofauna in areas not covered before, such as the polar regions, mangroves, and hydrothermal vents. As I attempted to keep up with the actual literature for the whole field of meiobenthos—taxonomy and ecology, marine and freshwater—I became a little discouraged upon noticing the flood of literature that had appeared in the few years after the publication of the first edition. Has there been a multiplication of new meiobenthologists or an inflation of their industrious efforts? How could I compile this plethora of new data; how to select, what to omit? The need to extract general information from the details, and to modify and amalgamate them within a greater context; this difficult “condensation” process was the key to my approach. It forced me to be selective, to focus on one goal: to write a readable compendium that will serve the interested biologist, the fellow benthologist and the student alike. Avoiding a style with constructions that are too sophisticated should also enhance the comprehension of those readers that are not natively familiar with the English language. Since the first edition, meiofaunal research has made, I believe, major progress in three general areas: (a) systematics, diversity, and distribution; (b) ecology, food webs vii
viii
Preface
and energy flow; and (c) environmental aspects, including studies of anthropogenic impacts. (a) In the area of systematics, diversity and distribution, molecular biological studies suggest that some of the “smaller” meiobenthic groups, such as Kinorhyncha, Gastrotricha and Rotifera, hold key positions in metazoan phylogeny, linking various invertebrate lines into new units (e.g., Ecdysozoa, Scalidophora, Cycloneuralia, Lophotrochozoa). Genetic fine-scale diversification has become an indispensable tool for understanding distribution processes and biogeographic patterns. With enhanced studies in exotic and remote areas, the meiobenthos continues to be a haven for the discovery of unknown animals, even of high taxonomic rank, e.g., Micrognathozoa. Reports on meiofauna from polar or tropical regions, deep-sea bottoms or hydrothermal vents were limited in the first edition due to the scarcity of pertinent studies. Recent comprehensive publications have now recognized these formerly exotic areas as being in the research mainstream, and are covered here in separate chapters. Problems of principal biological relevance, such as the study of distribution patterns or the relation of body size to distribution, have been tackled using meiofauna as tools. The high number of meiobenthic species found under even extreme or impoverished ecological conditions puts meiobenthos at the forefront of biodiversity and “census of life” studies. Taxonomic, functional and genetic diversity as influenced by ecological and/ or anthropogenic variables are widely acknowledged matters of concern. Molecular screening methods allow large numbers of species to be recorded upon expending reasonable effort. (b) Today, essays on aquatic environments mostly consider the relevant role of meiobenthos. Mucus agglutinations and microorganisms are increasingly recognized to be important components that structure the sediment texture and provide the basis for many meiobenthic food chains. Trophic fluxes can be followed using new techniques, such as by assessing isotopic signatures. Metabolic pathways visualized by fluorescence imaging enable us to broaden our limited knowledge of the physiology of meiobenthos. Combined with advanced statistics, such as multivariate analyses, we can achieve results that link meiobenthos to general ecological paradigms. (c) The reactions of biota to environmental threats are increasingly based on evaluations of the meiofauna, underlining their inherent advantages (small size, ubiquity, abundance). With improved processing and culturing methods, pollution experiments are now often based on meiobenthic animals, apply population dynamics and use micro-/mesocosm studies. Standardized bioassays include meiofauna and have become commercially available. The increased role of meiofauna in this field is reflected by new chapters on the impact of metal compounds and pesticides. The use of molecular techniques can alleviate the problem of rapid mass identification, e.g., in nematodes. All of these research fields tie meiobenthology closer to the “mainstream,” which should be a main goal of future meiobenthic research. If this second edition can
Preface
ix
synthesize these modern scientific achievements, meiobenthology could indeed play a key role in assessing the health of our environment, and will not just represent a playground for singular interests. Several comprehensive publications on meiobenthos published in the last few years are contributing to this goal. Of broad interest are monographic publications on freshwater meiobenthos (Hakenkamp and Palmer 2000; Hakenkamp et al. 2002; Robertson et al. 2000a; Rundle et al. 2002). The new edition of the classic treatise Methods for the Study of Marine Benthos (Eleftheriou and McIntyre 2005) contains competent contributions to sediment analysis, sampling strategies and meiofauna techniques (Somerfield et al. 2005). It also covers statistical and analytical methods that assess ecosystem functioning and measure energy flow through benthic populations. Therefore, in this edition of Meiobenthology I have condensed the information in some chapters referring to “Methods for the Study of Marine Benthos.” Lesser known are the meiofauna reviews of Galhano (1970, in Portuguese) and Gal’tsova (1991, in Russian), which were not mentioned in the first edition. In other chapters of this edition (e.g., on polluted sites), the scope has been expanded by adding short accounts of the impacts of metals and pesticides on meiobenthos. The most conspicuous novelty is the highlighted boxes, which either contain the essence of a particular section or comment on special aspects. The figures have been redesigned for higher clarity, and some outdated paragraphs have been shortened or omitted. To maximize readability not all of the publications on which I drew are cited; on the other hand, on several occasions the same publication is cited in a different context in order to make the chapters independently readable and understandable. The resulting reference list is meant to provide an archive of detailed studies in all fields of meiobenthology. A comprehensive index and a glossary explaining specific terms facilitate the use of this book. Because of their ease of accessibility for the general reader, I accentuate references in widely distributed, English-dominated journals. As much as all this may help to improve the distribution and didactic impact of this book, I especially hope, for the sake of the student reader, that Springer-Verlag publishes this new edition at a competitive price that is affordable to all interested in the great world of small organisms. I hope that this edition will be considered as readable and received as warmly by the readers as the 1993 edition. Despite all the care that I have taken, I could not consider every contribution, and so I apologize especially to those colleagues who have published in less common native languages or in journals with restricted distributions, whose results have not been considered here. My particular regrets remain realizing how much valuable knowledge is “hidden” to most of us in the numerous publications that have appeared in Russian over the last few years, much of it unnoticed by many of us. Mistakes in the first edition, for which I apologize, have hopefully been eliminated. I regret and take the responsibility for remaining omissions or erroneous interpretations. Should this book draw the attention of benthic ecologists to the relevance of meiobenthos and foster further research in this field, it has accomplished its goals. Perhaps it represents the last chance to write a monographic textbook that amalgamates bits of information into a coherent context before electronic databases,
x
Preface
pictures and information networks produce a glut of innumerable details and publications—an information jungle in which the beginner especially can easily become lost. Meiobenthology is now increasingly represented on the Internet: the International Association of Meiobenthologists (I.A.M.) and also many colleagues have often designed comprehensive homepages with address and publication lists. New editions of the I.A.M. newsletter Psammonalia are regularly published online (http://www. meiofauna.org/) and include pictures and even short movie galleries. Also, CD-ROMs and databases of computer-based pictorial identification keys have attained increasing importance (European Limnofauna; European Register of Marine Species, ERMS; separate databases for Nematoda, Harpacticoida, Turbellaria). With this book I conclude many of my activities in meiobenthology. To express my feelings I could do worse than adopting the words of a good friend and protagonist of meiobenthos research, Prof. Bruce C. Coull, who upon his retirement wonderfully characterized his feelings and probably those of many other fellow meiobenthologists of our peer group: “I maintain an interest in all things meiofaunal and it has been a great life studying them. I hope that the next generation of researchers will learn much more about these creature friends and that the researchers have as much fun as I have had trying to understand our ubiquitous and omnipresent aquatic denizens.” Acknowledgements The second edition has been carefully proof-read again by my friend Robert P. Higgins (Ashville, NC, USA). His dedication and encouragement constantly accompanied me while writing this text. Important chapters have been kindly reviewed by two other good friends and experts, Bruce C. Coull (Columbia, SC, USA) and Walter Traunspurger (Bielefeld, Germany). I owe a large intellectual debt to all those many colleagues who invaluably helped me by sending literature, giving comments and, most importantly, kept encouraging me to complete this work. There are far too many to mention them all here by name. I thank Mrs. M. Hänel for her detailed drawings and particularly Mrs. A. Kröger (both Hamburg) for her most valuable and patient computer skills when designing the figures. Finally, Springer-Verlag (Heidelberg, Berlin) is to be thanked for its continuous interest in this project and its “author-friendly” support throughout the correspondence.
Hamburg, July 2008
Olav Giere
Preface to the First Edition
Studies on meiobenthos, the motile microscopic fauna of aquatic sediments, are gaining in importance, revealing trophic cycles and allowing the impacts of anthropogenic factors to be assessed. The bottom of the sea, the banks of rivers and the shores of lakes contain higher concentrations of nutrients, more microorganisms and a richer fauna than the water column. Calculations on the role of benthic organisms reveal that the “small food web”, i.e., microorganisms, protozoans, microphytobenthos, and smaller metazoans, play a dominant role in the turnover of organic matter (Kuipers et al. 1981). New animal groups—even those of high taxonomic status—are often of meiobenthic size and continue to be described. Two of the most recent animal groups ranked as phyla, the Gnathostomulida and the Loricifera, represent typical meiobenthos. Up to now, a textbook introducing the microscopic organisms of the sediments, their ecological demands and biological properties has not existed, despite the significance of meiobenthos indicated above. A recent book entitled Introduction to the Study of Meiofauna (Higgins and Thiel 1988) gives valuable outlines for practical investigation, and Stygofauna Mundi, a monograph edited by Botosaneanu (1986a), focuses on zoogeographical aspects of mainly freshwater forms, but neither was intended to be a comprehensive text on the subject of meiobenthology. The purpose of this book is to provide a general overview of the framework and the theoretical background of the scientific field of meiobenthology. The first of three major parts describes the habitat of meiobenthos and some of the methods used for its investigation; the second part deals with morphological and systematic aspects of meiofauna, and the third part reports on the meiofauna of selected biotopes and on community and synecological aspects of meiobenthos. However, a monographic text cannot include an adequate survey of general benthic ecology, or be a textbook on the zoology of microscopic animal groups. The primary purpose of this text is to provide an ecologically oriented scientific basis for meiobenthic studies. Further advice for practical investigations is found in important compilations by Higgins and Thiel (1988), Holme and McIntyre (1984), and Gray (1981). Hence, aspects of sampling procedures and strategies, statistical treatment and fauna processing will be treated here only briefly. In these fields, the present work should be considered a supplement to the books mentioned above and instead focuses on some critical hints, methodological limitations, and a few neglected practical aspects. xi
xii
Preface
Writing this book was particularly difficult because the literature on meiofauna is so widely dispersed in journals and congress proceedings and has so rapidly increased in volume that complete coverage is impossible. Regardless of my efforts, therefore, there is no pretence that this text is absolutely comprehensive. Where it is important for the general context, the major chapters of the book contain some overlap in terms of information. This is deliberate; it provides the reader with chapters that are complete in themselves and avoids the need for too many crossreferences. Also, in order to maintain a readable, coherent style, citations of specific references had to be restricted. Thus, the “reference list” of this text does not represent all of the sources drawn upon during the production of this book. The selection of topics and the emphasis given to them is admittedly subjective. In particular, the brief treatment of freshwater meiobenthos (Chapter 8.2) by no means reflects the exhaustive achievements and importance of this field of meiobenthology. This book does not include the nanobenthos, since this represents a microbiota that is completely different from the meiobenthos in its size range, methodology, and taxonomical composition (mainly prokaryotes, often autotrophic protists and fungi). Where appropriate, references compiled in a “Recommended reading” paragraph are given at the ends of many chapters. They will serve as supplementary information and, hopefully, will compensate for my own subjectivity. Should incorrect or misunderstood data be reported in the text, I would be most grateful to be informed of this. This book resulted from a series of lectures for advanced students given by the author over a period of several years at the University of Hamburg. Studying the tiny organisms living in sand and mud fascinated many of the students and provided the encouragement and persistent stimulus needed to write this book. It will achieve its goal if it further promotes interest in the diverse and cryptic microscopic world of meiobenthic animals, emphasizes their ecological importance, from both theoretical and practical viewpoints, and contributes to the awareness that small animals often play a key role in large ecosystems, which are becoming increasingly threatened. Acknowledgements I am deeply obliged to Dr. Robert P. Higgins (Washington, DC), who critically reviewed the entire text, and not only for linguistic flaws. My thanks go out to my graduate students who supported me in selecting figures and designing graphs. I am grateful to several of my colleagues for their valuable comments on parts of the text, and for providing me with manuscripts that were sometimes still in press and for other helpful hints. It was my intention to include only originals or redrawn figures. This was possible through the patient work of A. Mantel and M. Hänel (both in Hamburg), for which I am most grateful.
Hamburg, July 1993
Olav Giere
Contents
1
2
Introduction to Meiobenthology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1.1 1.2
Meiobenthos and Meiofauna: Definitions . . . . . . . . . . . . . . . . . . . . . A History of Meiobenthology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 2
The Biotope: Factors and Study Methods . . . . . . . . . . . . . . . . . . . . . . . .
7
2.1
Abiotic Factors (Sediment Physiography) . . . . . . . . . . . . . . . . . . . . . 2.1.1 Sediment Pores and Particles . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Granulometric Characteristics . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3 The Sediment–Water Regime . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.4 Physicochemical Characteristics. . . . . . . . . . . . . . . . . . . . . . . Biotic Habitat Factors: A Connected Complex . . . . . . . . . . . . . . . . . 2.2.1 Detritus and Particulate Organic Matter (POM). . . . . . . . . . . 2.2.2 Dissolved Organic Matter (DOM) . . . . . . . . . . . . . . . . . . . . . 2.2.3 Mucus, Exopolymers, and Biofilms . . . . . . . . . . . . . . . . . . . . 2.2.4 Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5 Microphytobenthos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.6 Higher Plants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.7 Animals Structuring the Ecosystem . . . . . . . . . . . . . . . . . . . . Conclusion: The Microtexture of Natural Sediments . . . . . . . . . . . . .
7 7 9 14 22 37 38 40 41 43 48 53 53 59
Sampling and Processing Meiofauna . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
3.1
63 63 64 72 72 73 77 80 84
2.2
2.3 3
Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1 Number of Replicates and Size of Sampling Units . . . . . . . . 3.1.2 Sampling Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Processing of Meiofaunal Samples. . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Preserving Meiofauna in Their Natural Void System . . . . . . . 3.2.2 Extraction of Meiofauna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.3 Fixation and Preservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.4 Processing and Identifying Meiofaunal Organisms . . . . . . . . 3.3 Extraction of Pore Water. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xiii
xiv
4
Contents
Biological Characteristics of Meiofauna . . . . . . . . . . . . . . . . . . . . . . . . .
87
4.1
87
Adaptations to the Biotope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Adaptations to Narrow Spaces: Miniaturization, Elongation, Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.2 Adaptations to the Mobile Environment: Adhesion, Special Locomotion, Reinforcing Structures . . . . 4.1.3 Adaptations to the Three-Dimensional Dark Environment: Static Organs, Reduction of Pigment and Eyes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.4 Adaptations Related to Reproduction and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
87 92
97 99
5 Meiofauna Taxa: A Systematic Account . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.1
5.2
5.3
5.4
5.5 5.6
5.7 5.8
Protista (Protoctista) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.1 Foraminifera (Rhizaria: Granuloreticulosa) . . . . . . . . . . . . . . 5.1.2 Heliozoa (Actinopodia). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.3 Amoebozoa (“Rhizopoda”): Gymnamoebea, Testacea. . . . . . 5.1.4 Ciliophora (Ciliata) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cnidaria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.1 Hydroida (Medusae) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.2 Hydroida (Polyps). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.3 Scyphozoa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.4 Anthozoa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Free-Living Platyhelminthes: Turbellarians . . . . . . . . . . . . . . . . . . . . 5.3.1 Major Turbellarian Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Distributional and Ecological Aspects . . . . . . . . . . . . . . . . . . Gnathifera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Gnathostomulida. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.2 Rotifera, Rotatoria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.3 Micrognathozoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nemertinea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nemathelminthes: A Valid Taxon? . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.1 Nematoda (Free-Living) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.2 Kinorhyncha . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.3 Priapulida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.4 Loricifera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.5 Gastrotricha. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tardigrada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Crustacea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8.1 Cephalocarida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8.2 Anostraca: Anomopoda (“Cladocera”; “Branchiopoda”) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8.3 Ostracoda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
103 103 107 107 108 114 116 116 118 118 119 120 123 127 127 129 133 134 136 137 156 158 160 162 165 171 172 173 175
Contents
5.9
5.10 5.11
5.12 5.13
5.14
5.15 5.16 5.17 5.18
5.19 6
xv
5.8.4 Mystacocarida. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8.5 Copepoda: Harpacticoida . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8.6 Copepoda: Cyclopoida and Siphonostomatoida . . . . . . . . . 5.8.7 Malacostraca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chelicerata: Acari . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.1 Halacaroidea: Halacaridae . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.2 Freshwater Mites: “Hydrachnidia,” Stygothrombiidae, and Others . . . . . . . . . . . . . . . . . . . . . . 5.9.3 Palpigradi (Arachnida) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9.4 Pycnogonida, Pantopoda. . . . . . . . . . . . . . . . . . . . . . . . . . . Terrigenous Arthropoda (Thalassobionts) . . . . . . . . . . . . . . . . . . . . Annelida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.11.1 Polychaeta. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.11.2 Oligochaeta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.11.3 Annelida “Incertae sedis” . . . . . . . . . . . . . . . . . . . . . . . . . . Sipuncula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mollusca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.13.1 Monoplacophora and Aplacophora. . . . . . . . . . . . . . . . . . . 5.13.2 Gastropoda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tentaculata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.14.1 Brachiopoda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.14.2 Bryozoa, Ectoprocta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kamptozoa, Entoprocta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinodermata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.16.1 Holothuroidea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chaetognatha . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tunicata (Chordata) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.18.1 Ascidiacea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.18.2 Sorberacea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Meiofaunal Taxa: Concluding Remarks . . . . . . . . . . . . . . . . . . . . . .
180 181 189 190 201 201 205 205 206 207 207 208 215 218 221 223 223 225 226 226 227 228 229 229 230 231 231 232 233
Evolutionary and Phylogenetic Effects in Meiobenthology . . . . . . . . . . 235 6.1 6.2
Body Structures of Evolutionary Relevance . . . . . . . . . . . . . . . . . . . 235 Meiofauna in the Fossil Record . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
7 Patterns of Meiofauna Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 7.1 7.2
7.3 7.4
Evolutionary Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zoogeographic Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Mechanisms of Dispersal . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.2 Geological Structures and Processes . . . . . . . . . . . . . . . . . Ecological Aspects of Distributional Importance: Horizontal Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vertical Zonation of Meiobenthos . . . . . . . . . . . . . . . . . . . . . . . . . .
243 249 250 256 259 261
xvi
8
Contents
Meiofauna from Selected Biotopes and Regions . . . . . . . . . . . . . . . . . . . 267 8.1 8.2
8.3
8.4
8.5 8.6 8.7
8.8
9
Polar Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1.1 Sea Ice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marine Subtropical and Tropical Regions . . . . . . . . . . . . . . . . . . . . . 8.2.1 Tropical Sands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.2 Mangroves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Deep-Sea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.1 The Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.2 The Meiofauna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dysoxic, Anoxic, and Sulfidic Environments: Discussing the Thiobios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Reducing Habitats of the Thiobios . . . . . . . . . . . . . . . . . . . . . 8.4.2 Thiobiotic Meiobenthos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.3 Survival of Thiobios Under Anoxia and Sulphide – Mechanisms and Adaptations. . . . . . . . . . . . . . . . 8.4.4 Food Spectrum of the Thiobios . . . . . . . . . . . . . . . . . . . . . . . 8.4.5 Distribution and Succession of the Thiobios . . . . . . . . . . . . . 8.4.6 Diversity and Evolution of the Thiobios. . . . . . . . . . . . . . . . . 8.4.7 Chemoautotrophy-Based Ecosystems: Vents, Seeps, and Other Exotic Habitats . . . . . . . . . . . . . . . . . . . . . . Phytal Habitats and Hard Substrates. . . . . . . . . . . . . . . . . . . . . . . . . . Brackish Water Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Freshwater Biotopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.1 Running Waters: Stream and River Beds . . . . . . . . . . . . . . . . 8.7.2 The Groundwater System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.3 Standing Waters, Lakes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Polluted Habitats. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.8.1 General Aspects and Method Survey . . . . . . . . . . . . . . . . . . . 8.8.2 Selected Cases of Pollution and Meiofauna . . . . . . . . . . . . . .
268 270 276 278 280 284 284 287 296 296 298 302 307 308 309 313 317 324 328 329 338 344 349 349 361
Synecological Perspectives in Meiobenthology . . . . . . . . . . . . . . . . . . . . 373 9.1 9.2 9.3
9.4
Community Structure and Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1.1 Processes of Recolonization . . . . . . . . . . . . . . . . . . . . . . . . . . Community Structure and Size Spectra . . . . . . . . . . . . . . . . . . . . . . . The Meiobenthos in the Benthic Energy Flow . . . . . . . . . . . . . . . . . . 9.3.1 General Considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.2 Assessing Production: Abundance, Biomass, P/B Ratio, Respiration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.3 The Energetic Divergence Between Meiofauna and Macrofauna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Position of Meiofauna in the Benthic Ecosystem: A Compilation of Energy Fluxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.1 The Meiofauna as Members of the “Small Food Web” . . . . . 9.4.2 Links Between the Meiofauna and the Macrofauna . . . . . . . . 9.4.3 Meiofauna as an Integrative Benthic Complex. . . . . . . . . . . .
373 375 377 383 383 387 397 400 402 406 410
Contents
xvii
10 Retrospect on Meiobenthology and Outlook on New Approaches and Future Research. . . . . . . . . . . . . . . . . . . . . . . 417 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
423
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
503
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
513
Chapter 1
Introduction to Meiobenthology
1.1
Meiobenthos and Meiofauna: Definitions
The terms “macrobenthos” and “microbenthos” were already well established when in 1942 Molly F. Mare coined the term “meiobenthos” to define an assemblage of benthic metazoans that can be distinguished from macrobenthos by their small sizes (note that the Greek “µειος” means “smaller”). Therefore, the study of meiobenthos per se is a relatively new component of benthic research, despite the fact that meiobenthic animals have been known about since the early days of microscopy. This book will mainly focus on metazoan meiofauna, which mirrors the author’s field of expertise. Hence, the term “meiobenthos” is used here synonymously to “meiofauna.” However, an ecological picture cannot be drawn without also considering relevant benthic protists (e.g., ciliates, foraminiferans, amoebozoans), and microalgae (e.g., diatoms). Today, members of the meiofauna are considered mobile and sometimes also haptosessile benthic animals, smaller than macrofauna but larger than microfauna (the latter term is now restricted mostly to Protozoa). The formal size boundaries of meiofauna are operationally defined, based on the standardized mesh width of sieves with 500 µm (1,000 µm) as upper and 44 µm (63 µm) as lower limits: all fauna that pass through the coarse sieve but are retained by the finer sieve during sieving are considered meiofauna. In a recent move, a lower size limit of 31 µm has been suggested by deep-sea meiobenthologists in order to quantitatively retain even the smallest meiofaunal organisms (mainly nematodes). Using biomass as a measure, meiofauna (in freshwater) have been defined to include all mobile benthic organisms with masses of between 2 and 20 µg (Hakenkamp et al. 2002). What began as an arbitrarily defined size-range of benthic invertebrates has since been supported by studies on the size spectra of marine benthic fauna. Quantitative sizetaxon studies (Schwinghamer 1981a; Warwick 1984; Warwick et al. 1986a; Duplisea and Hargrave 1996—see Sect. 9.2) infer that the (marine) meiofauna represent a separate biologically and ecologically defined group of animals, a concept well known in the case of the (interstitial) meiofauna of sands (Remane 1933, see Sect. 1.2). In addition to the “permanent” meiofauna, members of the “temporary” meiofauna belong to the meiofaunal size category only as newly settled larvae that later grow
O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
1
2
1 Introduction to Meiobenthology
to become macrofauna. An exact upper size limit that will be passed by these temporarily small organisms (often juvenile molluscs and annelids) is difficult to define. Meiofauna are mostly found in and on soft sediments, but also on and among epilithic plants and other hard substrates (e.g., animal tubes). Even the surfaces of barren rocks with their biofilm and detritus cover are suitable habitats. Under each footprint of moist shore sediment we often find 50,000–100,000 meiobenthic animals! Indeed, it is unclear why the meiobenthos was not recognized earlier as a valid intermediate between the micro- and the macrobenthos. It seems inconsistent with the fact that the microscopic fauna in the water column had long been considered an established faunistic assemblage. Personally, I believe that bare sand bottoms and beaches and the often odiferous muds were considered unlikely habitats for diverse fauna of minute dimensions. More detailed reading: Warwick (1989), Palmer et al. (2006), Rundle et al. (2002).
Box 1.1 Meiofauna, Meiobenthos: Definitions The term “meiofauna” denotes microscopically small, motile aquatic animals living mostly in and on soft substrates at all depths in the marine and freshwater realm. Although originally restricted to small metazoans, ecological connections suggest that larger protozoans (ciliates, amoebozoans) should also be included in the scope of meiofauna. In the context of this book, this wider definition is used synonymously with meiobenthos. Formally defined by sieve mesh sizes of between 44 and 500 mm, meiobenthos is increasingly considered an ecological unit of its own, an important link between microand macrobenthos. In contrast to permanent meiobenthos, the newly settled larvae of many macrobenthic animals are temporary meiofauna.
1.2
A History of Meiobenthology
Taxonomic descriptions and biological investigations of minute benthic animals were being published by the mid nineteenth century. One of the first of these was on the discovery of a minute aberrant mollusc, the aplacophoran Chaetoderma by Lovén in 1844, then described as a new worm genus, and the Kinorhyncha described by Dujardin in 1851. In 1901, Kovalevsky studied Microhedylidae (Gastropoda) in the Eastern Mediterranean, and in 1904, Giard described the first archiannelid Protodrilus from the coast of Normandy. He even stated that the microscopic fauna were so rich “that it would take years to study them.” However, these pioneers of meiofauna considered only isolated taxa—often the exceptional species of known invertebrate groups—not their ecological niches and community aspects.
1.2 A History of Meiobenthology
3
Since then, field investigations were biased towards commercially interesting macrofauna. Consequently, a suitable methodology for specifically sampling the smaller benthic animals had to be developed. It was Remane who first used finemeshed plankton nets to filter the “coastal ground water,” and he used dredges with sacks of fine gauze to perform equally pioneering studies of the microscopic fauna of (eulittoral) muddy bottoms (“pelos”) and of the small organisms associated with surfaces of aquatic plants (“phyton”) ). Remane summarized this work in a monograph entitled Verteilung und Organisation der benthonischen Mikrofauna der Kieler Bucht (1933), where he first used the word “Sandlückenfauna.” The corresponding term “interstitial fauna,” introduced by Nicholls (1935), comprised all animals living in interstices, not only those of meiobenthic size, e.g,. polychaetes in a pebble beach. Aside from his important descriptions of new kinds of animals, the significance of Remane’s work is reflected by his contention that the meiobenthic fauna of sand were not merely a loose aggregation of isolated forms, but “a biocoenosis different not only in species number and occurrence, but also in characteristics of form and function” (Figs. 8.11 and 8.12). In his 1952 paper, Remane embodied this concept in the word “Lebensformtypus,” which has since been incorporated into the terminology of general ecology. The ubiquity and complexity of this smaller benthos became much clearer with the development of effective grabs (Petersen 1913) and dredges (Mortensen 1925) for sampling subtidal bottoms. With improved methods (e.g., Moore and Neill 1930; Krogh and Spärck 1936), studies on the small benthos soon emerged from many parts of the world. From Remane’s school came numerous German scientists of considerable influence in meiofaunal research, e.g., Ax, Gerlach, Noodt, to name just a few. Through their work Remane’s stimulus even proliferated to further generations of meiobenthologists (Westheide, Schminke, Riemann) in Germany. From Britain, Moore (1930, 1931), Nicholls (1935) and Mare (1942) initiated the study of meiofauna. At the beginning of the 1960s Boaden and Gray were among the first to perform experiments with marine meiofauna. In 1969, McIntyre compiled the first review, Ecology of Marine Meiobenthos, which is still a valuable source of information, particularly for data on meiofauna from tropical areas. By studying the fauna of the Normandy coast of the Channel, the Swedish researcher Swedmark focused attention on the rich interstitial fauna, and described many hitherto unknown species. His review The Interstitial Fauna of Marine Sand (1964) is considered a classic among early meiofaunal literature. Working along the shores of the Mediterranean Sea, Delamare Deboutteville concentrated his research into the meiobenthos on the brackish transition areas between the marine and freshwater realms. He was the first to conduct meiofaunal research along the African shores. His book Biologie des Eaux Souterraines Littorales et Continentales (1960) is another much-esteemed compendium of meiofaunal research. What about North America, now one of the main centers of meiofaunal research? The early marine meiofaunal studies were linked to just a few names, e.g., Pennak, Sanders, and Zinn, who discovered important new crustacean groups. Some European scientists working in the US also contributed to the further development of this field: the studies of the Austrians Riedl, Wieser and Rieger in the 1950–1980s
4
1 Introduction to Meiobenthology
stimulated several American students to become meiobenthologists. The 1960s saw the beginning of American investigations directed primarily at ecology (e.g., Tietjen), which continue to be a major thrust of American meiobenthology, and are mostly concentrated along the Atlantic and Gulf coasts of the United States. Beginning in the 1970s the school of Coull began investigating the soft-bottom meiofauna, often addressing environmental problems (disturbance, predation, pollution) and using field experimental methods in estuarine soft bottoms. Its impact drew the attention of general marine benthologists to meiofauna. The development of meiobenthology in the freshwater realm went separate ways, used different methods, and even produced a separate nomenclature. Still now, research on freshwater meiobenthos is not well coupled with its marine counterpart, although both Remane and Delamare Deboutteville often emphasized the connections between marine and freshwater meiofauna, especially those of a zoogeographical and evolutionary nature. Similar to the situation in the marine field, important taxonomic work was performed early in the nineteenth century, especially on benthic freshwater copepods (e.g., the works of Sars, Claus, Lang, Gurney), but freshwater meiobenthology, as an ecological discipline, started later. It developed independently with the Russian Sassuchin and colleagues (1927), who sampled at a river shore. They first described the “psammon,” i.e., the fauna and flora of sand. Today, this term is specified as “mesopsammon,” the fauna between sand grains (= interstitial fauna of sands), in contrast to the mostly macrobenthic “epipsammon” (i.e., species that live burrowing in the sand) and “endopsammon” (species that live burroweing in the sand). Wiszniewski (1934) conducted similar studies in Polish rivers and lakes that emphasized the important role of rotifers (see Sect. 8.7). While in England, Germany, France and Belgium early papers on the freshwater psammon remained rather isolated and mainly taxonomic in nature, it was the American Pennak who included a wider faunal spectrum in his ecological and faunistic considerations. His monograph Ecology of the Microscopic Metazoa Inhabiting the Sandy Beaches of Some Wisconsin Lakes (1940) is one of the classic publications in freshwater meiobenthology. His ecological comparison of freshwater and marine interstitial fauna (1951) provided valuable insights into the characteristics of these two biomes, an approach later continued in the USA by Palmer and Strayer. Related to the research of Delamare Deboutteville were the investigations of Angelier (1953) on the river shores and banks in the south of France exposed during the dry season. Detailed granulometric and physiographic descriptions of the biotopes are a characteristic of this work. The importance of the hydrological regime was the subject of the meiobenthos studies by Ruttner-Kolisko (beginning in 1953) in Austrian mountain streams and rivers. In Switzerland Chappuis started a series of investigations (beginning in 1942) on the fauna of the groundwater. He found the “stygobios” to be a distinct faunal element (see Sect. 8.2.1). The “hyporheic” biotopes beneath streams and rivers were the research domain of Karaman (1935), Orghidan (1955) and collaborators. They were attracted by the interesting subterranean fauna of karstic rivers in
1.2 A History of Meiobenthology
5
Southeast Europe and contributed much to the early knowledge of cave meiobenthos, today also termed “troglobitic” fauna. From the 1960s Danielopol worked intensively on hyporheic and lacustrine meiobenthos, mainly in Austria. Although specializing in ostracods, he and his colleague Stock from Holland also focused on general evolutionary aspects, discussing the colonization pathways for subterranean habitats (see Sect. 8.7.2). The ecology of groundwater fauna has been well covered in a volume edited by Gibert et al. (1994). A summary of methods for studying freshwater meiofauna has been provided by Palmer et al. (2006). Based mainly on lake meiofauna, Rundle et al. (2002) provided a competent review of freshwater meiobenthos. Meiofauna of lotic ecosystems (streams) is covered in a special volume edited by Robertson et al. (2000a,c). Enhancing our insight into their similarities and differences will hopefully reduce the historical separation between marine and freshwater meiofaunal research. Today, several hundred scientists are working to expand our knowledge of meiofauna from alpine lakes to the deep-sea floor, from tropical reefs to polar sea ice. However, despite an increasing number of meiobenthologists working in Africa, South America, Asia and Australia, the meiobenthos in these continents is as yet largely unknown. Studies of the deep-sea meiobenthos gain increasing momentum with the development of sophisticated maneuverable vehicles. As in other biological sciences, the structure of meiobenthological research evolved from isolated and individualistic taxonomic descriptions to assessments of abundance and distribution principles worked out by teams. These were the foundation for ecological research that, after implementing sophisticated statistical methods, could tackle complex problems such as pathways of distribution, community functioning and the impact of disturbances. From there, studies on environmental effects and on anthropogenic disturbance and pollution using meiofauna as sentinels were a logical consequence. The future of meiobenthology (see Chap. 10) will largely depend on how well we understand how to incorporate the specific potentials of meiobenthic animals into mainstream benthic research. The adoption of molecular methods will decisively contribute to future development. We should address the importance of global climate change and advocate more strongly than before the value of using the ubiquitous and speciose meiofauna to assess the health of ecosystems. Most meiobenthologists are members of the International Association of Meiobenthologists (IAM) (http://www.meiofauna.org/) and thus receive its newsletter Psammonalia for information on current fields of interest, members’ research projects and recent literature. The triennial conferences of the IAM are important occasions for the mutual exchange of results, experiences and developments, and members from countries that are now starting to perform meiobenthic research are increasingly participating in these conferences. The website provides information on upcoming events, new results and the e-mail addresses of all of the members. Scientists from remote places that are often cut off from the mainstream of meiofaunal research can also use such electronic media to easily contact their colleagues and access recent literature. The development of electronic species registers, iden-
6
1 Introduction to Meiobenthology
tification guides, and expert lists (e.g., the European Register of Marine Species, ERMS; NEMYS) has enabled easier access in order to solve the diversity problem of meiofauna. Thus, due to the increasing “globalization” of meiofaunal research through new technical achievements, meiofaunal research will be better dispersed into areas hampered by their social or geographical isolation. More detailed reading: Remane (1933); Pennak (1940); Swedmark (1964); Delamare Deboutteville (1960); Ax (1966); Schwoerbel (1967); McIntyre (1969); Coull and Chandler (1992); Gibert et al. (eds. 1994); Robertson et al. (2000c); Rundle et al. (2002).
Box 1.2 Meiobenthology: A Young Research History Meiofaunal research, especially meiobenthic ecology, as initiated by Remane, is a fairly young field. Aside from singular and scattered early descriptions of strange tiny organisms, the field of marine meiofaunal research originated in the first decades of the twentieth century in Europe, starting with taxonomic and basic ecological work. More complex ecological approaches were characteristic of research carried out between 1960 and 1980 in Europe and particularly in the US. Freshwater studies began independently in eastern European rivers, Swiss streams, and North American lakes. Marine and freshwater studies of meiobenthos developed along different lines and only recently prompted the ecological parallels a common nomenclature. Reasons for the relatively late start of multidirectional meiofaunal research may include the inconspicuous nature of meiobenthic organisms and their unspectacular habitats. This may have confounded the real phylogenetic and ecological roles of meiofauna. Today, the International Meiofauna Association and its triennial conferences bring together work in all fields of meiofauna research and most scientists that are studying meiobenthos.
Chapter 2
The Biotope: Factors and Study Methods
2.1 2.1.1
Abiotic Factors (Sediment Physiography) Sediment Pores and Particles
When describing the habitats of meiofauna, grain size is a key factor since it directly determines spatial and structural conditions and indirectly determines the physical and chemical milieu of the sediment. Poorly sorted sediment particles (e.g., sand mixed with gravel and silt) become tightly packed and the interstitial pore volume is often reduced to only 20% of the total volume. Well-sorted (coarse) sediments contain up to 45% pore volume. According to RuttnerKolisko (1962), most field samples of unsorted freshwater sand have 40% pore volume. Aside from pore volume, the external surface area of the sediment particles is an important determinant of meiobenthic life. It directly defines the area available for the establishment of biofilms (mucus secretions of bacteria, fungi, diatoms, fauna), which, under natural conditions, form the matrix into which the sediment particles are embedded. Thus, particle surface is an important parameter for microscopic animal life. This internal surface is unbelievably large: for a 1-m3 stream gravel it has been calculated to amount to about 400 m2. One gram of dry fine sand with a median particle diameter of 63–300 µm may have a total surface area of 8–12.5 m2; if it consists mostly of diatom shells, this value can even exceed 20 m2, whereas for 1 g of coarse-grained calcareous sand a value of just 1.8 m2 was calculated (Suess 1973; Mayer and Rossi 1982). In addition to size, the grain shape also determines the sorting of the sediment. Angular, splintery particles are packed tighter than spherical ones. A higher angularity leads to more structural complexity, less water permeability and usually higher abundance of meiofauna (Fig. 2.1; see Conrad 1976). A direct correlation between pore dimensions and body size of meiofaunal animals has been demonstrated experimentally (Williams 1972). In general, mesobenthic species moving between the sand grains prefer coarse sands, while endo- and epibenthic ones are
O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
7
8
2 The Biotope: Factors and Study Methods
Fig. 2.1 The pore system in sediments consisting of grains with a round shape (glass beads; left) vs. natural grains of angular shape (right); note differences in pore space due to different packing. (After Conrad 1976; modified)
mostly encountered in fine to silty sediments. These sediment differences affect the two major groups of meiobenthos, nematodes and harpacticoids. The finer sediments are preferred by most nematodes, while coarser ones are often favored by harpacticoids (Coull 1985). Within the nematode taxon, the preference for a specific grain size was found to relate to certain ecological types (Wieser 1959a). “Sliders” live in the wide voids of coarse sand; below a critical median grain size of about 200 µm, the interstices become too narrow. Thus, fine sand and mud will be populated by “burrowers” (Fig. 2.2). The particle shape determines the colonization of the sediment by meiofauna through indirect action via water content and by permeability (Sect. 2.1.3). The colonization of sand by meiobenthos is also determined by the grain structure, the roughness of edges, and the shapes of grain surfaces and cracks. These are important parameters that structure the microhabitats of different bacterial colonies (Meadows and Anderson 1966). Sand grains with diameters of >300 µm frequently have more plain surfaces than smaller particles; they also have a different bacterial epigrowth. This diversification has been shown to attract different meiofauna (Marcotte 1986a, Watling 1988). Likewise, in comparative experiments, cores of different grain sizes have been colonized by different meiobenthos. This emphasizes the capacity of meiofaunal species to chose and “recognize” their preferred sediment (Boaden 1962; Gray 1965; Hadl et al. 1970; Vanreusel 1991). Although the direct structural impact of the sediment particles is mostly confounded by other factors, e.g., biofilms, water flow, etc. (see Table 2 in Snelgrove and Butman 1994), there are strong affinities of specific meiofauna for specific sediments (Schratzberger et al. 2004). The structure and dimensions of the pore system are also directly correlated with the anatomy of the inhabitants and the functions of their organs (Ax 1966; Lombardi and Ruppert 1982).
2.1 Abiotic Factors (Sediment Physiography) >400 µm >400 µm <300 µm
8
< 200 µm intertidal height (feet)
Fig. 2.2 Distribution pattern of meiofaunal locomotory groups in the intertidal of a sandy shore. Black areas in circles or squares relate to the number of species per sample (250 cm3) that belong to the same locomotor type; lines indicate areas of identical median grain size. (After Wieser 1959)
9
4
<100 µm 0 −2
stations
coarse
fine sediment burrowers
1-3
4-6
7-9
>10 sliders
2.1.2
Granulometric Characteristics
2.1.2.1 Grain Size Composition Grain size analysis is fundamental to all ecological aspects of meiobenthic work. Although the fractionation of the sediment into different size groups does not reflect the natural composition, it provides a basis for reference and an important comparative framework. Techniques of sediment analysis are well covered in Bale and Kenny (2005); only some additional practical hints are presented here. Granulometry is usually based on the rather tedious procedure of the fractionated sieving of a sufficiently large sample. Recently sieving has been replaced by electronic procedures (modified Coulter counters, laser diffraction counters) with higher accuracy and throughput. Inherent inaccuracies with sieving (underrepresentation in the finer fractions) are based on effects of the adhesion of particles to the mesh fibers (Logan 1993). Salt-containing marine samples are mostly wet sieved, especially when fecal pellets consolidate fine sediment. However, the faster technique of dry sieving is often preferred (80 °C, 24 h) and is sufficiently accurate if agglutination
10
2 The Biotope: Factors and Study Methods
is avoided and the salt content of the sample is corrected for. The inaccuracies involved in these procedures are acceptable for most ecological questions. The siltclay fraction (“mud content”) passing through the 63-µm or 44-µm sieve is an important ecological parameter that determines the biological and mechanical properties of the sediment, but is usually not differentiated any further. After sieving, its proportion is determined by the loss of weight. However, it can be refined by performing a fractionated analysis of the settling velocity using elaborate soil science methods. The mesh sizes of the sieve set usually follow a geometric series (Wentworth − log x with x = particle size in mm scale) with 1.0 or 0.5 ϕ (phi) intervals, where ϕ = log 2 (Wentworth 1922; Krumbein 1939). Commonly for meiofaunal studies a series of sieves are used with mesh sizes (mm) 1.0/0.5/0.25/0.125/0.063/0.044 (= 0/ + 1.0/ + 2.0/ + 3.0/ + 4.0/ + 4.5 ϕ units). Very small meiofauna (e.g., some nematodes) would even pass through the 0.044-mm sieve and can only be quantitatively retained using a 0.031-µm sieve (= + 5.0 ϕ). Some animals with a smaller diameter than the mesh width are always retained lengthwise on the screen despite the wide meshes. A correction factor has been calculated to account for this inaccuracy (Tseitlin et al. 2001). With the increasing use of electro-optical devices this problem is has reduced in importance. Electronic calculations and illustrations of particle size mean that the ϕ notation is losing relevance. The simple process of sieving has some pitfalls that can render the procedure needlessly tedious or misleading: (a) It is important to weigh the whole sample as soon as possible to ensure the correct determinations of water content and salinity (see below). If this treatment is not possible shortly after sampling, care must be taken to keep the fresh core in a tight bag to minimize the outflow of water and evaporation. (b) Massive shaking of water-unsaturated cores during transport (e.g., due to motor vibrations on boats!) should be avoided because this can alter sediment structure and water saturation considerably. (c) If a sediment core contains a few coarse pebbles or shells in otherwise relatively homogeneous and fine sediment, these should be removed. Since calculations of character indices depend solely on weight, one or two massive particles can completely change the granulometric curve without having a relevant impact on the meiofauna. I believe that this alteration of conditions is justified in biological studies, provided that the manipulation is mentioned in the text. Block histograms or ternary diagrams (triangular coordinates) are the usual methods used to illustrate particle size distribution (Krumbein 1939; Gray 1981; Bale and Kenny 2005). The relevant granulometric parameters can be computer-calculated using specific software (e.g., Gradistat; Blott and Pye 2001) or calculated by simple mathematical methods: the fractions are computed as cumulative percentages starting with the coarsest fraction. These values are listed for further mathematical treatment or plotted as cumulative frequency curves (Fig. 2.3). It is apparent
2.1 Abiotic Factors (Sediment Physiography)
11
that the use of the ϕ notation (abscissa) has the advantage of giving relatively more detailed information on the important finer particles, and it also produces equidistant intervals that are relevant for the assessment of the following important statistical indices. The grain size composition of a sample is characterized by a few statistical parameters (see Table 2.3 in Bale and Kenny 2005) which can be read directly from the diagram or calculated. These include the median (Md) and the first (Q1) and the third (Q3) quartiles. The Md value corresponds to the 50% point of the cumulative scale (ϕ 50), Q1 to ϕ 25 and Q3 to ϕ 75. These values indicate the average grain size and the spread (scatter) of the grain size fractions towards both ends. The spread distance is defined by the sorting coefficient and conveniently expressed by the Quartile Deviation QD ϕ =
ϕ25 − ϕ75 . 2
A homogeneous sediment with a small QD enclosing only a few phi-intervals between the quartiles is regarded as “well sorted” (Table 2.1). An ideally sorted sediment would consist just of one grain fraction and would thus have QD = 0.
100
100
siliceous sand
siliceous sand 50
calcareous sand
cumulative weight %
cumulative weight %
75
25
84 75 50 25 5 calcareous sand 0
0 −1 0 1 2000 1000 500
a
2
3
250
125
grain size
4
−2
>9ϕ
63 <2µm
−1 ±0 1 2 3 diameter in ϕ units
4
b
Fig. 2.3 a–b Granulometric analysis of two exposed Atlantic beaches. Open squares, Portugal; solid circles, Bermuda. a Cumulative frequency curves. b The same granulometric data plotted on probability paper
12
2 The Biotope: Factors and Study Methods Table 2.1 Sediment sorting classes (Gray 1981) Sorting class [ϕ] Classification of sediment <0.35 0.35–0.50 0.50–0.71 0.71–1.00 1.00–2.00 2.00–4.00 >4.00
Very well sorted Well sorted Moderately well sorted Moderately sorted Poorly sorted Very poorly sorted Extremely poorly sorted
The frequency curve will only attain a sigmoid shape if the sediment fractions tend to follow a normal distribution. However, it will become “skewed,” i.e., it will have an asymmetrical slope when certain fractions are over- or underrepresented. The degree of curve symmetry is measured by the ϕ Quartile Skewness: Sk ϕ =
(ϕ Q1+ϕ Q3) − ϕ Md. 2
The above indices are based only on very few ϕ values, and they tend to neglect the “tails” of the curve. More precise computations comprise a wider portion of the fraction, i.e., the mean, M= the “graphic mean” M z =
(ϕ 25+ϕ 75) 2
(ϕ 16 + ϕ 50 + ϕ 84) 3
the “inclusive quartile deviation or “inclusive sorting coefficient:” QD I =
ϕ 84 − ϕ 16 ϕ 95 − ϕ 5 + , 4 6.6
or the “inclusive graphic skewness:” Sk I =
ϕ 16 + ϕ 84 − 2ϕ 50 , 2(ϕ 84 − ϕ 16)
From the grain size composition analyzed above and plotted as curves in Fig. 2.3a,b, the characteristic granulometric values can be derived (Table 2.2). All granulometric indices (e.g., the values for the median or quartiles) can also be computed mathematically (Hartwig 1973b) from the listed size frequencies by interpolation or by using computer software. In biological papers it is more illustrative to convert ϕ values into metric units. The use of a conversion chart (Page 1955; Fig. 3.2 in Buchanan 1971) or computer software is often recommended, although calculation is just as easy. The calculation of ϕ from x [mm]: −log x/log 2; calculation of x [mm] from ϕ : x[mm] = 2−ϕ.
2.1 Abiotic Factors (Sediment Physiography)
13
Table 2.2 Characteristic granulometric indices for the sediment samples plotted in Fig. 2.3a,b Granulometric index Siliceous sand from Portugal Calcareous sand from Bermuda Median, Md Lower quartile, Q1 Upper quartile, Q3 Inclusive sorting coefficient, QDI Inclusive graphic skewness, SkI
0.4 ϕ = 740 µm −0.2 ϕ = 1140 µm 1.0 ϕ = 500 µm 0.93 = moderately sorted
1.3 ϕ = 410 µm 0.9 ϕ = 528 µm 1.8 ϕ = 285 µm 0.74 = moderately sorted
0.003
−0.089
The occurrence of certain sediment types varies depending on the local geological and physiographical conditions. In temperate and boreal regions siliceous sands prevail, while in the warmer regions and on seamounts inhomogeneous biogenic calcareous sediments with more complex surface structures dominate (see Sect. 8.2). Black basalt and lava sand can often be found in volcanic areas. The deep sea floor is usually muddy and fluffy (unconsolidated), often consisting of foraminiferan (mostly calcareous) or radiolarian (mostly siliceous) skeletons. In shallow seas, offshore bottoms will usually consist of medium sand while nearer to the shore currents attenuate and will allow fine sand and mud to settle. In areas where ripple marks indicate strong currents, crests contain coarser sediments than troughs, where fine sand and often flocculent surface layers with a higher content of organic material tend to accumulate. The fine sediment in seagrass beds, where currents are
Box 2.1 To See a World in a Grain of Sand The size, shape and composition of sediment particles interact via the water flux with the physical and chemical regime of the sediment, the exposure to currents and waves as well as the general geological setting. In this network of abiotic factors that influence the habitat of meiobenthos, grain size plays a dominant role and can serve as the integrative key factor that characterizes the habitat of meiobenthos. Although we now know that communities and zones are not defined only by grain size composition, and that the differentiating factors are instead chemically and biologically controlled, granulometry remains an important foundation. Angular grains are packed tighter than round ones, but splintery, uneven surfaces are better for microbe colonization. Sediments with smaller grains offer less interstitial space and are preferred by different meiofaunal species to those in coarser sands. In general, the void system of sediments accounts for 20–45% of the total sediment volume. Careful granulometry should form the basis for every benthic ecological study. Modern data processing programs enable the relevant granulometric parameters, such as median, mean, quartiles, sorting coefficient and kurtosis, to be calculated automatically. These describe the granulometric basis for the living conditions of the biota and allow for abiotic structural comparisons.
14
2 The Biotope: Factors and Study Methods
weak, is enriched with leaf detritus. Near the shoreline the sediment structure may vary rapidly due to irregular water agitation, sedimentation and resuspension of shore vegetation and wrack material. These various sediment structures all represent different microhabitats for meiobenthic animals (Eckman 1979; Hogue and Miller 1981; Hicks 1989).
2.1.3
The Sediment–Water Regime
2.1.3.1 Exposure, Sediment Agitation, and Erosion Largely determined by the impacts of waves and currents, the exposure of a habitat is of eminent importance for the agitation and sorting of sediment particles, the flow of sediment water and fluxes of nutrients. Current velocity, sediment agitation and sorting interact in a complex way with the weights and surface structures of the particles and determine particle deposition and packing. These factors, in combination, control the “exposure” of a site, but a direct measurement of exposure is too complicated mathematically and instrumentally to be used by most biologists. Thomas (1986) and Hummon (1989) estimated the exposures of sandy shores from a fetch-energy index which was calculated using wave height and shoreline configuration, parameters which can be extrapolated from maps and data sheets. Eleftheriou and Nicholson (1975), on the basis of granulometry, discriminated exposed beaches from sheltered and semi-exposed ones via a critical median grain size of 230 µm. McLachlan (1980, 1989) attempted to create a general rating system for beaches based on a set of parameters including the height of the incoming waves. Muus (1968) and Doty (1971) related exposure to the weight loss of plaster test blocks distributed in/on the sediment. The dissolution of calcium sulfate was considered to be proportional to the velocity of the surrounding water currents, thus reflecting the exposure of the habitat. Similarly, Craik (1980) tried to derive the relative degree of (massive) exposure from the long-term scouring of cement blocks. Valesini et al. (2003) based their assessment of exposure on a set of seven quantitative environmental variables (e.g., fetch, steepness of shore, width of beach), which they analyzed using multivariate statistics, and classified several groups of beaches. However, in practice and in studies dealing with heterogeneous sites and topics, this computer-based grouping appears rather complex. Exposure remains a more or less summative often even subjective factor. Hence, benthologists are well advised to include a significant amount of comparative experience when deriving any measurements of the rate of exposure. Current velocity is not directly proportional to agitation and erosion. Turbulent water currents reduce particle suspension (McNair et al. 1997); particles with a diameter of approximately 180 µm are most easily eroded (Sanders 1958). A threshold of around 200 µm, earlier defined as a “critical grain size” for the occurrence of many animals (see Sect. 2.1.1), is of prime importance for the water contents of sediments. The lower average grain size threshold for the existence of an interstitial
2.1 Abiotic Factors (Sediment Physiography)
15
assemblage is often reported to be 150 µm. In freshwater sediments, 250 µm has been considered the size limit for the circulation of interstitial water (Rutter– Kolisko 1961). Neither tightly packed silt nor permanently agitated coarse sand offer favorable conditions for most meiofauna. In the rigid hydrographic regime of a North Sea estuary, increasing tidal ranges and current surges reduced nematode diversity in the sediment, while the biomasses of many species increased (Smol et al. 1994). Most but not all meiofauna react to strong currents and water surges by attempting to escape through downward migration (Steyaert et al. 2001; Sedlacek and Thistle 2006). Avoidance reactions of meiofauna to increasing currents and wave action, e.g., tidal wave fronts and concomitant vibrations of the sediment, have been documented and studied in experiments (Fig. 2.4; McLachlan et al. 1977; Meineke and Westheide 1979; Foy and Thistle 1991). Specialized species only will occur deep in the muds of sheltered flats or in the swash zones of exposed beaches (Menn 2002a, Gheskiere et al. 2005). Massive agitation of the sediment by storms apparently destroys the less agile meiofaunal groups. The erosion, shear strength and settling velocity of the sediment are not just influenced by abiotic factors. Biogenic factors such as the reworking of the sediment by intensive burrowing and pelletization as a result of defecation contribute considerably to the physical and biological properties of the sediments. Fecal pellets covering the bottom surface, especially in tidal flats, may reduce sediment shear strength and enhance erodibility by water currents, but they also tend to increase settlement velocity (Rhoads et al. 1977; Andersen and Pejrup 2002). Protruding tubes and plant culms may cause water turbulences and erosive forces, sometimes with negative impact on meiofauna (Coull and Palmer 1984; Hicks 1989). Agglutination by mucus (produced by bacteria, microphytobenthos and
High
12 Tide
Depth (cm)
8
16
20
Tide
Nematoda Harpacticoida
24 0
2
4
6
8
10
12
Low 14 (h) Time
Sunset
Fig. 2.4 Migration of beach meiofauna in relation to the tidal cycle. (McLachlan et al. 1977)
16
2 The Biotope: Factors and Study Methods
animals) as well as compaction in tube walls by small infauna will solidify the texture, increase sediment stability and diminish resuspension (Rhoads et al. 1978; Luckenbach 1986; Meadows and Tait, 1989; Decho 1990; Miller et al. 1996, Wiltshire 2000). When diatom populations became massively reduced by browsing so as to lower their mucus production, the shear strength and sediment cohesion decreased. There is a complex and dynamic correlation between the “biostabilizers” (mainly diatoms) and the “biodestabilizers” (mainly bivalves and polychaetes), between the “biosuspenders” (deposit feeders) and “bioirrigators” (tubicolous annelids and crustaceans) (Graf and Rosenberg 1997; Widdows et al. 2002; Meysman et al. 2006b).
2.1.3.2 Permeability, Pore Water Flow, Porosity, and Bioturbation Permeability. Permeability denotes the potential speed of water flowing through the sediment (volume of water flow per time, cm3 × s−1); in freshwater biology permeability is appropriately termed the “hydraulic conductivity.” It is calculated using a permeameter (see Fig. 5.4 in Giere et al. 1988). Directly influenced by the absolute size of the sand grains, it decreases as the proportion of small particles increases, especially those below 200 µm in diameter. The configurations of the individual pores and the structure and coherence of the void system all exert considerable influence on the permeability of the sediment, but permeability is not directly related to porosity (see below). Since water flow determines most chemical and physical factors in the sediment via the exchange rates of interstitial and supernatant water, permeability is responsible for supplying oxygen and dissolved and particulate nutrients, and so it largely controls the life conditions of meiofauna. Nutrient fluxes caused by pore water drainage through the tides easily exceed diffusive or bioirrigative fluxes (Billerbek et al. 2006). In exposed sandy beaches, the “tidal pump” creates a “stormy interstitial,” exposing the interstitial animals to high flow velocities. In near-surface layers, the wave-induced advective transport of pore water (>40 cm × h−1) exceeded diffusive transport by at least three orders of magnitude (Precht and Huettel 2003). Each lengthwise beach meter is percolated by several cubic meters of seawater each day. The yearly volume of the global shelf filtered (the “subtidal pump”) by the forces of percolation far exceeds the precipitation volume on land (Riedl and Machan 1972; Riedl et al. 1972). Berelson et al. (1999) calculated that within two hundred days the entire water column of Port Phillip Bay, Melbourne (Australia) passes through the sediment. In a tidal beach, 1 m2 of coarse sand filtered 14 L of water each hour (Rusch and Huettel 2000), a value also confirmed for Mediterranean shores (Precht and Huettel 2004). Even the small-scale topography of sandy bottoms massively influences advective water flux and particle transport. On the exposed sides of sediment mounds or ripple marks, surface water and organic particles penetrated about seven times deeper into the sediment than on the sheltered sides (Ziebis et al 1996; Huettel and Rusch 2000). The small-scale topography also directs the water flow, with intrusion occurring mainly in the ripple troughs and release occurring after filtration at the crests (Precht and Huettel 2004).
2.1 Abiotic Factors (Sediment Physiography)
17
The pore water velocity was first assessed by inserting heated thermistors into the sediment (Riedl and Machan 1972); microflowmeters based on minute thermistors were later used by researchers (LaBarbera and Vogel 1976; Davey et al. 1990). The cooling effect of the currents on a heated wire produces a voltage signal on a monitor, which, after calibration, indicates the microflow of water. Similarly, changes in the potential of a platinum wire used to measure the oxygen diffusion rate in sediments can be calibrated to record water microflows. Malan and McLachlan (1991) measured the pumping effects of waves and emphasized that most authors have underestimated wave-induced sediment water fluxes and their impact on the oxygen distribution. Long-term in situ records with oxygen microelectrodes have also indicated strong tide- and wave-dependent water pressure gradients (Weber et al. 2007). Precht and Huettel (2004) visualized the pore water flow in the field by applying fluorescent dye to the sediment and measuring it with an optical sensor (optode, see Sect. 2.1.4). Porosity. The total pore volume of a sediment core, its porosity or void ratio, depends in a complex way on the shape, sorting and mixing of the particles, and not just on the pore size available to the animals. Thus, it is not directly predictable from sieving data alone, but it is, of course, of relevance for physicochemical fluxes in the sediment. For mechanical measurements of porosity see Buchanan (1984) or Bale and Kenny (2005). Porosity profiles can also be calculated using electrodes, by measuring the resistivity of the sediment lattice (Archer et al. 1989). The velocity of the pore water flowing through the interstitial system does not depend solely on the hydrodynamics of the overlying water. The fluxes in the chemical milieu of sediments are strongly influenced by the sediment texture (see below). This, in turn, is controlled by bioturbation, sediment reworking, bioirrigation, and mucilage secretion of the benthic fauna (Graf and Rosenberg 1997; Pike et al. 2001; Berg et al. 2001; Murray et al. 2002; Meysman et al. 2006a,b), see below. Bioturbation. The biological reworking of sediments by endobenthic organisms, termed “bioturbation,” affects all sediments, limnetic and marine, from shallow tidal flats to deep-sea bottoms. In modern ecology, bioturbation is considered a major factor in the engineering of all benthic ecosystems (Meysman et al. 2006a) and in the creation of three-dimensional sediment structure (Lohrer et al. 2004). Because of its numerous biogeochemical implications, bioturbation probably had a massive influence on the archaic evolution of life (Bottjer et al. 2000; Dornbos et al. 2005; see Chap. 7). Among macrobenthos, bioturbation is mostly caused by the burrowing and digging of crustaceans and annelids. Bioturbative effects can extend to a sediment depth of >20 cm. Extrapolations suggest that in tidal flats the upper 10 cm of the sediment will become completely bioturbated once every three years. Depending on the population density, bioturbation can decrease compaction, and can even destabilize the bottom and increase its erodibility (Widdows et al. 2000). It provides a system of tubes and voids and enhances sediment mixing through particle exchange down to greater depths. Even sediment particles from a depth of 50 cm will be transferred to the surface. The burrows of the priapulid Halicryptus
18
2 The Biotope: Factors and Study Methods
spinulosus or the decapod Trypea (Callianassa) create a “secondary surface” of 0.7 m2 per surface m2 in muddy bottoms (Förster and Graf 1992; Powilleit et al. 1994). Transport rates of 40–50 g sediment per individual and day have been recorded through the burrow and void system. The subsequent increased water penetration accelerates the transport of solutes and gases more than diffusion (Diaz et al. 1994; Berelson et al. 1999; Berg et al. 2001). Twenty-five percent of the overall oxygen flux is attributed to the irrigational activity of burrowing animals (Booij et al. 1994). Bioirrigative water and solute transport into the sediment can exceed normal diffusion by a factor of ten (Aller 1988; Aller and Aller 1992; Kristensen 1988; De Deckere et al. 2001). Thus, benthic fauna markedly increases the flux of particles and modifies the physical processes (Graf and Rosenberg 1997). This has both beneficial and aggravating effects: organic matter and pollutants can be removed from the surface and buried into deeper layers where degradation is slow. On the other hand, the export of contaminated pore water is enhanced by bioturbators (Green and Chandler et al. 1994; Levin et al. 1997). By altering the geochemical system with animal tubes, and particularly through the import of oxygenated water by irrigational fluxes into anoxic layers, heavy metal precipitates (sulfides) that are buried at depth will become dissolved and released into the surficial, oxygenated layers (Green and Chandler 1994). Also phosphates and ammonium compounds are released in considerable amounts from the sediment by bioturbation and bioirrigation. The result is (often undesirable) eutrophication with an enhanced production of microphytobenthos in the overlying water (Monaghan and Giblin 1994). Bioturbative effects have also been shown to cause an undersaturation of calcites in the surficial layers, leading to increased shell dissolution and mortality (Green et al. 1998). Just as the sources of bioturbation are very diverse, their effects on meiobenthos are also very complex: negative impacts through disturbance and destabilization, positive ones through the oxygen and organic matter supplied (Green and Chandler 1994; Aarnio et al., 1998, Schratzberger and Warwick 1999; Thistle et al. 1999; Koller et al. 2006). Using radioactive isotopes and fluorescent dye as tracers, Bradshaw et al. (2006) found only minor chemical effects of bioturbation in Baltic Sea sediments compared to those of physical processes. On a general scale, the chemical impacts of the biogenic mobilization of buried chemical pollutants on meiofauna have not yet been sufficiently evaluated. While macrofaunal burrows affect the sediment, even the dense net of mmfine burrows of meiofauna can have a considerable influence on sediment structure and fauna colonization (Reichelt 1991, Fenchel 1996; Jensen 1996). Among meiobenthos, effective bioturbators are ostracods, nematodes and, particularly at the surface, harpacticoid copepods. Cullen (1973) experimentally demonstrated the bioturbative impact of meiofauna. He found that their burrowing activities alone eliminated all surface traces of macrofauna within 14 days. In average sandy sediment, the burrowing of meiofauna will completely displace the pore water in 1–3 years (Reichelt 1991). Because of meiofaunal bioturbation the transport of solutes with the subsequent stimulation of microbial mineralization was
2.1 Abiotic Factors (Sediment Physiography)
19
increased up to threefold compared to molecular diffusion (Rysgaard et al. 2000). Meiofaunal activity induces considerable microscale oxygen dynamics along the chemoclines of sediments, as documented by online registration with 2D planar optodes (Oguri et al. 2006). Distribution patterns of meiofauna, especially their colonization of deeper, anoxic horizons, have been shown to be highly dependent on the burrow system providing favorable microhabitats (e.g., Thomson and Altenbach 1993). Modern methods imaging the animal-made void system and bioturbative effects include the use of X-rays or fluorescence tracer techniques (Diaz et al. 1994, Powilleit et al. 1994). One (cost-intensive) method of analyzing the compositions of sediment cores and visualizing their biogenic tubes and burrows is computer-assisted tomography (Rosenberg et al. 2007). An indirect and elegant in situ method of demonstrating mixing processes due to animal activity on-line is the recording of oxygen changes by (expensive) optical sensors (Wenzhöfer and Glud 2004).
2.1.3.3 Water Content and Water Saturation The water content (mass of water in relation to the wet mass of a sample) is linked to grain size and permeability. Fine-grained sediments saturated with water have higher water contents than coarse sands. Mud cores often contain >50 weight % of water, while medium sand will only hold about 25%. Water content is considered by Flemming and Delafontaine (2000) to be a universal master variable that is relevant to any other sediment parameter (attention: inaccuracies may arise from the incorrect use of “content” and “concentration;” content denotes the mass per unit mass, while concentration is the mass per unit volume!). Water saturation and water flow play a dominant role in structuring meiofaunal settlement. If the water content fluctuates the pore water is replaced and the meiofauna are supplied with oxygen and nutrients, while in the deeper, permanently water-saturated layers the pore water flow is reduced. In tidal shores, the occurrence of meiofauna can become restricted because of insufficient water content. Moreover, because of their reduced capillary forces water-unsaturated surface layers cause steep gradients of many abiotic factors, such as temperature and salinity (see Sect. 2.1.4), often with negative impacts on the meiobenthos. Particularly in eulittoral shores at ebb tide, the degree of moisture or the desiccation stress often correlates with the distribution of meiobenthos. Lack of water in the surface horizons can force meiofauna into deeper horizons. Many eulittoral meiofauna species adapt to the regular tidal alterations of water content and concomitant salinity fluctuations with preference reactions and migrations (Fig. 2.4; see Sect. 7.3, 7.4). McLachlan and Turner (1994), based on Delamare-Deboutteville (1960) and Salvat (1964), suggested a generalized stratification pattern of (South African) beaches and their meiofauna in relation to desiccation and water saturation (Fig. 2.5).
20
2 The Biotope: Factors and Study Methods
dry sand swash zone
moist sand water-saturated
infiltration
low oxygen
brackish and fresh groundwater
Fig. 2.5 Stratification of a beach profile related to water content. (Compiled from various authors)
(a) An upper “dry sand stratum” is characterized by low water saturation and high fluctuations in temperature and salinity. Here, the prevalence of semi-terrestrial, specialized oligochaetes, mites and nematodes is contrasted with the scarcity of harpacticoids and turbellarians. (b) A partly underlying “moist sand stratum” (“zone of retention” in Salvat 1964) has an alternating water supply with fluctuations in temperature and salinity that decrease in the deeper strata. Due to the perpetually well-oxygenated conditions in this zone, meiofaunal abundance and diversity, particularly those of harpacticoids, increases. (c) In the “water table stratum” around the ground water layer, the sand is always water-saturated. In more sheltered beaches, restricted oxygen content and often brackish salinities lead to a reduced meiofaunal diversity and abundance. (d) The “low oxygen stratum,” where oxygen deficiency can extend down to a considerable depth, develops in beaches with a high content of organic matter; this zone can harbor meiofauna adapted to temporary oxygen depletion (see Sect. 8.4). Variations in this four-strata pattern primarily depend on the beach slope and result in different patterns of wave energy, particle size and nutrient supply: reflective, dissipative and intermediate beaches (Fig. 2.6). The tidal rhythm, local geography, high temperatures and different amounts of organic content will modify the above gradients. In flat-profiled and sheltered “dissipative” shores with medium-to-fine sand, the zonation is less developed.
2.1 Abiotic Factors (Sediment Physiography)
21
sand particle size physical gradient flow, moisture chemical gradients O2 , Eh relative importance of waves
relative importance of tides
reflective
intermediate
dissipative
drying
rentention
resurgence saturated
RPD
reduced
Fig. 2.6 Different beach types and their factor patterns. (After McLachlan and Turner 1994)
Conversely, in the exposed shore conditions of an “erosive” shore, the coarsegrained high-energy beaches have a “reflective” profile. Here, the waves prevent the occurrence of a low-oxygen stratum in the swash zone (“zone of resurgence,” Salvat 1964). This zone is characterized by intensive infiltration and circulation of interstitial water, and by a specialized interstitial fauna of reduced diversity and abundance. This physically controlled assemblage contrasts with the rich, often biologically controlled meiofauna of dissipative shores (Menn 2002b; see Sect. 9.4). Usually, moderately well-sorted medium sands provide the habitat with the most diverse meiofauna. In coarser sand, the species richness can be relatively high but population density may be low. Muddy sediments are more chemically controlled and often characterized by rich populations of a limited number of species restricted to the surface layer. In sublittoral sediments rich in organic matter, meiofaunal communities may be structured by the lack of pore water-flow and the resulting poor oxygen supply. In general, the correlation between hydrodynamic patterns, sediment structure and meiofaunal distribution is strong enough, particularly in littoral areas, to dominate all other factors. It often relates directly to the abundance and diversity of meiofauna, particularly nematodes (Vanreusel 1991, Menn 2002b, Gheskiere 2005). More detailed reading: Hylleberg and Henriksen (1980); Yingst and Rhoads (1980); Gray (1981); Buchanan (1984); Aller (1988); Giere et al. (1988a); Kristensen (1988); Watling (1988); McCall and Tevesz (1982); Hall (1994); McLachlan and Turner (1994); Snelgrove and Butman (1994); Graf and Rosenberg (1997); Widdows et al. (2000); Pearson (2001); Cadée (2001); Murray et al. (2002); Reise (2002); Bale and Kenny (2005); Meysman et al. (2006a).
22
2 The Biotope: Factors and Study Methods
Box 2.2 The Sediment–Water Regime: Exposure, Permeability, Water Circulation, Bioturbation The exposure of a habitat, which depends on the impacts of waves and currents, determines agitation, erosion, suspension, sorting of the sediment and flow of interstitial water. The degree of exposure, a complex parameter, is difficult to assess and is often only estimated. The erosion and permeability of the sediment are influenced by the hydrodynamic system, the sizes and shapes of the particles, and their material (quartz or biogenic calcium carbonate). The presence of organic matrices (biofilms, fecal pellets) is also important. The sediment represents a huge filter system of the water flow that provides the benthos with particulate organic matter from the water column and with dissolved nutrients. If saturated, the water content, determined by the capillary (adhesive) forces between the particles, is high in silt and mud (>50%) and low in sand (about 25%). In tidal shores, meiobenthic living conditions are influenced by pore water exchange. The sediment as a habitat is further complicated by biotic factors like bioturbation. This burrowing activity of animals massively reworks and irrigates the sediment and enhances pore water flow and primary production. Compounds bound to the sediment can become dissolved, resulting in eutrophicating or polluting effects. Secretions and tubes compact the bottom. A complex web of particle mixing, in- and outflows, biosuspension and biocompaction links the sediment and water column. Various stratification patterns have been suggested for tidal shores. Based mainly on the beach slope, wave energy and tidal regime, dissipative accreting shores can be distinguished from reflective erosive beaches by grain size, water content, nutrients and oxygen supply. As meiofauna avoid strongly agitated sands, intermediate or dissipative beaches will be populated by more diverse and richer meiofauna.
2.1.4
Physicochemical Characteristics
2.1.4.1 Temperature Meiofauna are present in polar ice and tropical coral reefs, in the constantly cold deepsea and in the supralittoral fringe with frequent temperature fluctuations. Nevertheless, extremes of temperature can have a structuring impact on meiofauna, particularly in exposed tidal shores with their steep vertical thermal gradients. However, in sublittoral bottoms the influence of temperature on meiofaunal distribution is normally negligible. The steepness of the temperature gradient is strongly related to permeability (see Sect. 2.1.3). In water-saturated boreal mud flats of low permeability, surface and deeper layers can diverge widely in temperature, particularly at ebb tide. While summer temperatures can rise to >40 °C at the surface, those in the
2.1 Abiotic Factors (Sediment Physiography)
23
a
b Fig. 2.7a–b A typical temperature distribution in a boreal beach. a Summer aspect. b Winter aspect. (After Jansson 1966a)
depths are only 10–15 °C because of a strong vertical dampening. In wintertime, even under thick ice cover, the frozen ground at the surface does not extend beyond the uppermost 5 cm (Fig. 2.7). This dampening effect with depth, which is particularly evident when calculating monthly ranges (Table 2.3), is important for the existence of meiofauna in climatically harsh biotopes where sensitive species often perform vertical migrations if other conditions like oxygen supply are favorable. On the other hand, many meiofaunal animals are highly resistant to frost, either by supercooling or protective dehydration. In Lake Taimyr (Siberia) nematodes and oligochaetes have been reported to regularly survive for months frozen in ice at temperatures of −10 °C or less (Timm 1996). The Alaskan ice worm, a black enchytraeid oligochaete, lives permanently in crevices of ice at temperatures of below 0 °C (Goodman and Parrish 1971). The (terrestrial) Antarctic nematode
24
2 The Biotope: Factors and Study Methods
Table 2.3 Monthly temperature ranges in 1964 at various sediment depths in a Scandinavian beach (Jansson 1967) Depth March April May July October Air −9.0 19.9 19.7 18.8 14.5 Surface −13.0 34.2 38.4 33.5 24.1 2 cm −8.7 26.0 33.0 24.0 12.8 10 cm −4.1 9.8 11.2 9.9 4.2 25 cm −1.0 2.8 3.4 – – 70 cm −0.5 0.8 0.5 0.6 1.0 Note: Values represent the differences between the lowest and highest temperatures (minus signs indicate when the sum of the minimum and maximum temperatures had negative values)
Panagrolaimus davidi survives temperatures as low as -80˚C and freezing down to >80 % of its water bodies (Smith et al. 2008). Polar sea ice is a permanent habitat for a rich, specialized “sympagic” meiofauna of ecological importance (Gradinger 1999a, Gradinger et al. 2005, see Sect. 8.1.1). Temperature can be conveniently and routinely measured with a variety of pointed semiconductor probes connected to electronic (field) instruments. Since only the narrow surface of the thin metal probe is temperature-sensitive, in situ measuring is possible even at a considerable penetration depth without much compaction or displacement of the sediment.
2.1.4.2 Salinity As with temperature, meiofaunal organisms exist under all salinity regimes from freshwater to brine seep areas, from brackish shores to deep-sea bottoms. Because many species are able to adapt to a wide range of salinities, there is often even a diverse meiofauna in those critical brackish water zones where Remane (1934) described a minimum number of species, mainly for macrofauna. Yet, depending much on the frequence of variations, salinity gradients can strongly determine occurrence and species composition of meiofauna (Ingole et al. 1998; Richmond et al. 2007). Habitat-adapted ranges of salinity tolerance or preference have been experimentally found in various meiofaunal species (Giere and Pfannkuche 1978; Ingole 1994; Moens and Vincx 2000b); the physiological capacity for salinity regulation was elegantly recorded for some littoral nematodes by Forster (1998) using an optical method based on the interference pattern of body fluids. Today the international salinity unit is PSU (Practical Salinity Units) which corresponds to ‰ S. In African volcanic lakes high conductivity (often together with extremes of pH, see below) not only structures the occurrence of different meiobenthic assemblages, it locally excludes the existence of meiofauna (Tudorancea and Taylor 2002). In tidal shores, the steep vertical and horizontal salinity gradients strongly depend, as with temperature, on the water permeability of the sediment. In muds with their water-saturated fine sediments and much reduced vertical water exchange capacity, the surface salinity at ebb tide can rise up to hypersaline conditions due to evaporation.
2.1 Abiotic Factors (Sediment Physiography)
25
After heavy rainfalls it can suddenly drop to almost zero. The effects of these fluctuations are greatest in the uppermost centimeter. At a depth of 2–5 cm, the fluctuations in salinity (as well as temperature) are much dampened and often remain amazingly constant. This offers mobile meiofauna a favorable refuge zone, while a drastic decline in meiofaunal abundance after flooding rains or tropical monsoonal rains has been reported for the surface layers (Alongi 1990b). At the highwater line of a North Sea mud flat, salinities of up to 40‰ PSU have been measured on a warm summer day, while immediately below at 5 cm depth, 30‰ PSU was not exceeded. At depths approaching the ground water level, the salinity often drops to brackish water conditions (about 20%). In contrast, on coarser, well-drained sandy beaches with a high permeability, precipitation can affect salinity to a depth of 30 cm (Reid 1932–33), creating adversely lowered salinity conditions for meiofauna, even in the depths. Consequently, in sandy beaches the drainage system is complicated by the effects of both precipitation and ground water currents. Even in fully marine shores without any direct freshwater influx at the surface, groundwater is often markedly reduced in salinity depending on the local geological, climatic and geographical conditions. In sublittoral bottoms, the salinity fluctuates less and is usually identical to that of the overlying water, and so it is hardly a limiting factor for meiobenthic populations. Only beneath brine seeps (Gulf of Mexico) and in brine basins (Mediterranean Sea) are meiofauna exposed to adversely high (>150 PSU) salinities. In the high shore, salinity gradients are difficult to record in detail because the amount of water available is often restricted. Two methods, microtitration and electrical conductivity measurements, are often used. Pointed conductivity electrodes measure the electric current between two platinum rings, which is modulated by the conductivity of the pore water. After calibration, this measurement is used to indicate the salinity of the water. In sediment with little moisture, conductivity measurements may be problematic. Here, the salinity of sediment water can be determined refractometrically. Special salinity refractometers are provided with a scale that is converted and calibrated for direct readings of salinity from only one drop of water extracted from the sediment. The precision of these instruments (±1 ‰ S) is usually sufficient for ecological studies of meiobenthos. As with all methods of salinity measurement, in brackish water the precisions of both the conductive and the refractometer methods suffer because the altered ion composition will cause deviations from the constant relationship that classically defines salinity measurements in pure ocean water.
2.1.4.3 Acidity/Alkalinity (pH Value) Water acidity, recorded in pH-units, used to play only a minor role for meiobenthos in the marine biome. The slight alkalinity of seawater (pH 7.5–8.5) makes it well buffered against pH fluctuations. Only in anoxic, hydrogen sulfide-containing sediments does the pH drop below 7, and rarely below 6. Hence, pH recordings are essential when sulfide concentrations are calculated (see below). On the other hand, in the surface layers of tidal flats, intensive assimilation of the abundant microalgae can increase the pore water pH to >9. In tropical tidal flats pH values of up to 10
26
2 The Biotope: Factors and Study Methods
have been measured. Here, high pH in combination with other stress factors such as extreme temperatures and salinity can be detrimental. However, in the future there is a global risk that the dramatic rise in CO2 will increase the acidity of seawater, with negative effects on the benthos. pH and pCO2 are master variables in the formation of carbonate species (Cai and Reimers 1993), the relevance of which has been underlined by Green et al. (1998) based on manipulating the saturation state of carbonate. Hence, particularly calcareous shells and calcifying processes will be impaired by elevated acidity, as has been shown for macrobenthos. In spite of the high resilience of the sedimentary system, undersaturation of calcite (which is typical of many littoral sediments) massively increases mortality and causes shell dissolution in Foraminifera. The authors conclude that dissolution by undersaturation of calcite “likely represents an unrecognized source of mortality for carbonate-bearing meiofaunal-sized organisms.” Under stable deep-sea conditions, experimental CO2 deposition caused a decline in sediment pH of up to 0.75 units and significant defaunation among meiofaunal nematodes and protozoans (Thistle et al. 2005, 2006, 2007a). Even mild acidosis with a decrease in pH of only 0.1–0.2 caused mortality among meiobenthos of up to 30% in the affected area (Barry et al. 2005; see Sect. 8.3). In natural freshwater biotopes, extremes of pH can occur in mires and limestone waters, but also in spring lakes of volcanic origin. Many freshwater bodies are, of course, exposed to anthropogenic pollution, which often causes drastic fluctuations in the pH level. However, the buffering capacity of the sediments dampens the fluctuations in the pore water compared with the overlying water. Subterranean fauna is well adapted to the slightly acidic conditions in continental groundwater aquifers. Thus, a negative impact of acidity or alkalinity on freshwater meiofauna has rarely been demonstrated (Pennak 1988). A reduction in diversity and abundance does however occur in volcanic lakes, where extremely ionic conditions, e.g., alkaline and saline conditions (“soda lakes”), have been found to reduce the diversity or even to limit the occurrence of (nematode) meiofauna (Tudorancea and Zullini 1989; Muschiol and Traunspurger 2008). pH also has a major influence on the balance of many physiological reactions that are not related to calcification, e.g., respiratory processes. Since the effects of pH changes are often confounded by concomitant parameters (e.g., metabolic processes, oxygen binding) in the field, the specific impact of acidosis or elevated pH should be studied experimentally, separate from other factors. The measurement of pH in situ is not problematic when glass insertion electrodes with the reference electrode combined in the same shaft are used. Usually, the recording is done in parallel with redox potential measurements performed with a mV-meter calibrated for pH. Modern electrode design even prevents clogging of the diaphragm. Correct readings require that the temperature of the electrode’s internal filling equals that of the ambient water or sediment to be measured. Internal and external air pressures must also be equalized. A special pH electrode with a recorder integrated into its shaft has been developed to allow for measurements in a drop of pore water filling a small hollow. This even enables correct measurements of pH to be obtained in water-unsaturated shore sediments. pH electrodes in combination with pCO2 microelectrodes have been designed for combined recordings (Green et al. 1998).
2.1 Abiotic Factors (Sediment Physiography)
27
Box 2.3 Temperature, Salinity, Acidity In their normal ranges, temperature, salinity and acidity are not limiting to meiofauna. However, in polar or tropical shores, tidal flats or sea ice, and in volcanic waters, extreme values can become limiting for meiofauna and favor resilient communities of relatively low diversity. A global increase in CO2 will result in the lowering of seawater and sediment pH values and will influence the physiological pH balance and the formation of calcareous structures in sealife. The consequences for shallow-water meiofauna remain to be assessed. Dumping CO2 into the deep sea as a way of solving the CO2 problem seems an inappropriate and risky approach for benthic life.
2.1.4.4
An Interacting Complex: Redox Potential, Oxygen, and Hydrogen Sulfide
Increasing temperatures and eutrophication cause a reduced oxygen supply, mostly associated with increases in hydrogen sulfide and ammonia in many areas. This is a particular problem for benthic biotopes, with their rich content of degradable, oxygen-consuming matter (see reviews by Diaz and Rosenberg 1995, Wu 2002). Redox potential. Before the development of suitable oxygen and sulfide electrodes, platinum electrodes were used to measure the overall reducing or oxidizing capacity of a sediment, the redox potential. The resulting values were simple to record but difficult to interpret. Nevertheless, redox values have become one of the most frequently provided environmental parameters in meiobenthic studies and are sometimes still used. However, redox potential recordings only vaguely reflect the supply of oxygen in the pore water, and the measurement of redox potentials can easily be deceptive: all of the controlling redox couples that occur in the sediment, in addition to those induced by free oxygen, are integrated in a redox potential reading and will influence the measurement. Moreover, a basic understanding of electrometry and some information about inherent flaws and errors is needed to avoid bias and misinterpretation. The electrical potential between an outer platinum electrode and an internal reference electrode is recorded with a mV-meter, which must be corrected to yield the “Eh value.” The response of the electrode depends to a high degree on the properties of the platinum surface. Eh measurements and their interpretation become even more of a problem because of the low reproducibility of replicate recordings from directly adjacent spots. This problem is largely caused by microchambers of decaying organic matter, entrapped air bubbles or animal tubes encountered by the pin-pointed electrode. In the field one can encounter values between +550 and –300 mV. In exposed sandy bottoms that are typically yellowish in color, positive values are obtained throughout, while in soft muds with rich organic content, the gray-to-black layers underneath a thin brighter surface layer will yield clearly negative redox values. The transition zone between positive and negative redox values is termed the
28
2 The Biotope: Factors and Study Methods
“redox potential discontinuity layer” (RPD layer). The RPD layer can periodically move up or down with an increase in the assimilation activity of surficial diatoms or with the incoming tide. Relation of redox values to sediment color. It is relevant to note (but is often disregarded, even in recent meiofaunal studies) that, for chemical reasons, the RPD threshold does not eo ipso parallel the shift from bright sediment layers to dark ones. Detailed Eh measurements have shown that the change from light (yellowish) to dark instead indicates the transformation of ferric iron to ferrous iron rather than the RPD (the redox potential can still be +125 mV, see Fig. 2.8a, and data in Sikora and Sikora 1982; Jørgensen 1982). Only in those cases where the transition layer is narrow and the change from bright to deeply black is abrupt (in muds) can the color change coincide with the RPD. In other words, a wide and diffuse transition from bright over gray to black just indicates the gradual disappearance of free oxygen to a layer with oxidized compounds, albeit without free oxygen. This is then followed by the fully reduced layers without any traces of free or bound oxygen. The value of an optical sediment indicator for the presence of oxygen is further limited in many calcareous sediments, where the iron-based change of the sediment color (brownish = ferric iron, grayish = iron sulfides) may be absent because of the low iron content in the sediment. For a long time it was considered a general and practical rule for interpreting Eh values that values of > +100 mV would indicate the presence of oxic pore water in the sediment, and < −100 mV its absence. Today, oxygen microelectrodes have shown that free oxygen can be absent in sediments with redox potentials as high as +300 mV. Differentiating between oxic (i.e., with free dissolved oxygen available to animals) and oxidized sediments (with compounds in an oxidized chemical state) is of prime importance for correct ecological interpretations (Sikora and Sikora 1982; Jørgensen 1988). In a coastal sandy sediment, Revsbech and Jørgensen (1986) demonstrated that the oxic zone was only 2 mm thick, while an oxidized sediment layer with a positive redox potential extended down to 3.5 cm. Microelectrode measurements have shown that the oxygen content of the pore water can be “zero” in sediment layers that are still brownish to yellowish, i.e., oxidized (Fig. 2.8a). Likewise, the presence of toxic hydrogen sulfide is not automatically indicated by grayish to dark sediments (see below). Only when oxygen or sulfide ions occur in excess and are not chemically bound will the pore water become oxic or sulfidic, respectively. Today it is preferable in meiobenthic studies to measure the oxygen concentration directly with small oxygen electrodes (see below) rather than using the redox potential as a difficult substitute that cannot demonstrate the microdistribution of oxic/anoxic niches. Oxygen. Oxygen is the predominant factor among the abiotic parameters determining the habitat conditions and the presence of meiofauna. Meiobenthic organisms have relatively large surface areas and mostly high oxygen demands; only a few specialized forms will prefer hypoxic conditions (see Sect. 8.4). Thus, the distribution of most meiofaunal communities can be correlated to the oxygen supply of the pore water. Technical progress has largely changed our conception of oxygen supply to the benthos. The classic picture of a vertical oxygen stratification must be abandoned.
2.1 Abiotic Factors (Sediment Physiography) 0 depth 0 [cm]
29
0.1
0.2
O2[mM]
O2
1 Eh
2 3 4 5 6
H2 S
7 −100 0
a
0
+100 +200 +300 Eh[mv]
0.2
0.4
H2S[mM]
Oxygen and Sulphide (µmol l−1) 0
200
400 0
Oxygen
1000
2000
Oxygen
Depth (mm)
0
Sulphide
1
pH
Sulphide
2
DARK b
7
LIGHT 8
7
8
9
Fig. 2.8a–b Oxygen and hydrogen sulfide gradients in two sediment profiles from tidal flats. a Showing relationship with the redox potential (Eh). b Showing relationship with the acidity (pH) and assimilation-induced changes. (a is after Revsbech and Jørgensen 1986; b is own recordings)
30
2 The Biotope: Factors and Study Methods
New microelectrometric techniques have revealed a much-differentiated pattern of oxic conditions in the sediment. Numerous micro-oxic niches occur within anoxic sediments, creating temporary microenvironments with a potential distributional effect for meiofauna (Forster et al. 1996) in the same vertical layer. Numerous geochemical cycles and microbial processes together with the bioturbative impact of sediment fauna and tidal fluctuations create a complicated three-dimensional net of oxic/anoxic microniches. Many sands and muds are only oxygenated in the uppermost millimeter-thin layer (Revsbech and Jørgensen 1986; Visscher et al. 1991). In intertidal flats densely populated by meiobenthos, recordings of oxygen and hydrogen sulfide taken with a set of microelectrodes (see below) showed a very strong periodicity in the fluctuations in oxygen and sulfide concentrations (De Beer et al. 2005a; Weber et al. 2007): At ebb tide, the oxygen content declined rapidly and anoxia (plus hydrogen sulfide) often even reached the surface layer. When bioturbation/irrigational fluxes are interrupted, rapid oxygen depletion through bacterial consumption will soon create hypoxic and anoxic conditions in previously oxic burrow walls, the preferred habitat of meiofauna (Wetzel et al. 1995, Fenchel 1996). The incoming tide reoxidizes the interstitial sediment water, even before the flats become inundated again. Even with tidal regularity, these intermittent fluctuations of oxygen and sulfide exert considerable ecological stress on meiofauna that was not realized before. Aside from the tidal fluctuations, the diurnal fluctuations of light exposure cause changing micro-oxic stratifications via the assimilation activity of the phytobenthos (Fig. 2.8b). Similarly variable oxygen conditions have also been reported for other biotopes (e.g., Archer and Devol 1992; Förster and Graf 1992). In seagrass beds, anoxic periods can be particularly destructive and long-lasting (Guerrini et al. 1998). Buried seagrass and algal remains caused black patches on tidal flats devoid of fauna (Neira and Rackemann 1996) and with a delayed recolonization by meiofauna (De Troch et al. 2005). Other microscale oxygen recordings have revealed that biogenic microtopographic surface structures such as small mounds modify the hydrodynamic and oxidative micropatterns (Ziebis et al. 1996). In a North Sea tidal flat, diffusion of oxygen from Arenicola burrows extended only about 1 mm into the surrounding fine sand (Wetzel et al. 1995), results which were also confirmed by Fenchel (1996) for burrows in the mm range (Fig. 2.9). Many microgradients of oxygen available to meiobenthic animals are, in fact, so steep that changes from fully oxic to anoxic conditions can occur within the µm range of microbial mats. The narrow transition zones from oxidized to anoxic layers can imply that aerobic and anaerobic microbial processes can take place at same depths, and that micro-oxic conditions can temporarily persist beside sulfidic ones (Fenchel 1996; Meyers et al. 1988). The view that the micro- and meiobenthos beneath the upper few millimeters live a largely anaerobic life (Revsbech et al. 1980a,b) has to be refined. In fact, these organisms seem to continuously oscillate between (micro)oxic islets, they tolerate short anoxic events which can momentarily cease, and their habitat changes temporarily under the influence of assimilation and tidal influx. Probably, many meiofauna are forced to intermittently gain energy by rapidly switching from aerobic metabolism
2.1 Abiotic Factors (Sediment Physiography)
31
250
O2 (µM)
200
150 100
50 0 0
1
2
3
Distance from center of burrow (mm)
Fig. 2.9 Decrease in oxygen concentration around a thin worm burrow in a muddy bottom. (After Fenchel 1996a)
to phases of passive outlasting or anaerobic metabolism. The oxygen flux is also very limited in deeper shelf and slope sediments. (On the other hand, the deep-sea ooze, with its low content of organic particles, can stay oxidized all the way down to several centimeters in depth.) The fine-scale assessment of oxygen stratification directly in the pore water system became possible within the last few decades through the use of thin oxygen microelectrodes, which were introduced into experimental ecology mainly by Revsbech and his group (see Revsbech and Ward 1983; Revsbech and Jørgensen 1986). Protected from abrasion by a layer of semipermeable silicon rubber, and provided with a sturdy glass shaft, field versions have been designed (Fig. 2.10; Revsbech and Ward 1983). By ensheathing the electrode in a thin stainless steel shaft (Helder and Bakker 1985; Visscher et al. 1991), the risk of breaking this device during field use is further reduced. In contrast to polarographic oxygen electrodes, microelectrodes do not require any water flow and, if gold-coated, the presence of dissolved sulfide does not tarnish the minute measuring surface. However, these microelectrodes, with their critical limits of detection (5–10 mmol oxygen), hardly allow for discrimination between truly anoxic and hypoxic conditions—a problem of particular relevance in respiration experiments. Oxygen concentrations in this low range might be still relevant for adapted micro- and meiobenthos with high oxygen affinity (Watling 1991). The measuring threshold in oxygen recordings was minimized upon the development of novel optical oxygen electrodes (“optodes,” Klimant et al. 1995). These fiber-optic microsensors measure the extremely sensitive luminescence quenching of certain chemical dyes that emit light when oxidized. Their highest sensitivity occurs at an ecologically important low oxygen concentration, and their detection
32
2 The Biotope: Factors and Study Methods
guard silver cathode
epoxy
platinum wire
silver wire cathode
Ag /AgCl anode
sensing gold cathode
electrolyte
electrolyte silicone rubber membrane
10 mm
10 µm
Fig. 2.10 Oxygen microelectrode with a sturdy glass shaft. (After Revsbech and Ward 1983)
limit can be 0.1 µM (or 0.5 ppb). The most advanced recording method visualizing oxidative processes online and in situ is the development of 2-D planar opt(r)odes (Glud et al. 1996, 2001; Wenzhöfer and Glud 2004), by which the small-scale dynamics of oxygen distribution, caused by changes in water flow, assimilation or bioturbation, can be continuously and precisely registered on-line (Oguri et al. 2006). The inserted tiny oxygen-quenchable fluorophore creates an on-line, precise and highly stable “oxygen picture” over time. If constructed for field use, these optical sensors can bridge the gap between punctuated microelectrode measurements and the continuous, integrative recordings of chamber incubations. The minimal dimensions and the precision of these microelectrodes allow for the construction of novel microchambers and non-invasive, exact measurements of meiofaunal respiration to be obtained (Moodley et al. 2008; see Sect. 9.3.2). The major disadvantage of this instrumentation is its cost. In recent years microelectrodes and optodes have been adopted and modified by various working groups (Weber et al. 2007), but there is still a need to construct sturdier versions that are suitable for use by nonspecialists in field operations at a reasonable price.
2.1 Abiotic Factors (Sediment Physiography)
33
WATER benthic microalgae
flow
O2
O2 O2
oxic
O2 O2
O2
O2
O2
O2 O2
anoxic
O2
benthic fauna
O2
O2 O2 O O2 2
O2
O2 O2
O2
SEDIMENT
Fig. 2.11 The microdistribution of oxygen in a shallow flat. (Courtesy Wenzhöfer)
The principally new conception of oxygen conditions in the benthic environment is illustrated in Fig. 2.11. The oxygen distribution is extremely patchy and variable, primarily due to the tidal cycle, bioturbation and organismic activities. The overall volume of oxic microniches grossly exceeds that of the “classical” two-dimensional sediment surface. Considering the complexity and dynamics of the “oxygen realm” and its ecophysiological implications for the small benthos, further microscale investigations, especially those performed under natural field conditions, are needed. It is obvious that oxygen recordings from pore water samples obtained with syringes or by centrifugation (see Sect. 3.3) cannot yield an accurate resolution of the oxygen conditions, whether performed by a micro-Winkler titration (Bryan et al. 1976; Peck and Uglow 1990) or with a polarographic electrode. The fetch area of the pore water is clearly not reproducible. These methods, often still used today, yield a misleading oxygen distribution and should be replaced by direct measurements with microelectrodes. The same holds true for the “oxygen diffusion rate” (ODR, see Jansson 1966a), an historical method that is difficult to interpret and of little relevance today. Hydrogen sulfide. Hydrogen sulfide (H2S) is, oxygen aside, perhaps the most relevant environmental parameter in benthic habitats. Under normal marine pH conditions, hydrogen sulfide is predominantly dissolved as HS− ion. Only in more reducing habitats at slightly acidic pH can the particularly toxic undissociated H2S molecules prevail. Because it (reversibly) blocks the cytochrome C oxidase of the intracellular respiratory chain, hydrogen sulfide is toxic to animals at just nanomolar to micromolar concentrations and profoundly influences the distribution of the benthos. Especially in marine biotopes, H2S can develop in high concentrations
34
2 The Biotope: Factors and Study Methods
through anaerobic microbial reduction of sulfate. Sulfide production can be maintained by sulfate reducers, even under low-oxygen conditions (Jørgensen 1988; Fukui and Takii 1990; Jørgensen and Bak 1991). Plant roots (e.g., Spartina) may enhance sulfide oxidation through both oxygen production and catalytic effects (Lee et al. 1999; Lee 2003). Because of the chemical balance between iron and sulfide ions, in temperate regions a considerable proportion of the generated sulfide will precipitate as iron sulfides, forming a grayish-to-black layer in the depths of the sediment. Particularly in the warm season, the deeper layers of muddy intertidal flats, rich in organic matter, develop high hydrogen sulfide concentrations that sometimes exceed 1 mmol l−1 (author’s unpublished data; Rey et al. 1992). In polluted sediments, dissolved hydrogen sulfide has an indirect but important impact in fixing toxic heavy metals such as cadmium as solid precipitates. In calcareous sediments of warm-water regions, with their low iron contents, this precipitation process is limited. Here, high concentrations of toxic dissolved hydrogen sulfide can develop without any blackening of the sediment (see above). In freshwater biotopes, periods of anoxia rarely give rise to comparably high concentrations of H2S. In this environment, sulfate ions are rare and sulfide is derived mostly from the organically bound protein sulfur. Despite its ecological relevance (see Sect. 8.4), hydrogen sulfide has mostly been ignored in field work and in experiments. In part, this may be due to the widely reported but prejudiced notion that H2S-smelling sediments are azoic. The difficulty involved in measuring this labile substance quantitatively may also have caused it to be neglected. Smell is a good criterion for detecting even low concentrations of hydrogen sulfide, because human olfactory organs sense hydrogen sulfide concentrations as low as 0.1 µm H2S. Today, hydrogen sulfide in sediment pore water is usually measured spectrophotometrically after the addition of methylene blue (Cline 1969; GilboaGarber 1971). There are some modifications of this method that mostly concern the calibration and the range of concentration. Pore water is obtained by suction corers (see Sect. 3.3) and injected right after extraction into the prepared vials filled with alkaline lead acetate, the precipitate is later dissolved in the laboratory. Although somewhat cumbersome, this “pore water method” with subsequent photometrical analysis remains a versatile and frequently used method of measuring hydrogen sulfide. As with oxygen, the stratification pattern of hydrogen sulfide in the sediment column is three-dimensional, extremely complex and characterized by sulfidic microniches directly adjacent to oxic spaces and by steep gradients (see Fig. 2.8b). Microelectrodes have been developed in order to obtain a realistic picture. Several companies offer a silver sulfide electrode combined with a reference system of calomel or Ag/AgJ. The recorded voltage (mV) is directly proportional to the logarithm of the S2− partial pressure. Therefore, it is essential to convert all hydrogen sulfide in its various pH-dependent stages of dissociation to S2− and to calibrate the electrode in relation to the ambient pH. In particular,
2.2 Abiotic Factors Csediment Physiography
35
the micro versions of this electrode, designed originally by Revsbech and Ward (1983), are of interest as insertion electrodes for field use. Early constructions suffered from technical problems (mechanical fragility and abrasion of the silver sulfide coating). Mounting the sulfide microelectrodes in the steel injection needles from medical syringes has made recent versions sturdier and sufficiently protected, even for use in sandy substrates (Van Gemerden et al. 1989). A particularly promising tool for simultaneous recordings of the closely interacting compounds combines an oxygen and a hydrogen sulfide electrode within one thin metal needle (Visscher et al. 1991). The construction of a needle-pointed amperometric sulfide microelectrode combines increased sensitivity with broad applicability (Kühl et al. 1998). Its field use in profiler experiments enabled minute gradients in microprofiles to be recorded. Visman (1996) designed an automatically controlled experimental oxygen/sulfide system with variable pH and temperature conditions that enables complex ecophysiological experiments with sulfide. Although even these new designs do not make measurements of hydrogen sulfide a simple and convenient task, they will hopefully help to ensure that this dominant factor in benthology is no longer neglected in the future. More detailed reading: Giere (1992); Grieshaber et al. (1994); Grieshaber and Völkel (1998); Wetzel et al. (2001).
Box 2.4 An Interwoven Complex: pH, Redox Potential, Oxygen, and Hydrogen Sulfide Oxygen availability and exposure to hydrogen sulfide often set the limits on the distribution of benthic animals. In turn, the development and stratification of these chemofactors depend significantly on biotic processes. The microbiological depletion of oxygen to zero restricts most meiobenthic organisms to the surface layer as a suitable habitat in many sediments. However, the boundary to the anoxic strata, which are often dominated by toxic hydrogen sulfide, changes its position because of tidal fluxes and assimilatory oxygen production. This is the preferred biotope for rich microbiota. Bioturbative processes performed by macro- and meiofauna and microtopographic structures (tubes, mounds) locally alter this vertical stratification, creating a complex threedimensional pattern of oxic and sulfidic microniches. This is often the basis for the notoriously patchy meiofaunal distribution. Measurements with thin microsensors give a more accurate picture of this microscale oxic/sulfidic regime. Their high sensitivity makes it possible to measure even the critical micro-oxic and microsulfidic ranges around the chemocline, which are relevant to many fauna in hypoxic and low-sulfidic sediments. Long-term measurements in benthic chambers have revealed that this micropattern of oxic/sulfidic conditions in the sediment is continuously changing, a scenario that has completely changed the former conception of a two-dimensionally layered system. (continued)
36
2 The Biotope: Factors and Study Methods
Box 2.4 (continued) Because of advances in oxygen microelectrodes and optodes, the use of redox potential as an indicator of oxic/sulfidic conditions is losing importance. Eh measurements in sediments are, at best, inexact and their interpretation is vague. An operational parameter with an integrative character, Eh does not reflect the fine-scaled and dynamic oxic/anoxic/sulfidic system. Commercially available oxygen microsensors remain expensive considering their fragility during field use. Technical attempts to produce robust constructions have been rare.
2.1.4.5 Inorganic Nutrients and Pollutants Nutrient enrichment and contaminants from anthropogenic pollution are factors that today have an important impact on the meiobenthos in many parts of the world (Nixon 1995). Because of physical adsorption and chemical bonds, nutrients and pollutants in the sediment are enriched by several orders of magnitude compared to the overlying water. Absorption of inorganic and organic nutrients, especially in carbonate sediments, is very high (Rasheed et al. 2003a). Once bound in the sediment, these substances are only slowly released into the water column. Retarded microbial decomposition in the oxygen-deficient deeper layers of many bottoms combined with high absorptive forces can cause long-lasting or chronic negative effects on the benthos. Thus, sediments often represent sinks for pollutants. Here, the intensive transport of oxygen in plant roots can have a phytoremediative effect (Lee 2003). Measurement of nitrogen and total phosphorus are summative indicators and are hardly adequate for assessing the situation. One potentially powerful method indicating the extent by which the meiofauna are aggravated by contaminants might be gene or genome analyses of exposed animals (Staton et al. 2001). In any case, eutrophication and pollution have been linked to patchy distribution patterns of meiofauna (Lambshead and Hodda 1994). In many coastal areas, while terrigenous eutrophication may have been abated in recent years, fish and mussel farms in formerly often oligotrophic, barely exposed environments (Norwegian fjords, Mediterranean Sea bights) create an ever-increasing source of nutrient input, which has consequences for meiofauna, especially in sheltered coastal sediments. Another widespread man-induced environmental factor in the meiobenthic habitat is sediment pollution by heavy metals, antifouling compounds (TBT), and pesticides. Because they usually have only a local impact, chemical agents are not considered here as general habitat factors. The reaction of meiofauna to chemical pollution will be detailed in the synecological part of this book (Sect. 8.8). More detailed reading: Kristensen (1988); Watling (1991); Huettel and Gust (1992); Chester (2002).
2.2 Biotic Habitat Factors: A Connected Complex
37
Box 2.5 The Abiotic Environment of the Meiobenthos In the natural environment all of the single factors described in the above section interact with counteracting and synergistic effects. Attractive nutrients are enriched by the highly absorptive bonds at the grain surfaces, as are noxious pollutants. Steep oxic gradients favor microbial life, but anoxia and (often co-occurring) hydrogen sulfide are highly toxic. Erosive water currents act negatively on meiofauna, but they also transport solutes and oxygen into the system. These interactions create a three-dimensional pattern of small-scale favorable microniches or adverse patches that represents the abiotic environment of meiobenthic animals. It is a system governed by currents and sediment composition as key factors, which influence a cascading, multifactorial network to which meiofauna respond in varying patterns and sequences. Experiments have revealed some of these reactions (e.g., to temperature or salinity), but the impacts of more complex factors (nutrients, toxicants) and particularly the role of the elusive chemical complexations are far from understood. In addition, in the field the numerous abiotic factors are constantly interacting with the equally complex system of biotic factors (see the following section).
2.2
Biotic Habitat Factors: A Connected Complex
Studies on the impact of biotic factors on meiobenthic communities have increased since about 1975, while work on abiotic factors dominated studies of the 1960s and early 1970s. Today, it is recognized that biotic factors can have a massive influence on the population structure of meiofauna and the benthic habitat in general (Woodin and Jackson, 1979a,b). The array of biotic factors in sediments extends from dead organic matter (detritus) and biogenic structures to mucus aggregates, sedimenting plankton, bacteria, and, of course, living organisms. Detritus—the remains of plants and animals—is the main component of particulate organic matter, but fecal pellets, mucous agglutinations from exudates and excretions, as well as dissolved organic substances also contribute to this complex. In nature, this organic matter is inseparably connected to bacterial growth, colloids, aggregate and biofilm production and remineralization (Jørgensen et al. 1981; Kepkay 1994; Azam 1998). In a wider sense, the activities of other fauna, such as disturbance, competition and predation, also represent biotic habitat factors for meiofauna (Figs. 15, 16). Community aspects of this subject will be discussed in Chap. 9. Watling (1988, 1991) pointed out that the classical methods of sediment analysis, derived from geology, give a misleading, denatured and biologically rather irrelevant picture of the sedimentary habitat. The world of benthic animals is not characterized by the mineral particles but by the delicate, flocculent organic matter that binds them into a highly adsorptive interwoven fabric. Consideration of the organic material, bacterial colonization, biofilm production, pore-water chemistry and bioirrigation is necessary to achieve a natural conception of the real sediment.
38
2 The Biotope: Factors and Study Methods
Abiotic factors will predominate in extreme biotopes only, such as in the swash zones of exposed beaches (Hockin 1982b). However, even in the wrack zones of tidal flats, trophic or predator–prey relationships are of significant importance (Giere 1973; Reise 1987; Menn 2002b). Living organisms (bacteria, microalgae, other animals) also represent important environmental factors that structure the habitats of meiobenthic animals. Among the biotic factors, probably the least known component with a potential influence on the meiobenthos are the microscopic fungi. Biotic factors interact in complex ways and are difficult to separate and measure. This is probably why they have been investigated far less than abiotic parameters. For better elucidation of this biotic–habitat interaction, the various components are presented separately below. Interpretations of data from such investigations have often been controversial and thus there are few meaningful generalizations. While this has been the case for the better studied macrobenthos (see reviews by Pearson 2001, Wilson 1991), it is even more valid in the case of meiobenthos.
2.2.1
Detritus and Particulate Organic Matter (POM)
Dead organic particles from any trophic level of an ecosystem, as well as organic input from external sources, regardless of its size, are considered to be detritus. Much of the detritus found in sediment samples is derived from dead plankton organisms trapped by the huge filtering systems of shores (Riedl 1971; McLachlan 1989; Berelson et al. 1999; Ólafsson et al. 1999; see Sect. 2.1.3). Decaying phytoplankton blooms can also result in the deposition of a fluffy layer of phytodetritus on the sea floor in coastal, sublittoral and deep-sea areas. These unconsolidated organic deposits, often agglutinated by mucous secretions, enhance the bacterial activity after relatively short time periods (a few days to weeks) and can subsequently cause a significant increase in meiofaunal abundance and diversity (Thiel et al. 1988/89; Fleeger et al. 1989; Lambshead and Gooday 1990; Riemann 1995; Witte et al. 2003; Vanaverbeke et al. 2004). In Kiel Bight (Baltic Sea), the benthic degradation of a phytoplankton bloom was completed after only three weeks (Graf et al. 1982). Especially in surface-feeding meiobenthos, the positive response to these blooms is very direct, while subsurface feeders tend to show a more indirect reaction (onset of reproduction periods, etc.) (Ólafsson and Elmgren 1997). In shallow, well-illuminated bottoms, benthic macroalgae and seagrass meadows also provide an ample source of detritus known to promote abundant meiofaunal populations (Novak 1989; Blanchard 1991; Urban-Malinga et al. 2008). In return, meiofauna have been shown to stimulate the decomposition of plant litter (Findlay and Tenore 1982; Alkemade et al. 1992). Experimental observations indicate that the detritus is not indiscriminately ingested by meiobenthic animals: debris of brown algae is preferred over red algae (Giere 1975; Rieper-Kirchner 1989), while mangrove litter (with its rich tannin content) is less attractive than other plant debris (Alongi 1987b). The differing origins, the multitude of stimulating and inhibiting substances contained in detritus, and the diversity of the degradation processes led Tenore et al. (1982)
2.2 Biotic Habitat Factors: A Connected Complex
39
to concentrate on the “available,” attractive detritus in his ecological studies. The organic remains of decaying macrofauna have been shown to be attractants for saprobiotic meiofauna, especially nematodes (Gerlach 1977; Ólafsson 1992), causing patchy meiofaunal aggregations. As far back as 1942, Mare stated that the amount of particulate organic matter (POM) had a significant influence on the distribution of meiofauna, a fact which has since been confirmed by many authors (Lee 1980a; Tenore and Rice 1980; Tietjen 1980; Warwick 1989). In silty muds, the dry weight of the organic particles can reach 10% of a sample, while for sandy shores this value is often <1%. However, it has been demonstrated that even the microtopography of the bottom (tubes, ripples) influences the content of organics. Ripples enhance the particle flux by a factor of >2 (Huettel et al. 1996), while organic particles accumulate in the troughs and the depressions around tubes (Hogue and Miller 1981; Hicks 1989). To obtain a gross measurement of the total organic matter, the ash-free dry weight of the dried sample is generally the parameter most commonly used. This parameter is the mass loss observed upon the combustion of the dried sample at about 400 °C to constant weight, usually for 2 h (for a comparison of methods and a suggested standard procedure for marine sediments see Beyers et al. 1978). Care must be taken to ensure that the combustion temperature does not exceed 580 °C (400 °C according to other authors) in order to avoid the volatilization of sediment carbonates and thus incorrect results. This “loss on ignition” approach yields inaccurate results for sediments with a high content of clay. A more accurate discrimination between the different components of organic matter (e.g., organic carbon, proteins, lipids, carbohydrates) requires high-temperature oxidation (Bale and Kenny 2005). Another method to determine the organic carbon is to measure the reductive potential of the organic matter by titration. A problem inherent in the assessment of organic content and, consequently, relevant to the analysis of meiofaunal nutrition and distribution is the inclusion of living organisms—microalgae and animals—in the bulk measurement of organic carbon. This is prone to cause a biased conclusion for the microbial and trophic potential of the sediment. Separation of the detritus-linked substances from the live organisms is possible by measuring the ATP content. The procedure for extracting ATP from sediment samples performed by Karl and La Rock (1975) uses the sensitive reaction of the luciferine test for oxidizing substances. However, the ATP contents of organisms vary with their living conditions and ontogenetic events. Consequently, the procedure must be carefully calibrated and the data replicated. The difficulty involved in many of these methods is not so much the technical procedure or the sensitivity of the measurement, but the correct interpretation of the data obtained. For an appropriate conversion of ATP content to weight in (nematode) biomass studies, Goerke and Ernst (1975) report an average ATP concentration of 1.35 mg ATP g−1 wet wt of meiobenthos (nematodes). Unless the oxygen conditions in the sediment are not adversely diminished, it can be generally assumed that an increase in organic matter will enhance meiofaunal abundance but will also change the community composition and the microdistribution (Ólafsson 1992; Creutzberg et al. 1984). Austen and Widdicombe (2006) were able
40
2 The Biotope: Factors and Study Methods
to experimentally show that, at intermediate doses, organic enrichment leads to an increase in the number of meiofauna, confirming that Huston’s general model (Huston 1979) is also valid for meiofauna.
2.2.2
Dissolved Organic Matter (DOM)
Though often disregarded, dissolved organic matter represents a huge reservoir, 20–30 times greater than that of POM. It originates mainly from bacterial excretion and decomposition (Van Oevelen et al. 2006b), but also from the leaching of decaying plant and animal materials or exudation from bacteria and plants. Therefore, DOM in the mud of Spartina salt marshes reaches particularly high concentrations (Gardner and Hanson 1979, see Table 2.4). High DOM values are often due to essential fatty acids derived from plants. Secretion by meiofauna such as nematodes is another source of DOM (Moens et al. 2005). Labile nonhumic substances like sugars (glucose, galactose, sucrose), free amino acids (alanine, glutamic acid, aspartic acid, ß-glutaric acid) and the refractory humic acids are the principal organic molecules which become highly enriched in the pore water; their concentrations in pore water are often one or two orders of magnitude higher than those in the overlying water. In freshwater habitats, where the concentration of DOM is even higher than in marine environments, dissolved organic matter plays substantial roles as both food and a possible source of attractants and releasers of developmental signals (Tranvik and Jørgensen 1995; Thomas 1997). The fluvial input into the oceanic pool is considerable. Neritic coastal zones are richer in DOM than the deeper oceanic sediments (Table 2.4). Dissolved free amino acids (DFAA) are Table 2.4 Dissolved free amino acids (DFAA) and total dissolved organic carbon (DOC)—a comparison of open water and pore water concentrations DFAA (mmol l−1) Seawater Oceanic regions:
Coastal regions: Mud Estuarine regions Spartina salt marsh Fjord, bight Fjord, 40 m depth Freshwater Lewes Brook, UK Lake Balaton, Hungary
In the water column
In pore water
Reference
0.5–1.0
12–50
0.06–6.0
0.5–12.5
Jørgensen (1979); Jørgensen et al. (1980) Thomas (1997)
1.0–3.0 1.8–28.5 8.9 0.5–12.5 1.3–2.6
15–220 16.0–56.0 3.9–28.5 8.7–28.5 0.3–3.8
Stephens et al. (1978) Jørgensen (1979) Gardner and Hanson (1979) Thomas (1997) Jørgensen et al. (1980) Landén and Hall (1998)
0.6–1.3 0.04–0.5
30–60 40–90
Thomas (1997) Thomas (1997)
8.1
Farke and Riemann (1980)
DOC (mg l−1) Coastal regions (silt)
1.7
2.2 Biotic Habitat Factors: A Connected Complex
41
particularly enriched in the upper 0–2 cm and often disappear in sediment strata below 10 cm. Why is the concentration of DOM in the upper sediment layers so high compared to the overlying water despite these releasing processes? The deposition of degradable detritus is higher on the bottom and in the sediment while degradation is generally lower because of the frequent lack of oxygen. Hence, the sediment particles with their relatively large surface areas and considerable adsorptive forces “retain” DOM in the pore water of the sediment. “New” sand grains with sharp edges have been found to adsorb more glucose than older ones with “smoothed” surfaces (MeyerReil et al. 1978). The multitude of substances present in DOM necessitate detailed and complicated chemical analyses (Volk et al. 1997; Burdige 2002). DOM is actively absorbed by either transepidermal uptake or “drinking” (intestinal uptake). Thus, intensive and permanent contact with the pore water system favors the utilization of DOM by bacteria and meiobenthic organisms (see Tranvik and Jørgensen 1995). This explains why sediments with high DOM concentrations are favored by meiobenthos, primarily by the soft-bodied ciliates, turbellarians and annelids with large relative surface areas (Petersen et al. 1988); but significant uptake also occurs in nematodes. An interesting new aspect of DOM uptake is the capacity of some nematodes to utilize acetate, which they can metabolize into polyunsaturated fatty acids, a pathway usually restricted to algae and bacteria. Dissolved organic matter (DOM) is released from the sediment into the water column. The slow process of physical diffusion is accelerated by hydrodynamic forces (storms, currents and waves) and by sediment reworking through meio- and macrobenthic animals. This meiofauna-related increase in solute transport seems to be in the range of 1.5–3-fold (Aller and Aller 1992; Rysgaard et al. 2000). Through the concomitant exchange processes, the activities of benthic animals can enhance the physical diffusion of sediment-bound substances by orders of magnitude, thus counteracting the adsorptive and accumulating sedimentary processes mentioned above (see also Hylleberg and Henriksen 1980; Aller and Yingst 1985; Kristensen et al. 1995). Habitats with high DOM concentrations in the sediment pore water such as flats, deltas and estuaries in the marine realm and eutrophic lakes in freshwater often attract meiobenthos, since DOM is intensively and primarily taken up by bacteria (Moriarty 1980), from which, in turn, the bulk of DOM is recycled (Van Oevelen et al. 2006a). Today, uptake rates of DOM are best indicated by radiotracer methods. Aside from the trophic aspects, the following section will point out that extracellular organic substances have an important structural effect, they cause agglutination of sediment particles, can increase sediment stability, and serve as signal substances that have informative value to aquatic organisms (Thomas 1997; Bale and Kenny 2005).
2.2.3
Mucus, Exopolymers, and Biofilms
In both marine and freshwater systems, exopolymer secretions (EPS) and mucous aggregates are considered key structures for the concentration and exchange of nutrients, for the formation of flocculent matter, often termed “snow,” and for the
42
2 The Biotope: Factors and Study Methods
nutrition and even transport of meiofauna (Yallop et al. 1994; Shanks and Walters 1997; Azam 1998; Wotton 2005). Organic carbon bound in colloids seems to represent a major carbon reservoir in seawater (Farke and Riemann 1980, Kepkay 1994). Microbial activity and coagulation by physical forces transform the colloids into aggregates to which flocculent detrital material adheres, forming nutrient-rich bioreactors. Mucus consists mainly of carbohydrates, especially polysaccharides and glucoproteins; a minor portion consists of loosely bound and labile amino and fatty acids (Meyer-Reil 1994). Passow (2000) regards transparent exopolymer particles as a “distinct group of polysaccharides” formed from dissolved precursor material. They coagulate particularly after phytoplankton blooming periods, but can also be experimentally generated. Condensed as macroaggregates (Logan et al. 1995), they contribute to the “marine snow/lake snow phenomenon” to which bacteria and meiofauna are closely associated (Shanks and Walters 1997; Heissenberger et al. 1996; Simon et al. 2002; Wotton 2005, see Chap. 7). Adhering detritus particles solidify this unconsolidated matrix and accelerate sedimentation. The mucous biofilm that develops on the bottom is excreted mainly by benthic microorganisms, especially bacteria and diatoms (Meyer-Reil 1994; Smith and Underwood 1998), as well as by benthic meio- and macrofauna. Particularly the mucus trails secreted by nematodes, but also those secreted by harpacticoids, enhance bacterial growth (Moens et al. 2005; De Troch et al. 2005). Decho and Lopez (1993) speak of an “exopolymer microenvironment” of bacteria. Its rich nutrient content (especially polysaccharides) and its capacity to affect the flux of dissolved organics favor colonization with rich bacterial stocks. An EPS coating on sediment particles increases their bioavailability for particle-ingesting meiofauna. The specific composition of the biofilm seems of particular relevance to the growth, reproduction and developmental phases of meiofauna (Brown et al. 2003; Dahms et al. 2007). This underlines the considerable trophic importance of exopolymer secretions for the micro- and meiofauna, from allogromid foraminiferans in tidal flats (Bernhard and Bowser 1992) to harpacticoids copepods and nematodes (Koski et al. 2005; Moens and Vincx 1997a; Riemann and Helmke 2002; Wotton 2005), from freshwater ecosystems to the deep sea. Labile components such as free amino acids are even thought to act as sources of information and communication for the microbenthos (Meyer-Reil 1994; Thomas 1997). While this superficial mucous biofilm is relevant nutritionally to meiofauna, it also modifies habitat characteristics, enhancing the cohesiveness and reducing the erodibility of the substratum (Yallop et al. 1994; Miller et al. 1996; Black 1997). Through this biostabilization of sediments, dissolved organics bound in mucus derivates also contribute to the heterogeneous spatial and temporal small-scale occurrence of meiofaunal populations. Decho (1990) contended that extracellular polysaccharides form “an extensive matrix of amorphous organic material which may provide the bulk of carbon sources for many benthic organisms.” In his seminal article, Azam (1998) states that microorganisms do not experience water as their ambient medium in the sediment, but rather a gel-like matrix of suspended polymers and colloids. He speaks of an “organic matter continuum” connecting DOM via colloids, aggregates and particle-embedded “suprapolymers” to POM. The carbohydrates that represent the
2.2 Biotic Habitat Factors: A Connected Complex
43
bulk of colloidal exopolymers can be readily measured (Underwood et al. 1995), but the physical structures of these mucoid and gelatinous substances in the void system, which determines the “world” of meiobenthos, are hard to demonstrate. The development of sophisticated methods (often fluorescence dyes) was required in order to visualize, manipulate and study these delicate mucous aggregates (Heissenberger and Herndl 1994; Heissenberger et al. 1996; Schumann and Rentsch 1998; Mari and Dam 2004; Neu et al. 2002). Today, it is conceivable that the study of mucus secretions and mixed colloidal aggregations and the comprehension of their ecological role will gain importance in future meiobenthological research. It should be promoted more, despite the inherent technical problems (see Murray et al. 2002). Discussing the role of dissolved organics and exopolymers in freshwater, Thomas (1997) stated “The assumption that the energy flow in aquatic ecosystems can be quantified solely by measuring rates of photosynthesis, ingestion of solid food and its digestion by higher organisms, is invalid.” This certainly also holds true for the marine realm (Meyer-Reil 1994).
2.2.4
Bacteria
Bacterial abundance and biomass are several orders of magnitude higher (abundance: 1000×) in the sediment than in the water column. This huge bacterial stock is closely linked to organic debris (detritus) and usually accounts for 4% of the total organic carbon (Jørgensen et al. 1981; Kemp 1990; Schallenberg and Kalff 1993). Experiments have demonstrated that it is the bacterial film, not the detrital substrate, that is preferably utilized by the “detritivorous” meiofauna (Fenchel 1969, 1970; Hargrave 1972; Meyer-Reil and Faubel 1980). Hence, the rich coating of detritus with bacteria may attract high concentrations of meiofauna. As the plant debris ages, the bacterial colonization grows and the protein content increases. This, in turn, makes aged detrital particles more attractive to meiofauna (Warwick 1989). Also, animal remains decomposed by bacterial degradation can attract meiofauna (nematodes) in the sediment so long as oxygen is present (Gerlach 1977a; Ólafsson 1992; Debenham et al. 2004). Other centers of bacterial growth include gradients, especially at the water/ sediment interface and in the sediment at chemical gradients. Many sulfideoxidizing microbes tend to concentrate in the oxic/anoxic interfaces of decomposing detrital particles, where they are exposed to irrigational fluxes and sediment reworking by infauna (Fenchel and Riedl 1970; Fenchel 1996). Here huge populations of “sulfur bacteria” are of considerable importance to meiofauna (Yingst and Rhoads 1980; Jørgensen and Bak 1991). Sulfur bacteria also thrive in fecal pellets produced by meio- and macrobenthos. Meadows and Tait (1985) found that bacterial numbers in fecal pellets in deep-sea sediments were several orders of magnitude higher than those in the surrounding sediments. In the anaerobic depths of nutrient-rich sediments, sulfate reducers and mat-building cyanobacteria develop rich populations (Jørgensen and Bak 1991; Stal 1991, Ramsing et al. 1993).
44
2 The Biotope: Factors and Study Methods
Attracted by nutrient aggregations, bacteria often occur in micropatches. These, in turn, are often the basis for the notorious patchy field distributions of bacterivorous meiofauna, even in seemingly homogeneous sediments (Blackburn and Fenchel 1999). In lentic coastal sediments the rich detritus/bacteria complex especially favors nematodes and other taxa linked to the “detritus/bacteria-based food chain.” In contrast, the more “microalgae-based” harpacticoids are less dependent on these factors (Montagna et al. 1989). A warmer climate or season favors bacterial growth. In temperate regions, peak populations occur in summer. In spring and summer, bacteria develop a characteristic extensive mucus coating and provide a rich mucoid nutritive source for deposit feeders (see below). The capacity of meiofauna to distinguish between various microbes (bacteria, microfungi) and to select certain groups or even strains as food has repeatedly been shown experimentally. This is probably the case for all meiobenthic groups, e.g., ciliates (Fenchel 1969), nematodes (Tietjen and Lee 1977; Tietjen 1980a; Moens et al. 1997, 1999b), harpacticoids (Carman and Thistle 1985), oligochaetes (Chua and Brinkhurst 1973; Dash and Cragg 1972), and polychaetes (Gray 1966a,b, 1971). Montagna (1995) found that meiobenthic rates of feeding on bacteria vary with the developmental stage, in terms of both the amount and the quality of the microbial food. In some cases, the functional correlation between the structures of the mouth parts or buccal armatures and the shapes of the bacteria have been demonstrated in more detail (Wieser 1959, 1960; Jensen 1983, 1987a for nematodes; Marcotte 1984, 1986a; Romeyn and Bouwman 1986 for harpacticoids). Selective bacterivory of meiobenthos is thought to result in a bacterial-induced zonation (mainly nematodes), as demonstrated in laboratory sediment tanks (Boucher and Chamroux 1976), or is a mechanism for microniche segregation (Moens et al. 1999b). Considering the extremely high bacterial productivity (on average 324 mgC m−2 d−1, Kemp, 1990) the bacterial stock does not seem to be limited under natural conditions by meiofaunal grazing (see Sect. 9.3). Quantifying the community structure and the abundance of bacterial microorganisms is a difficult task and not without serious methodological flaws. Bacterial number or biomass is typically underestimated and potentially inaccurate by a factor of two (Kemp 1990). The sonification of homogenized sediment in order to count bacteria is only 65–95% efficient. Direct counting of stained (vital dyes, fluorescent dyes) bacterial cells on the particle surface, although tedious, remains one of the more reliable procedures (DeFlaun and Mayer 1983). Even DAPI staining resulted in a considerable underestimate of the bacterial abundance (70%) and biovolume (60%) (Suzuki et al. 1993). Bacterial volume, a neglected parameter of relevance to metabolic processes, can be calculated more reliably through ultrastructural scanning methods (Kaye 1993). According to a detailed working protocol elaborated by Epstein and his group, the combination of careful sonification with tritiated thymidine or CTC labeling will result in particularly efficient enumeration of about 90–95% of the bacteria present (Epstein 1995; Epstein and Rossel, 1995a,b; Epstein et al. 1997). Moriarty (1980) recommends the determination of muramic acid as a good basis for calculations since this substance is a cell wall component of almost all prokaryotes. Another indirect method used to quantify sediment bacteria is the calculation of their biomass by phospholipid or ATP analysis (Findlay et al. 1989, Köster
2.2 Biotic Habitat Factors: A Connected Complex
45
and Meyer-Reil 2001). Schmidt et al. (1998) showed that scaling bacterial abundance to the fluid volume of pore water within the sediment yields a much greater consistency than traditional relations to dry sediment mass. Molecular screening techniques with general bacteria probes or the incorporation of radiolabeled markers (e.g., tritiated thymidine) are new methods that are increasingly being applied for quantitative recordings (e.g., Ramsing et al. 1993). Homogenization and subsequent Percoll flotation (Sect. 3.2.2) are also valuable improvements. The numbers of sediment bacteria can vary considerably depending on the evaluation method, on the local microtopography and physiography, and on the sediment quality, water content and climate, but are mostly in the range of 108–109 cells per ml sediment. Hence, the figures in Table 2.5 probably cannot be generalized. Only in the past decade have modern methods yielded more reliable quantitative abundance data. Bacterial biomass, biovolume and productivity are more indicative than abundance data of the importance of bacteria as an eminent ecofactor for meiofauna. In many sediments, bacterial biomass and production is equal to or exceeds that of the macrofauna (Kemp 1990; Bergtold and Traunspurger 2005). Usually, the bacterial density in sediments corresponds to the amount of degradable organic matter and is reciprocally related to the degree of exposure and the particle size of the sand fraction (Köster and Meyer-Reil 2001). Muddy bottoms and sea grass beds are microbially richer than sand, just as the wrack zone of a beach is richer than its surf zone or its sublittoral bottom. Moriarty (1980) found five times the amount of bacteria in seagrass beds than in the adjacent open sediment. The dependence of bacterial abundance on granulometry is demonstrated by the vertical profile in the North Atlantic (Vanreusel et al. 1995b) from the shelf to the continental slope (70–1,500 m depth). The sandy slope samples contain several orders of magnitude fewer bacteria than the deeper slope (Table 2.5). A rich supply of detritus and oxygen, for most microbes, makes the surface layers of the bottom more attractive than groundwater layers or the anoxic depths. Light-dependent cyanobacteria aggregate in mostly sandy sediments underneath a thin surficial mucus film (Yallop et al. 1994). On the surfaces of seagrass beds, 18% of all organic substances measured were contributed by live bacteria (Moriarty 1980), a significant nutritive amount, even for larger animals. In a sandy tidal flat, almost all of the carbon input was attributed to microbes (Joergensen and Mueller 1995). Typically, bacterial biomass equals about 4% of the total organic carbon (Kemp 1990). However, the world of meiobenthic animals is determined by a three-dimensional pattern of microniches and particle surfaces. Even at the microscopic scale, sediment particles are colonized by qualitatively and quantitatively different bacteria; this in turn implies differences in the compositions of bacterivorous meiofauna. Colonization of sand grains seems to be proportional to surface area. One mm2 of grain surface may be populated by up to 260 × 103 bacterial cells (Anderson and Meadows 1969)! This enormous number of bacteria is concentrated on only a very small portion of the huge overall surface (between <1 and 5%). The restriction may be caused by the limited diffusion of oxygen and nutrient molecules through the tortuous and long distances between densely packed particles (Watling 1991), and by micropatches of nutrients, as demonstrated by Blackburn and Fenchel (1999).
Mangroves Subtidal Antarctic Seagrass bed Mediterranean Coral reef sand Continental shelf slope Deep-sea mud Mud Lake sediment
Brackish Estuary
Salt marsh
Mud to fine sand Tidal flats
Beach sand
Alongi (1990a)
Danovaro (1996) Koller et al. (2006) Alongi (1990a) Schmidt et al. (1998) Schmidt et al. (1998)
1–6 × 108 0.4–5.2 × 109 1.6 × 109 1.1–1.5 × 109 2–5 × 108
Meyer-Reil et al. (1978) Findlay et al. (1989)
Westheide (1968) Anderson and Meadow (1969) Meyer-Reil et al. (1978) McLachlan (1985) Moriarti et al. (1985) Epstein et al. (1997)
~10 × 1010
2 × 1010 1 × 109
9.8 × 10 14 × 107 to 1.1 × 109 4 × 108 to 28 × 108 0.3 × 108 to 2.5 × 108 0.5 × 109 1 × 109 6
Table 2.5 Abundance of bacteria in various marine sediments Habitat Bacteria (per g dry sed.) Reference
Uthicke (1998) Vanreusel et al. 1995b Thiel et al. (1988/89) Tietjen (1992) Schallenberg and Kalff (1993) Starink et al. (1996)
Dietrich (1999)
1.5 × 105
1 × 108 1.2–3.0 × 106 2–3 × 108 0.1–3 × 107 6.3 × 109 4–12 × 1010
Epstein and Shiaris (1992) Kemp (1988) Steward et al. (1992) Steward et al. (1992) Dietrich (1999) Cammen (1982) Schmidt et al. (1998) Kemp (1988)
McLachlan (1985)
0.5 × 108 to 3.9 × 108
7.5 × 108 2 × 109 1.2 × 109 (mud) 1.7 × 109 (sand) 3 × 109 2 × 109 4–17 × 109 1.7 × 109
Boaden (1964) Kemp (1990)
Reference
50 × 103 to 500 × 103 108 to 109
Bacteria (per cm3 wet sed.)
46 2 The Biotope: Factors and Study Methods
2.2 Biotic Habitat Factors: A Connected Complex
47
However, mechanical forces also seem to be responsible for the patchy colonization of bacteria. Microscopic inspections of sand grains have revealed the heterogeneous colonization of grain surfaces, with rich microbial clusters in depressions and cracks and barren areas along the edges (Fig. 2.12; see Meadows and Anderson 1966, 1968; Weise and Rheinheimer 1978; DeFlaun and Mayer 1983). On these exposed edges, the destruction of bacteria through the intense agitation of grains by waves or heavy rainfalls is particularly important, while they have more protection in areas of low relief. Thus, exposure and particle size may contribute to the relative scarcity of bacteria in the exposed upper eulittorals of beaches compared to sheltered flats. Large grains (>300 µm diameter) with fairly smooth surfaces are inhabited by bacterial flora that differ quantitatively and qualitatively from those on smaller particles with many crevices and depressions (Marcotte 1986a) and those of the water column. The relatively small surfaces of silt particles <10 µm diameter were found to be devoid of bacteria (DeFlaun and Mayer 1983). According to Azam (1998), many bacteria are found in the transparent gelatinous fabric filling the void system, and are not attached to solid substrates. This would explain why the fresh sediment volume or the fluids contained therein are more appropriate as units to relate to bacterial numbers than the sediment dry mass (Schallenberg and Kalff 1993; Schmidt et al. 1998). The large fluid component may particularly contribute to the rich bacterial populations in fine, silty sediments. Data on the production and carbon flux of sediment bacteria are given in Sect. 9.4, but in nature only a small proportion represents active cells that contribute to production (Novitzky 1983), which reduces the value of bacterial counts. In the profound experiments by Van Oevelen et al. (2006 a,b) from an intertidal flat, which integrated tracer experiments and modeling, it became evident that most bacterial production (65%) was respired and remineralized back to dissolved organic carbon, while only 3% was consumed by meiobenthos and 24% by macrobenthos. Considering the
a
b
Fig. 2.12a–b Colonization of a sand grain by microorganisms. a Overall aspect. b Detail. (After Meadows and Anderson 1966)
48
2 The Biotope: Factors and Study Methods
extremely high bacterial productivity (on average 324 mg C m−2 d−1, Kemp 1990), the bacterial stock does not seem to be limited under natural conditions by meiofaunal grazing (see Sect. 9.3). In microcosm experiments (Traunspurger et al. 1997), grazing by freshwater nematodes stimulated bacterial activity rather than having a strong grazing impact. In sediments of the North American Great Lakes, bacteria were calculated to contribute about half of the organic matter taken up by metazoa (Schallenberg and Kalff 1993). The numerous relationships between bacteria and meiofauna, reviewed by Coull and Bell (1979), Tietjen (1980a), Kemp (1990, 1994) and Montagna (1995), will be detailed later when dealing with the position of meiofauna in the benthic food web (Sect. 9.4). Here, it is concluded that the significant impact of bacteria on the habitat of meiofaunal organisms makes cooperation between meiobenthologists with sediment microbiologists indispensable to obtaining a profound ecological understanding. More detailed reading: Detritus: Warwick (1987); dissolved organic matter: Jørgensen (1976); Sepers (1977); Stewart (1979); Aller and Yingst (1980); Ferguson (1982); Stephens (1982); Reichelt (1991); Aller and Aller (1992); Thomas (1997); mucus and exopolymers: Decho (1990); Azam (1998); Simon et al. 2002; Krumbein et al. (1995); Wotton (2005); especially meiobenthos: Tempel and Westheide (1980), Montagna (1984); bacteria: Tenore and Rice (1980); Tietjen (1980); Kemp (1990, 1994); Epstein (1997a, b); reviews: MeyerReil (1994); Murray et al. (2002); freshwater: Schallenberg and Kalff (1993).
2.2.5
Microphytobenthos
Unicellular microalgae, mainly diatoms and dinoflagellates, act, together with bacteria, as biotic habitat factors for meiofauna through their trophic value and mucus production. In coastal sediments, the microphytobenthos has a central role in carbon flux (MacIntyre et al. 1996, Middelburg et al. 2000). In a study from a Mediterranean seagrass bed, microalgae accounted for 25% of all organic carbon, while bacteria only contributed 5% (Danovaro 1996). Microphytobenthos provided 15–20% of total primary production in brackish muds (Gerbersdorf et al. 1999). Even in the brine network of Arctic sea ice, diatoms have been calculated to represent 26% of the total biomass (Gradinger et al. 1999; see Sect. 8.1.1). Hence, microphytobenthos has been considered a “secret garden” of enormous ecological importance in shallow water habitats (Miller et al. 1996). In recent years relevant papers on benthic diatoms underlined this most important food source for the benthic fauna. Diatoms prefer the upper, light-exposed layers of sheltered, often silty bottoms (Wiltshire 2000) where they can grow to considerable densities (20 × 103 cells cm−3 sand or up to 16 × 103 cells cm−2, and sometimes even up to 7 × 107 cells cm−3: Fenchel and Straarup 1971). In flats on the island of Texel, about 5 × 106 diatom cells were counted per cm2 in mud, and only 2 × 106 cells per cm2 in sand (Yallop et al. 1994). Other authors found densities in the same range (see Round 1971). From coral reef sand, 1.7–2.4 × 10−6 cells cm−3 dry sand has been reported (Uthicke 1998). In the polar sea ice, diatoms preferentially grow in the deeper layers. Since the penetration
2.2 Biotic Habitat Factors: A Connected Complex
49
of light is only 1–2 mm in mud and up to about 2 cm in coarse sand (long-wave red light; for details and modern methods see Lassen et al. 1992, Kühl et al. 1994), in the presence of light, the uppermost 0.2 mm harbored most of the diatoms, while in the dark the distribution was more homogeneous (De Brouwer and Stal 2001). However, living diatoms and flagellates are repeatedly found down to 5 cm, and have even been recorded at horizons as deep as 20 cm (Taylor and Gebelein 1966; Steele and Baird 1968; Fenchel and Straarup 1971; Wasmund 1989). Correspondingly, at a depth of 5 cm, 20–50% of the surficial chlorophyll concentrations can still been measured. The layer between 0.5 and 2.0 cm depth still provided 40–60% of the diatom biomass (De Jonge and Colijn 1994). Viable diatoms (and other algae) have also been discovered in considerable numbers in the sediment of aphotic subterranean aquifers (Kuehn et al. 1992). For beaches, this unusual distribution for photosynthesis-dependent cells is interpreted as being a result of sediment turbation; however, vertical migrations have also been considered (Janssen et al 1999). Long-lasting phases of passive resting and intermittent metabolism have also been suggested. There is even supposition that some groups of diatoms can live heterotrophically. Detailed information on the vertical fine-scale distribution of diatoms and associated chemical gradients has been obtained through the use of a “cryolander,” a corer which, in combination with freezing techniques, can retrieve undisturbed mm-samples (Wiltshire et al. 1997). In tidal flats, the highest population densities of diatoms have been recorded in spring and partly in autumn (Asmus and Bauerfeind 1994). In the Danish Wadden Sea, temperatures correlated with the photosynthetic rate of diatoms, while high salinities and pH values decreased it (Rasmussen et al. 1983). In superficial sediments and depending on the season, microphytobenthos contributed between 3 and almost 17 g C per m2 y−1 (De Jonge and Colijn 1994) in estuarine superficial sediments; Round (1971) reported a production of marine diatoms of 100–200 g C m−2 y−1 in the intertidal. Considerably lower values were reported by Guarini et al. (1998) from a French bay, but even so, about 30–90% of all of the chlorophyll a suspended in the water column was contributed by microphytobenthos. Seasonally varying values of between 90 and 170 g C m−2 y−1 were recorded as the net production of diatoms in coral reef sand (Uthicke 1998). Microphytobenthos in tidal flats apparently have a specific spatial microdistribution (Round 1971) which can be associated with various macrobenthic zones (Asmus and Bauerfeind 1994). Saburova et al. (1995) stated that the differentiated spatial structure of diatoms and dinoflagellates in a polar tidal flat was due to the interaction of the biogenic, interspecific effects (small scale) with the granulometry (meso scale) and the emersion period (large scale); see Fig. 2.13. Many ciliates, harpacticoids, nematodes and oligochaetes often feed exclusively on diatoms. With their specialized mouth structures they can differentiate between various algal species of different shapes (see Fig. 5.20); Brown and Sibert 1977; Pace and Carman 1996; Azovsky et al. 2005; De Troch et al. 2005). The close linkage of many meiofauna, especially the “microalgae-based” harpacticoids (Montagna et al. 1989), to diatoms as a predominant food source was experimentally demonstrated by the rapid and preferred uptake of labeled diatom carbon (Riera et al. 1996; Middelburg et al. 2000), and is evidenced by the numerous test remains in copepod fecal pellets.
50
2 The Biotope: Factors and Study Methods INTERTIDAL
LEVEL
SPECIES Lower
Middle
Upper
Nitzschia sp. Amphiprora paludosa Gymnodinium sp. Amphidinium britannicum Amphora proteus Navicula salinarum Achnanthes sp. Diploneis sp. Navicula pupula Amphora coffeaeformis 20 %
Fig. 2.13 Zonation of benthic diatoms along an intertidal profile in the White Sea. (After Saburova 1995)
In the upper sediment layers numerous spatial and temporal microscale processes tightly link microalgae and meiofauna. The occurrence of diatoms can determine the microdistributions of meiobenthic food specialists (e.g., peaks of naidid oligochaetes in tidal flats coincide with spring blooms of diatoms, Giere and Pfannkuche 1982). Diatom patches were found to correlate directly with aggregations of meiofauna (Montagna et al. 1983; Blanchard 1990). This is understandable considering the intensive utilization rate of microphytobenthos by meiofauna: Escaravage et al. (1989) calculated that 27% of all benthic microalgae were being devoured by meiofauna in a shallow lagoon, 38% in oyster beds, and as much as 89% in seagrass beds. In a Louisiana tidal flat, Pinckney et al. (2003) even found that the meiofaunal grazing rate sometimes exceeded the biomass of benthic microalgae, although, in terms of productivity, the meiobenthos grazed only 17% of the net primary production. The corresponding value in a polar tidal flat was 3–11% (Azovsky et al. 2005), and in a temporarily closed South African estuary 11% (Nozais et al. 2005). In sublittoral sites Montagna et al. (1995) found indications of a diet shift of meiofauna away from the preferred microalgae to bacteria when the algal supply became scarce. This nutritive linkage also explains the decreases in various meiobenthos after experimental inhibition of diatom growth (Hentschel and Jumars 1994). The attractiveness of microalgae as food for meiofauna may be based on their high content of polyunsaturated fatty acids (PUFAs) (Coull 1999, see Sect. 2.2.2). A high proportion of 16:1ω7 fatty acids can even be used as a marker indicating that diatoms are a preferred nutritional source.
2.2 Biotic Habitat Factors: A Connected Complex
51
However, even during periods of high grazing pressure, microalgal productivity was more than sufficient to sustain the meiobenthic consumers. The degree of control meiobenthos exerted on the diatom stock varied greatly. The diatom production in a network of brine channels in Arctic sea ice was in the same range as the ingestion rates of the numerous sympagic (sea-ice-living) meiofauna. Thus, in this special habitat, it was thought to be controlled by the meiofauna (Gradinger et al. 1999). Apparently the meiofaunal impact depends on regional and climatic factors, but also on the specialization for particular food species (Azovsky et al. 2005). Microphytobenthos is not only of considerable nutritional importance to meiofauna; it also alters the structure and chemistry of the sediment. Through their intense production of mucous substances, the microphytobenthos compact the sediment and enhance its cohesiveness, reduce erosion, and generally modify the activities of the fauna (DeFlaun and Mayer 1983, Madsen et al. 1993, Miller et al 1996; Black 1997; Widdows et al. 2000). This biostabilization by diatoms has an indirect trophic effect. Mucus can form dense coatings on the sediment which, in turn, promote bacterial growth (see above). On the other hand, massive macrofaunal bioturbation can increase the production of microphytobenthos by enhancing nutrient fluxes (Lohrer et al. 2004). The assimilatory oxygen production of microalgae lowers the oxic/sulfidic threshold diurnally with considerable consequences for the vertical distribution of meiofauna (Fig. 2.8b; page 29). Studies by Ólafsson et al. (1999) and Josefson and Hansen (2003) drew attention to another group of microalgae in the sediment. In Scandinavian coastal sediments they found that common planktonic species occurred regularly in the upper sediment layers, thus providing an additional trophic source for meiobenthos. Especially for copepods and ostracods, but also for nematodes, settling phytoplankton seems to be an important food source (Modig et al. 2000), either directly or via an increase in bacterial decomposers. Microphytobenthos is frequently considered to consist solely of benthic diatoms. However, it is becoming increasingly clear that benthic flagellates, mainly heterotrophic Dinoflagellata, are also relevant (Patterson et al. 1989; Hoppenrath 2000; Patterson et al. 1989), and must be considered when dealing with phytobenthos in meiobenthic habitats. In North Sea flats of the Netherlands, there were 50–300 × 103 flagellates per ml surface sediment (Bak and Nieuwland (1989), which is somewhat more than the measurements obtained by Epstein and Shiaris (1992) for flagellates on a muddy flat near Boston (10 × 103). Mangrove mud contained about 150 flagellates per cm2 (Alongi 1986). In brackish bacterial mats, Fenchel (1969) often found a flagellate abundance of 100 to >100 × cm−2. Voids of sand, especially exposed sand, seem less densely populated: Helsingør beach (Denmark) yielded several 100 × cm−2; sandy flats of the island of Sylt only about 100–200 interstitial flagellates cm−3 (Hoppenrath 2000), belonging to about 140 taxa; and a beach at Roscoff (France) exhibited up to 2,000 mesopsammal flagellates cm−2 (Dragesco 1965). Sea ice harbors a considerable number and biomass of active flagellates, with a production only somewhat lower than that of diatoms. Freshwater sites also seem to be rich in flagellates, but numbers fluctuate heavily with season: in eutrophic limnic sediments, between 1–16 × 103 cells cm−3 were retrieved (Gasol 1993; Starink et al. 1996, see
52
2 The Biotope: Factors and Study Methods
also Kemp 1990). In oligotrophic lake sediments mean densities of about 80 × 103 flagellates were observed (Bergtold and Traunspurger 2004). As with bacteria, quantitative data may vary by orders of magnitude depending on the extraction methods used. Flotation methods yielded efficiencies of between 73 and 100% (Alongi 1986; Starink et al. 1994). Fixation can cause considerable shrinkage of both diatoms and microflagellates and has to be corrected for in order to avoid inaccurate estimates of biovolume and thus carbon biomass (Menden-Deuer et al. 2001). Closely linked to bacteria, most of the dense populations of flagellates in tidal sediments are heterotrophs grazing intensively on bacteria. However, at least in freshwater, the immense amounts of bacteria produced appear to be only marginally consumed (up to 5%, Kemp 1990; Starink et al. 1996; Hondeveld et al. 1992; Hoppenrath 2000). Using a new fluorescence technique, Starink et al. (1994) calculated that in lake sediments up to 70 bacteria were consumed per protist per hour, while in marine sediments the corresponding consumption was only around five bacterial cells. Microdistribution and abundance patterns also suggest a close trophic correlation between flagellates and ciliates (Santangelo and Lucchesi 1995), a link that had been observed earlier by Fenchel (1969). Kemp (1990) emphasized the preferred occurrence of flagellates in organic-enriched detrital surface layers of the sediment, while ciliates were more linked to the interstices of sand. A differentiating trophic specialization within the microphytobenthos was evident in tidal sand: Epstein et al. (1992) found that ciliates devour 93% of the dinoflagellate production, in contrast with only 6% of the diatom production. Dietrich (1999) contends that in brackish sediments rich in organic matter, about half of the flagellate stock is consumed by meiobenthos. More detailed reading: Round (1971); Patterson et al. (1989).
Box 2.6 Microphytobenthos: The Garden of Meiofauna About 20% of all organic carbon is produced by diatoms and flagellates in the uppermost millimeters of shallow sediments. Some 103 to 104 cells live in each cm3 of sediment. Combined with their high nutritive value, this emphasizes their central role in the world of meiofauna, where they often mediate between bacteria and meiofauna. Through their oxygen production and light-dependent migrations they provide the sediment layers with oxygen. Many meiofaunal groups have specialized on diatoms as a food source and linked their population dynamics to the annual peaks in their production (spring and autumn in temperate regions). Benthic, mostly heterotrophic flagellates are important bacterivores, however, their nutritional role for meiobenthos is not well known. Particularly in the surface layers, where both microphytobenthic groups co-occur, they can regulate the temporal and microdistributional patterns of meiobenthos. Due to their rich mucus secretion, diatoms produce a nutrient film for bacteria and an effective coagulant that protects sediment particles against erosion.
2.2 Biotic Habitat Factors: A Connected Complex
2.2.6
53
Higher Plants
Sessile macroalgae and seagrasses have a physically structuring effect that influences meiofaunal settlement and distribution (Wieser 1959b). The reduction of sediment agitation and the enhancement of particle suspension under a plant canopy favors meiofaunal abundance. Culms, thalli, mangrove pneumatophores, and holdfasts provide numerous niches and protection for small animals. These plant structures expand the available living space for meiobenthos from the sediment into the water column and into the phytal (see Sect. 8.5). Delicately branched algae or the fuzzy culms of seagrasses are more densely inhabited by meiofauna than the smooth thalli of algae or the blades of seagrass. Conversely, minute epigrowth organisms on plants (e.g., mucous tubes) favor meiofaunal colonization (Peachey and Bell 1997; Gwyther and Fairweather 2002). Plant roots and shoots have a similar structural effect in the sediment. Thus, structural complexity is often positively correlated with meiofaunal abundance and diversity (Remane 1933; Hicks 1985; Hall and Bell 1988; Hull 1997). This has also been experimentally tested using artificial substrates of various complexities (Atilla et al. 2005). In addition, there are also chemical and nutritional effects by which plants can influence the habitat conditions of meiofaunal biotopes. Enhanced bacterial growth at the frequently damaged and leaching frond ends of plants indirectly promotes the trophic possibilities for meiobenthos. Live plant roots have been found to metabolically create a favorable micro-oxic gradient system in their surrounding sediments (Teal and Kanwisher 1961; Lee et al. 1999), enhancing the density and heterogeneity of (nematode) meiofauna (Osenga and Coull 1983). Algal cover on soft bottoms was found to favor the development of meiofauna in mesocosm experiments (Ólafsson et al 2005). When decaying, the leaves of seagrasses represent an attractive and important source of valuable detritus (see Sect. 2.2.1). An aggravating impact on the meiobenthos by mechanical disturbances has been shown by Hicks (1989). His field experiments indicated that (artificial) seagrass not only promotes meiofaunal populations, but can also disturb mainly epibenthic meiofauna assemblages, probably through the sweeping action of the blades and alteration of the microtopography (depressions, ripples). Lower numbers of meiofauna on mangrove pneumatophores as compared to mimics may result from the secretion of anti-fouling substances produced by the plants (Gwyther and Fairweather 2005). A discussion of other effects of plants relevant to meiofaunal habitat will be presented in Sect. 8.5, which deals with the phytal.
2.2.7
Animals Structuring the Ecosystem
The effects of macrobenthos on the habitat conditions of meiobenthos are extremely variable, species-dependent and are often not clearly delimited (Ólafsson 2003). They comprise both negative interactions and positive, facilitative effects.
54
2 The Biotope: Factors and Study Methods
Among the negative interactions, we can differentiate: (a) mechanical disturbance; (b) reduction by sediment ingestion; and (c) alterations of the chemical milieu. (a) Mechanical disturbances can be produced in areas with water currents by protruding macrofaunal tubes, which, enhance the boundary friction of the sediment and thus can exert an erosive force that negatively affects meiofauna (Eckman 1979; Gamenick and Giere 1994; Widdows et al. 2000; Ólafsson 2003). The spacing, the density and radius of the tubes largely govern whether the network of animal tubes will impair or promote meiofaunal life (see below). Those meiofaunal groups with a more passive, strictly sediment-bound lifestyle (nematodes, annelids) will be less affected, while the more epibenthic and temporarily suspended harpacticoids are more strongly affected (McCall and Fleeger 1995). Mechanical disturbance is caused also by disruption and reworking of the sediment, due to the digging of horse shoe crabs, rays, crabs, molluscs and sea urchins, the pipetting of Tellina (Bivalvia) siphons, and reworking by anthozoans, polychaetes, priapulids, amphipods, echinoids, fish and birds (Creed and Coull 1984; Reise 1987; Warwick et al. 1990 b; Hall et al. 1991; Ólafsson and Moore 1992; Ólafsson and Ndaro 1997; Aarnio et al. 1998). Meiofauna is usually considered relatively insensitive to disturbance and less persistently affected than macrofauna (Alongi 1985; Austen et al. 1989; Hall et al. 1991). Nevertheless, in open tidal flats, lugworm activity (Arenicola) led to a decrease of about 20% in meiofaunal density, as shown by exclusion experiments (Reise 1987, 2002). Physical disturbance (experimental raking) had a clearly negative effect on intertidal meiofauna (Austen and Widdicombe 2006). A study performed by Warwick et al. (1990b) enabled a discrimination of mechanical disturbance that reduced the stability of the sediment from pollution stress. The impact of shore crabs (Carcinus) on nematodes seems to be sediment-related (Schratzberger and Warwick 1999b): disturbance dominated in muddy bottoms, while predation prevailed in sands. Usually, disturbance-induced losses of meiofauna in tidal flats soon recover (see Sect. 9.1 on recolonization; Sherman and Coull 1980; Ólafsson and Moore 1990; Warwick et al. 1990b; Hall et al. 1991). The recovery potential of meiofauna in the deep sea is unknown, but disturbance-induced negative impacts of large epifauna on meiofauna have been documented (Thistle et al. 2008). (b) Sediment ingestion, the second impact type, occurs wherever deposit-feeding macrobenthos prevails: the feeding galleries of Arenicola marina contain lower numbers of nematodes (Jensen 1987a; Reise 1987); a negative impact of intensive digging was also noted for the priapulid Halicryptus (Aarnio et al. 1998). However, the reduction apparently depends on the intensity of the reworking and was not noted for other sediment feeders. The negative effects of disturbance by sediment ingestion can gradually merge into predator–prey relationships, particularly when meiofauna and macrofauna interact (meiofauna vs. burrowing polychaetes, crustaceans or fish, see Sect. 9.4.2). Disturbance (and predation) will mainly affect the meiofauna at the sediment surface and the upper sediment layers (Bell 1980). Some groups may react by performing
2.2 Biotic Habitat Factors: A Connected Complex
55
downward migration, which creates surface layers with reduced meiofaunal abundance. Effects of direct predation on meiofauna by macrofauna will be considered elsewhere (see Sect. 9.4.2). (c) Alterations in the chemical milieu, the third type of impact of macrobenthos on meiobenthos, affect meiofauna in mussel and oyster beds where thick layers of organic debris, fecal pellets and decaying algal mats accumulate. Here, oxygen soon limits the meiofauna in the layers beneath (Dittmann 1987; Dinet et al. 1990; Neira and Rackemann 1996; Reise 2002). Brominated compounds excreted by some echiurids and enteropneusts into their tube-wall linings have been suggested to exert a toxic impact on bacteria and meiobenthos (King 1986; Jensen et al. 1992b). In deep-sea bottoms, mangenese diagenesis may be affected by meiofaunal oxygen consumption (Shirayama and Swinbanks 1986). Among meiobenthic animals, the competition (mostly) caused by identical nutritional resources can structure the ecosystem. Trophic competition can cause spatial niche segregation and can ultimately lead to amensalism or mutual exclusion (e.g., Fenchel 1968a for ciliates; Joint et al. 1982 for nematodes). This exclusion has been documented as a within-group effect among meiobenthic species (nematodes, Ott 1972a; Alongi and Tietjen 1980, Santos et al. 2008a; harpacticoids, Chandler and Fleeger 1987). However, competitive exclusion has even been described between taxonomically distant meiofaunal groups (foraminiferans vs. harpacticoids, Chandler 1989; oligochaetes vs. turbellarians, Dörjes 1968; capitellid polychaetes vs. nematodes, Alongi and Tenore 1985; ciliates vs. nematodes, Bergtold et al. 2005). Competition can also result in a shift in life history characteristics (Heip 1980a, Marcotte 1983). A good example is the mutual exclusion of two species of the ciliate genus Condylostoma, C. arenarium and C. remanei, which have contrasting population dynamics, with maximum numbers occurring in June and November, respectively (Hartwig 1973b). Many negative interactions among certain meiofaunal taxa are difficult to analyze. Is the mutual exclusion of enchytraeid oligochaetes and turbellarians in the upper sandy beach, reported by Dörjes (1968), a result of trophic competition or merely an unknown, animal-mediated factor? Why is there a negative interrelation between the naidid Amphichaeta sannio and the nematode Tobrilus in the freshwater flats of the River Elbe (Schmidt 1989)? What causes the negative correlation between ciliates and freshwater nematodes (Bergtold et al. 2005), or that between the two bacterial symbiotic nematodes Catanema and Astomonema in mangrove muds (Bezerra et al. 2007). These questions remain largely unanswered. The causative factors of the inverse relationship between the harpacticoid Tisbe furcata and nematodes (Warwick 1987) and the contrasting population fluctuations of the gastrotrich Turbanella hyalina and its annelid counterpart Protodrilus symbioticus (Boaden and Erwin 1971) are also unknown. Similarly, the negative interaction between the foraminiferan Ammonia beccari and the harpacticoid Amphiascoides limicola in muds from tidal flats (Chandler 1989) has not been conclusively explained. Of particular relevance and widespread occurrence is the mutual regulation between permanent and temporary meiofauna (Elmgren 1978; Warwick 1989), which is in fact a series of effects ranging from predation to sediment reworking.
56
2 The Biotope: Factors and Study Methods
Permanent meiofauna, through their more effective and intensive feeding activities, initially attenuate the population development of newly settled polychaetes, bivalves, and very young amphipods. Subsequently, the growing macrofaunal species, through their intensive sediment reworking and their direct predatory impact, have an aggravated negative influence on permanent meiofauna (Bell and Coull 1980; Watzin 1986). The extent to which large beds of suspension-feeding bivalves reduce suspended meiofaunal populations in tidal flats needs to be experimentally verified. First studies do not suggest a density reduction (Boeckner, Recife: O 10). Examples of positive mutual effects between species arose from experiments by Reise (1983), where the bioturbation of lugworms enhanced solute fluxes and acted positively on the development of bacteria and meiofauna. Facilitative effects among nematode species were also reported by Santos et al. (2008a,b). Other effects are of a more indirect nature, yet their impact on the structure of the habitat, increasing habitat heterogeneity, can be considerable. Levin et al. (1997) considered the rapid subduction of plant material into the sediment by nonselective macrobenthic deposit feeders (some polychaetes or sipunculids) to be a “keystone resource modification” which massively influences the structure of the bottom and the fate of settling organic matter (see also Sect. 2.2.1 and 9.4). This example shows that besides the animals themselves, animal-produced structures can also positively influence meiofauna (Murray et al. 2002; Ólafsson 2003). Tubes and burrows and mounds piled up by benthic animals can have a massive impact on the habitat conditions of meiofauna. Around the tubes of macrobenthic polychaetes, nematode numbers were up to five times higher than in the unstructured controls. The small mounds around the openings created by many crustaceans in soft sediments have been documented to massively influence the hydrodynamics of the water on a microscale and thus to modify the influx of oxic surface water and the drainage of pore water (Hüttel et al. 1996; Ziebis et al. 1996). Nematode numbers were doubled along the tubes of thalassinid decapods in sublittoral Mediterranean sand, and foraminiferans increased as much as a hundredfold in number (Koller et al. 2006). In the deep sea, where animal-created structures tend to persist for long periods, distinct habitat patches of polychaetes fostered the abundance of harpacticoid copepods (Thistle et al. 1993). Intensive promotion of meiobenthos colonization by macrofaunal burrows and tubes was found to be responsible (among other factors) for meiofaunal spatial patchiness. This close interdependence has been simulated by mathematical models that show a good agreement between the model and sample situations (Pfeifer et al. 1996). Such modeling needs to be used more commonly in meiofaunal research. One important stimulus is the enhanced bacterial growth resulting from tube flushing effects, which profoundly influence the porewater circulation and the geochemistry of the bottom (Alkemade et al. 1992; Webster 1992; De Beer et al. 2005a; Weber et al. 2007). Burrows of decapod crustaceans, especially those of thalassinids, can be termed “hot spots” for bacteria and meiofauna (Förster and Graf 1992 in the North Sea; Koller et al. 2006 in the Mediterranean; Dittmann 1996 in Australia). However, meiofaunal populations also more than doubled the transport
2.2 Biotic Habitat Factors: A Connected Complex
57
rates of solutes (Aller and Aller 1992; Rysgaard et al. 2000), an increase that is particularly important for the transport of oxygen in the deeper layers (Zühlke et al. 1998). In freshwaters, the bottom is often densely covered by protruding tubes and mounds of tubificid oligochaetes, which exert a corresponding positive role on the geochemistry at the sediment/water interface. Another stimulus with a positive effect on meiofaunal populations is the secretion of exopolymers and the enrichment with fine organic particles. This will solidify the texture and enhance organic content and bacterial stocks (Eckman et al. 1981, Dobbs and Guckert 1988; Meadows and Tait 1989; Meadows et al. 1990; Decho 1990; Nehring et al. 1990; Jensen 1996). Macrofaunal tubes can also provide protection for meiofauna from predators and thus have a positive effect (Bell and Coen 1982). As a result of biogenic structures in/at the bottom, meiofaunal species richness increased (Ólafsson 2003). A network of more or less permanent tubes can also be produced by some meiofauna, preferably in silty sediments, but also in mucous biofilms (harpacticoids, Chandler and Fleeger 1984; Williams-Howze and Fleeger 1987; nematodes, Cullen 1973; Riemann and Schrage 1978; Platt and Warwick 1980; Nehring et al. 1990; Jensen 1996; Fenchel 1996a; Mathieu et al. 2007). As a result of biogenic structures in/at the bottom meiofaunal species richness increased (Ólafsson 2003). Considering this array of factors that are beneficial to meiofauna, it is understandable that the positive effects of biogenic structures have been suggested to yield 10–50% of the meiofaunal colonization of tidal flats (Fig. 2.14), with even higher figures for harpacticoids and gnathostomulids (Reise 1981a). In exclusion experiments, the presence of lugworms was estimated to promote the abundance of >90% of meiofauna (Reise 1983). In recent years, various large-scale, man-made physical disturbances have affected the biotopes of meiofauna with increasing intensity and frequency, including maintenance dredging and habitat (beach) enhancement. Climatic extremes, intense fishing with bottom trawls and dikes and other constructions along the shores and waterways have escalated erosive forces in the last few years. Maintenance dredging and habitat (beach) enhancement have become regular counteractions by which huge masses of unconsolidated sediment are mechanically distorted. Amazingly, the meiofauna appear to recover much more quickly from these maximal disturbances than macrofauna, as micro/mesocosm experiments and large-scale field surveys have shown (Schratzberger et al. 2006; Bolam et al. 2006). Supported by meticulous statistical analyses, they disclosed that the impact of this mechanical sediment distortion was mitigated by the concomitant intensive transport of meiofauna with slush water and intensive migration activities. The meiofaunal species richness recovered from an initial reduction after relatively short periods (weeks), while density reductions remained for longer periods (>1 year) than in the reference areas. In sandy habitats recovery was quicker than in mud, and nematodes were less affected than harpacticoids. The recolonization processes of meiofauna in the freshly consolidated sediments must be differentiated. They not only follow passive settling, but also depend on the water transport capabilities of the taxa and their reproductive potentials (see Sect. 7.3).
unstr.
head
tail
0
10
Arenicola
cm
20 0 20
Nematoda x cm−3
unstr.
cm
0
5
10 20 0 20
Pygospio
Corophium
Fig. 2.14 Impact of biogenic structures on meiofaunal density. Nematodes around burrows of various tidal flat macrofauna in comparison with unstructured (unstr.) reference sites. (After Reise 1981a)
Box 2.7 Animals as Habitat Factors Direct interactions between meiobenthic animal groups often result in competition, niche partitioning or even mutual exclusion. Structuring effects through food competition are hard to delineate from predator–prey relations, especially when dealing with macrofauna–meiofauna interactions, but have even been documented between unrelated meiofaunal groups. Disturbance (the other animal-mediated structuring source), on the other hand, is due not only to predation but also to bioturbative effects. Mechanical disturbance of meiobenthos is mostly the result of sediment reworking by digging or bulldozing macrofauna, and has mostly adverse effects on meiofauna. However, only extreme and long-lasting disturbing effects tend to degrade the meiofauna for a long time. Short disturbance events are quickly compensated for, making meiobenthos fairly resistant to environmental stress. Burrows of benthos usually have a benign effect on meiobenthic populations; they diversify the habitat structure and ameliorate the mechanical and nutritional properties of the sediment by compaction. This positive structuring is enhanced by the greater influx rates of oxic water and attractive mucus secretions. Beyond that, protruding macrofaunal tubes provide shelter and foster the settling of organic particles, modulating the hydrodynamics at the surface.
2.3 Conclusion: The Microtexture of Natural Sediments
2.3
59
Conclusion: The Microtexture of Natural Sediments
Natural sediments are characterized by a net of intricately interacting abiotic and biotic (biogenic) factors rather than just the grain size, porosity or sorting of the particles. The interactive and multiple nature of numerous determinants in the field, illustrated in Fig. 2.15 and schematically in Fig. 2.16 was nicely demonstrated in a small study performed by Warwick et al. (1986b), who investigated the impact of the macrobenthic, tube-dwelling polychaete Streblosoma on the meiofaunal assemblage. Around the tubes that extend slightly above the surface there is an area with rich meiofauna, probably because of the improved flux conditions caused by the tube and also because of the worm’s mucus secretion and concomitant microbial activities. Slightly further away, in the grazing range of the polychaete, the meiofauna was impoverished through mechanical disturbance and perhaps uptake by Streblosoma. Mainly through the activation of geochemical fluxes and microbial activity, an inhomogeneous small-scale topography is created (Fig. 2.15) which supports the aggregation and a patchy distribution of meiofauna, even in superficially uniform sediments (Sun and Fleeger 1991). The mucus film secreted by bacteria, phytobenthos and many burrowing animals like nematodes and annelids decreases the amount of erosion and meiofaunal suspension from currents. This entire web of biotic ecofactors which influence the occurrence of meiofauna is intricately combined with and influenced by the multitude of abiotic parameters described in Sect. 2.1. Any schematic attempt to illustrate the complexity of the “meiobenthic habitat” (Fig. 2.16) is too static to adequately reflect the dynamic interactions of the components. In addition to the more regular factorial system, depicted in Fig. 2.16, the meiobenthic ecosystem is, of course, subject to stochastic or “accidental” factors, such as local irregular and temporary disturbances (e.g., storms, pollution events) and benefits (e.g., food input through the settling of larval forms or decaying macrofauna). These erratic alterations, even those of a small-scale nature, may influence the system unpredictably. They certainly contribute to the notoriously patchy distribution pattern of meiofauna and support their high diversity, two characteristics of meiobenthic communities that make generalization very difficult. A high diversity is also maintained by the well-developed nutritional selectivity of meiofauna, which seems to exceed that of macrofauna. The resulting differentiated resource partitioning of the available food stock renders biotic (trophic) factors more relevant than physiographic parameters: the occurrence and distribution of meiofauna appear to be controlled by a multifactorial dynamic network in which the biotic factors in particular must be considered. More detailed reading: Round (1971); Coull (1973, 1986), Coull and Bell (1979); Rhoads et al. (1977), Eckmann (1985), Warwick (1989), Decho (1990); Watling (1988, 1991); Reichelt (1991); Krumbein et al. (1995); Meyer-Reil (1994); Snelgrove and Butman (1994); Fenchel (1996a); Reise (2002); Murray et al. (2002); Ólafsson (2003); Meysman et al. (2006a).
animal tubes
faecal pellets
animal burrows
geochemical fluxes
increased stability
oxic
decaying organisms
sulfidic, anoxic
disturbance
mud, silt
Fig. 2.15 An illustration of the biotic factors structuring the occurrence of meiofauna in a tidal flat sediment. (Compiled from Meadows 1986; Anderson and Meadows 1978, and other authors)
sand
sediment water interface
wrack
erosion
60 2 The Biotope: Factors and Study Methods
2.3 Conclusion: The Microtexture of Natural Sediments
61
sublittoral: water currents, wave action
sedimentary complex
physicochemical complex
grain size composition
H 2S
eulittoral: atmospheric exposure of sites, tides
permeability
porosity
O2
water flow
To
pH
S‰
water supply (eulittoral)
H2O content
Meiofauna structure and distribution
dissolved organic matter
biogenic structures biogenic complex
mucus production
biofilms
biotic complex
bacteria
food
bioturbation predation
particulate organic matter, detritus
phytobenthos
disturbance
macro-zoobenthos
Fig. 2.16 A schematic factorial web structuring the habitat of meiobenthos
62
2 The Biotope: Factors and Study Methods
Box 2.8 Sediments: The Real Picture The natural sedimentary fabric is an interacting, living system of organic components, mucous excretions, detrital agglutinations and microbial films interwoven with the inorganic particles. Thus, the real habitat of meiobenthos cannot be defined by grain size and sorting or characterized by sieving. Particulate organic matter (POM) and debris represent major components that are inseparably linked to bacterial coatings and biofilms produced by bacteria, diatoms and fauna. When oxygen is present, this organic complex is highly attractive to many meiofauna and is often the basis for their patchy microdistribution. Dissolved organic matter (DOM), which consists mainly of carbohydrates and amino acids and is released by many microorganisms, is mostly bound to colloidal complexes and mucous layers. It represents valuable nutrition for meiofauna. All of these components influence geochemical fluxes, microbial activities, the degree of biological reworking (bioturbation), and fauna-mediated bioirrigation and pelletization (Watling 1988, 1991; Meysman et al. 2006b). The huge surfaces of this nutrient-rich sedimentary web are ideal substrates for rich microbial colonization. Bacterial stocks that are orders of magnitude richer in the sediment than in the open water represent a dominant nutritive source for meiobenthos. The metabolic waste products of the meiofauna, in turn, attract bacteria. The spatially and temporally variable interactions between fauna, plants, bacteria and detritus become further complicated by the release of substances which exert specific stimulatory or inhibitory effects on all components of the system (Tenore and Rice 1980, Meadows 1986). Meiofaunal colonization is further controlled by biogenic microstructures such as animal burrows and tubes, mucus tracks, and fecal pellets. These are not only the nutritional basis for many meiofauna, but also act as mechanically important biostabilizers, reducing erosion. The complexity of this network renders the measurement of single parameters difficult. Recent advances in simulating and measuring have been accomplished using microelectrodes and on-line experiments. These innovative approaches will contribute to understanding the dynamics of the living sediment.
Chapter 3
Sampling and Processing Meiofauna
Since methodology is the domain of Introduction to the Study of Meiofauna (Higgins and Thiel 1988) and the updated version of Methods for the Study of Marine Benthos (Eleftheriou and McIntyre, 2005; herein for meiobenthos: Somerfield et al. 2005), this chapter should be read in conjunction with those treatises. In order to avoid too much overlap, only the more important sampling methods will be presented here. In some cases, I will include supplementary hints of practical importance which are rarely mentioned elsewhere. Additional information, particularly more detail on sampling design and devices for sampling soft-bottom sediments, may be found in the methodological recommendations of Elmgren and Radziejewska (1989) and the compilation by Blomqvist (1991). Since sampling design, strategies and statistical evaluation vary with the scientific problem being addressed, it is impossible to describe universally applicable methods here. For freshwater, valuable sampling hints are given in Palmer et al. (2006). In addition to the usual textbooks for (biological) statistics, the critical review on the calculation of diversity by Hurlbert (1984) and the compilation by Underwood and Chapman (2005) are recommended. The problem of control sites, which is important when evaluating disturbances, is competently covered for meiofauna in Eskin and Coull (1984).
3.1
Sampling
Prior to each investigation, it is important to carefully consider the area to be sampled and the equipment to be used. Among other factors, this depends greatly on sediment characteristics, the animal taxa in question and their specific adaptations. The abundance of animals, their turnover and the rate of fluctuations will influence the appropriate sample size and sampling strategy.
3.1.1
Number of Replicates and Size of Sampling Units
Patchiness, a notorious and fundamental characteristic of meiofaunal distribution (Sun and Fleeger 1991), requires that parallel samples should be taken in order to achieve O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
63
64
3 Sampling and Processing Meiofauna
reliable quantitative meiofauna data. In practical work, it is not the mathematically unrealistic optimal number of samples but the cost-effectiveness (time and manpower) that has to be considered (Esteves et al. 1997). As a general rule, variation between replicates should be less than the variation between sampling stations. Instead of taking one large meiofaunal sample, it is more advisable to take several small subsamples and evaluate them separately (for details see Underwood and Chapman 2005). For a reliable assessment of mean temporal abundance, the strong temporal fluctuations in many meiofaunal groups require repeated samplings over short time intervals (e.g., weekly, fortnightly; see Armonies 1990). The classical procedure used to ascertain the optimal sample size (which kinds of samples require the least effort to evaluate and still yield reliable results?) is to initially count a larger unit and then compare it with data from subsamples of defined, smaller units. The counted data should not deviate by >10% from the expected ones. The calculation of species versus effort curves is a helpful means to avoid unnecessary input (Smith et al. 1985). However, because of the extreme heterogeneity in meiofaunal distribution patterns, it remains questionable whether this method is always applicable. There is a rule that the surface sampled by the corer should exceed the patch size, but in order to follow it the latter must be known. The relevance of this rule is illustrated in Fig. 3.1. A large number of small cores (0.5–1.0 cm2) are required to assess the heterogeneous microscale distribution pattern of meiofauna (Findlay 1982). Valuable guidelines for the application of the right sample size and the resulting test power are given by Sheppard (1999) and Underwood and Chapman (2005). Sample size and, accordingly, sampling gear have to be adapted to the wide size range of meiofaunal groups. If just one size class is considered (e.g., small but numerous ciliates or fairly large but infrequent oligochaetes), sampling can be optimized by using specialized corers and evaluating different sample volumes via different extraction methods. For studies on the entire assemblage of meiofauna within one sample site, either samplers of various sizes must be used in parallel, or one must compromise when selecting the sampling gear. In any case, presampling and, where possible, examining live meiofauna gives valuable information about the qualitative composition (of those taxa destroyed by fixation too) and local distribution of the meiofauna.
3.1.2
Sampling Devices
A wide variety of sampling devices have been designed for the collection of the various meiofaunal taxa, which differ in mobility, size and habitat. For detailed lists of sampling methods, sampling gear, suitability and efficiency, see Tables 5.1 and 5.2 in Eleftheriou and Moore (2005). Specifically for meiobenthos, Somerfield et al. (2005) compile collection methods and extraction techniques in their Table 6.1. Wells (1971) recommended 10 cm2 (= 3.6 cm ø) as the minimal sample area. This small version of a tube is applicable mostly for ciliates and other abundant small
3.1 Sampling
65 403.9-527.5 527.5-650.7 650.7-774.3 774.3-897.5 877.5-1021.1
2 cm
a
b
c
411.8-550.7 550.7-689.6 689.6-828.6 828.6-967.5 967.5-1106.4
382.1-532.1 532.1-682.5 682.5-832.5 832.5-982.8 982.8-1132.8
d
e
Fig. 3.1a–e Assessment of the meiofaunal distribution on an area of about 15 × 15 cm (0.2 m2) and its dependence on the sampling design. a “Actual” distribution of meiofauna on the sampling square, assessed by 81 cores. b, c Different sampling strategies resulting in different distribution patterns (d, e). (After Findlay 1982)
66
3 Sampling and Processing Meiofauna
meiofauna (nematodes). It is often constructed from disposable syringes whose lower ends have been cut. For analyses of heterogeneous small-scale patterns (e.g., in ripple marks), even straws with surface areas of <1 cm2 are recommended. Sampling problems associated with all tube or piston corers include the shortening of the core by sediment compaction, and, under water, the disturbance effects of shock waves, which tend to flush away the important sediment surface (see Jensen 1983). These shortcomings of sampling are particularly serious in soft, flocculent bottoms and can be minimized only by using corers with fairly wide diameters (8 cm or more, McIntyre and Warwick 1984). However, for most purposes, this yields an undesirably large quantity of sediment. Rutledge and Fleeger (1988) tested corers of between 2.6 and 10.5 cm diameter and obtained no major errors in the results when compared with controls provided that the corer is gently and slowly pushed into the sediment. In shallow water this can be done best by SCUBA divers. Holopainen and Sarvala (1975) found that cores of 4.4 cm diameter, obtained by hand in this manner, contained 20% more fauna than cores of the same size obtained mechanically. This reduced efficiency caused by mechanical sampling was not compensated for until the corer diameter was increased to 8.3 cm! In a detailed investigation on the efficiency of core sampling, Blomqvist (1985, 1991) identified as prime sources of sampling error effects such as the shortening of the core, sediment resistance, lateral extrusion (bow wave effect), and repeated penetration of the corer due to wave action. Multiple corers such as the Craib corer or the Kajak corer (see below) as well as the large “Aberdeen corer” are superior to the bulky box corers in terms of reliability of capture since their bow waves are less destructive (Bett et al. 1994; Shirayama and Fukushi 1995). Since they provide effective replication of samples, they are also more efficient in yielding representative community figures. However, there is one caveat: replicates obtained by multiple corers do not represent statistically independent and random samples, but rather nested two-stage pseudoreplicates (see Hurlbert 1984). They are not equivalent to repeated samples obtained with a single corer. The sampling efficiencies of grabs, whether the “Petersen grab,” the “Ekman grab,” the popular “van Veen grab,” or the “Smith–McIntyre grab,” are often not reliable for quantitative approaches (Ankar 1977; Blomqvist 1985, 1990), and so results from grab samples should be used with caution. Although their construction can be improved (Håkanson 1973; Blomqvist 1990), even enhanced grabs do not penetrate deeper than 5 cm at their edges into the sediment (Riddle 1989), and the bite profiles are often not quantifiable (Eleftheriou and Moore 2005). Somerfield and Clarke (1997) pointed out that the variability of the results also depends on weather conditions, sediment type, size of samples, etc. Discussing the efficiencies of different corers, Long and Wang (1994) recommend that rather than the absolute figures of animals retrieved, the ratios between the means and standard deviations of the respective replicates should be compared. The application of this “signal/noise” ratio would reduce the impact of variances in sample number. Depending on the sampling locality, the research objectives and the prevailing animal group(s), meiofaunal studies are based on a variety of specially constructed
3.1 Sampling
67
sampling equipment. Only the most commonly used and generally available sampling gear will be described and commented upon here. For a review of benthic sampling gear, see Blomqvist (1991) and Somerfield et al. (2005). A valuable overview, especially of methods used in freshwater studies, is given in the review by Palmer et al. (2006) and Strayer (1996).
3.1.2.1 Qualitative Sampling - Straining groundwater. The simplest and most extensively used (“classical”) method since Remane’s studies of eulittoral sandy shores in the 1920s is to dig a funnel-shaped excavation into the sediment down to the groundwater depth, allow water to fill the bottom of the pit, and then strain the inflowing water rapidly with a small hand net (mesh size 45 or 63 µm) with swift, circulating movements. Meiofauna from the surrounding sand layers are washed into the pit and collected by the net. Repeating this procedure several times will result in a qualitative, albeit not for all taxa representative overview of the meiobenthic inhabitants from a relatively wide sampling area. A semiquantitative modification of this method is commonly used in groundwater meiobenthology: a defined water volume is collected after stirring and then poured through a 45-µm net, and the net content is fixed for later evaluation. This Karaman–Chappuis method is appropriate for quick assessments of the meiofauna when exact extractions are not possible or necessary. In comparison with vertical coring, this filtering of seepage water yields no difference in (nematode) species diversity, but species composition may differ (Gourbault and Warwick 1994). - Dredging. From sublittoral grounds, a fine-meshed dredge (e.g., Ockelmann 1964; Higgins 1964), sometimes combined with a pushing plate (Muus 1964) is used to scoop the uppermost cm of the sediment. The Muus sampler is sometimes considered to be semiquantitative and may even catch some of the more mobile meiofauna and small macrofauna (e.g., isopods). - Suction corer. Only small, hand-held suction and piston corers are sufficiently gentle gadgets to collect meiofauna from soft bottoms. They allow for quantitative sampling and even vertical profiling of the core (Taylor et al. 1995; Tita et al. 2000a). In contrast, large “air lifts” or “slurp guns,” which evacuate the bottom using a connected pressure tank, damage most meiofauna.
3.1.2.2
Quantitative Sampling
For meiofauna, tube corers are the most versatile instruments and are superior to most box corers and bucket grabs. The classical meiofaunal corer is a perspex tube with an externally beveled lower edge and an internal surface of between 10 and 20 cm2 (= 3.6–5.0 cm diameter, for size restrictions see Sect. 3.1.1). Holes can be drilled into the perspex wall to allow the insertion of thin electrodes for measuring abiotic parameters at various depths (Riedl and Ott 1970). A rubber stopper con-
68
3 Sampling and Processing Meiofauna
nected with a hose helps to reduce friction when inserting the corer by lowering the pressure above the core (Fig. 3.2a), and also facilitates the sliding of the sediment column out of the tube for fractionated subsampling. In directly accessible (eulittoral) muddy shores, the open tube can be pushed into the bottom, and after its upper end is closed the core can be removed with ease. In sandy areas with potentially more compact sediment, insertion of the corer is more difficult (greater resistance), requiring the application of greater force (such as that
a
b
Fig. 3.2a–b Quantitative meiofaunal corers. a Perspex corer with rubber stopper and hose (after Riedl and Ott 1970). b Square steel corer for sampling deep sand layers. Left: cross-section. (After McLachlan et al. 1979)
3.1 Sampling
69
obtained with a rubber hammer). In unconsolidated sediments, a closing device is required for the precise, quantitative removal of the core (see the piston stopper in Blomqvist and Abrahamsson 1985), unless the tube can be dug out carefully and closed from below by hand. One device which reduces friction in sandy bottoms and allows deep sampling consists of a square stainless steel tube with one side fitted for an interlocking sliding blade (Fig. 3.2b). Extrusion of the core is accomplished by using a piston (see below) or air pressure. Subsampling is usually done by fractioning the reversed core with a broad spatula or a thin blade (beryllium bronze is considered best; the blades of normal knives are inconveniently narrow). A particularly useful fractioning device has been described by Danielopol and Niederreiter (1990). Another construction for serial sectioning (Blomqvist and Abrahamsson 1987) is probably more useful for muddy than for sandy substrate. In corers opened from one side (Fig. 3.2b), tightly fitting blades can slice the subsamples horizontally. In any case, subsampling of the core should be done immediately after the retrieval of the sample in order to avoid bias due to faunal migration. For studies on bigger or rare meiofaunal animals (e.g., Isopoda, Gastropoda), larger corers with diameters of up to 20 cm can be obtained. When sampling intertidal oligochaetes, the author used a square 10 cm × 10 cm metal box corer with sharp edges at its lower end and a removable side wall. In sandy sediments, the appropriate length of the corer (usually between 20 and 30 cm) is determined by the depth to which it can be driven into the bottom without major disturbances, compression or otherwise altering the depth distribution of meiofauna. Core sampling in exposed beaches where such taxa as nematodes and tardigrades occur in greater depths (up to 180 cm deep, Kristensen and Higgins 1984) must be incremental, e.g., short core samples must be taken repetitively at increasing horizon depths. The gadget of McLachlan et al. (1979) depicted in Fig. 3.2b is designed to allow sampling in sandy beaches down to a depth of 1 m. Exact vertical subdividing of the core is facilitated by ring-marks (e.g., every 2 or 5 cm) on the outer surface of the corer. If the study object requires a finer subdivision, i.e., examination of meiofauna–microbial interactions, the corers designed by Joint et al. (1982) with screw micrometers attached can serve as valuable tools. For distributional studies of microbiota in intertidal sediments, Wiltshire et al. (1997) applied a shock-freezing technique (liquid nitrogen), which enabled the sediment to be sliced into fractions of <1 mm. The sampling and subdividing of submersed sediment cores is difficult because, once lifted above the surface, the core, or at least its lowest sediment fractions, tend to be washed away by the overlying water in the tube. Therefore, the overlying water covering the core plus the flocculent surface layer should be carefully siphoned off onto a fine sieve prior to subdivision of the sediment in order to avoid distortion by percolating water flow. In shallow water, tightly sealing the tube with a stopper immediately after its withdrawal from the sediment, while the corer is still under water, will secure the core. In order to quantitatively obtain any remaining centimeter fractions of the core, the use of a metal piston with a sturdy handle is recommended. By pressing this handle against the ground or onto another solid
70
3 Sampling and Processing Meiofauna
support, the sediment core can be gently pushed up to the required height and sliced away level to the upper edge of the core. In shallow sublittoral muddy or fine-sand bottoms, the pole corer of Frithsen et al. (1983) combines simple handling with reliable closing (a lid from above is triggered on contact with the sediment by a bottom plate). For coring in deeper sublittoral or deep-sea sediments, remotely operated corers are required. Since the often unconsolidated, fluffy surface layer is particularly rich in meiofauna, a tight closing device that retains the flocculent fraction is imperative. In order to achieve a reliable quantitative sample without producing a bow wave effect from the corer itself, the coring tube should be deployed in a hydraulically damped supporting device which, after touching the bottom, allows the tube to slide slowly and vertically into the sediment (Jahnke and Knight 1997). When connected to a pinger, the coring apparatus can electronically record the bottom surface and be gently lowered during the last meters of descent. Reliable gravity corers for sublittoral sampling include the “Kajak” or the “Haps corer” (Kanneworff and Nicolaisen 1973, 1983), along with several modifications of it (Jensen 1983; Blomqvist and Abrahamsson 1985; Chandler et al. 1988), some of which allow multiple sampling with three or four parallel tubes. In all versions the tube is tightly closed from above by a lid released when the corer comes into contact with the sediment (Fig. 3.3). Since the corers of the Kajak type lack a closing valve at the lower end, they function best in fine sediments. In the commonly used Craib corer (Craib 1965) a ball valve is used as the lower closing device. Valuable modifications of the Craib corer have been designed by Fenchel (1967) and Jackson (1986). For sampling freshwater substrates, Danielopol and Niederreiter (1990) combined the lid device of the Kajak corer with the ball valve of the Craib corer. The multiple coring frame of this sampler (six tubes) enhances its value for reliable quantitative work. When working at greater depths from a larger research vessel, the multiple SMBA corer, first developed by Barnett et al. (1984), has become a standard tool (Somerfield et al. 2005). Later modifications of this “multicorer” entail, besides a solid supporting frame, a set of elaborate techniques (hydraulically damped, controlled penetration of the tubes, an electronic release system, optional visual inspection by video camera surveillance) that provide undisturbed quantitative samples with replicate cores, even from great depths. They have been used successfully on board many research vessels for meiofaunal sampling (see Sect. 8.3). Most of the deep-water corers are equipped with weights and guidance wings for swift and vertical descent to the bottom. When both macro- and meiobenthos need to be studied simultaneously, e.g., in many ship-board situations, the collection of meiobenthos is most often accomplished by subcoring the sediment taken with larger box corers (e.g., the USNEL corer), which are needed to obtain the macrobenthos (Coull et al. 1977; Shirayama 1984; see Eleftheriou and Moore 2005). This allows unimpeded water flow during descent and a lowering control is used that avoids the negative bow wave effect (but see the cautionary results of Bett et al. 1994; Shirayama and Fukushi 1995 above).
3.1 Sampling
71
b
a Fig. 3.3a–b A modified Kajak corer. a Overall aspect. b Schematic detail of the closing mechanism (after Hakala 1971)
When the hoist line becomes taut, the corer is tightly closed with a “closing arm” that is pulled below the square sampling box from the side. More detailed reading: Hurlbert 1984; Riddle 1989; Blomqvist 1985, 1991; Andrew and Mapstone (1987); Burd et al. 1990; Somerfield and Clarke (1997); Eleftheriou and McIntyre (2005); Underwood and Chapman (2005).
72
3.2 3.2.1
3 Sampling and Processing Meiofauna
Processing of Meiofaunal Samples Preserving Meiofauna in Their Natural Void System
The actual positions of meiofauna in their microhabitat can only be preserved and demonstrated by applying rather complicated methods immediately after sampling which prevent the animals from changing their position. Shock freezing of the core (even in liquid nitrogen) is not advisable since it has been proven to cause considerable distortions of the natural stratification (Rutledge and Fleeger 1988). Moreover, the assessment of stable isotope values, today often used for energy flux analyses, is biased by bulk freezing (Dannheim et al. 2007). Microwave irradiation of the undisturbed (sub)sample has been tested and was found to give moderately good results (Berg and Adams 1984); however, it requires subsequent fixation (e.g., by hot formaldehyde vapors). Haarløv and Weis-Fogh (1953) found that infiltrating a hot agar solution into a soil sample will, after cooling and hardening, leave the soil fauna undisturbed in their original positions. Furthermore, the step-wise dehydration of the core by ethanol enables serial sections to be obtained with a razor blade. Slices 750 µm thick obtained from a sandy soil sample have been stained and then finally mounted in glycerol for inspection or photography. Another method used in sedimentology involves embedding the core with resin (Williams 1971, Rieger and Ruppert 1978; Watling 1988). The sediment core is drained under negative pressure (exsiccator!) with a water-soluble resin to which an appropriate stain is added to make the animals in the sediment more recognizable (see Sect. 3.2.2). After hardening, the block must be prepared for microscopical inspection by trimming, sectioning and polishing. Such permanent sections reveal the natural structure of the pores with the animals included. For foraminiferans, Bernhard and Browser (1996) developed a fixation method with fluorescent dyes that enabled the subsequent assessment of their natural positions in cores under a fluorescence microscope. It remains to be seen whether this is also applicable to other meiofaunal groups. When examined under a microscope on a glass slide, all movements of interstitial animals seem clumsy and ineffective, in contrast with their swift and efficient behavior in their natural microhabitats. Only through the microscopic inspection of active meiobenthic animals are their intricate anatomical and functional adaptations revealed. In order to understand the functions of structures and their ecological niches, “microecological” observations in appropriate microscopic observation chambers (often special microscopic slides) are needed (see Marcotte 1984; Gilmour 1989; Shirayama et al. 1993). Direct observations of meiobenthic animals in their undisturbed habitats are usually prevented by the optical inaccessibility of sediment samples. A microvideocamera for insertion into (coarse) sand and gravel, as developed by Danielopol and Niederreiter (1990), might be a promising tool for direct inspection. One optical obstacle to such microscopic applications is the uneven surface of any natural sediment surface. There have been several attempts to mimic the sedimentary structure of the meiobenthic habitat for analysis. One method involves an artificial sand system cast in
3.2 Processing of Meiofaunal Samples
73
transparent resin (Giere and Welberts 1985). The near-natural structure is achieved by photographing the genuine sand and subsequently producing a plastic cliché by etching. Another cast technique has been used to analyze the void structure of sea ice (Weissenberger et al. 1992), and might also be an interesting approach for casts of the pore system in sand.
3.2.2
Extraction of Meiofauna
3.2.2.1 Sample Staining Prior to fauna extraction it is often advisable to selectively stain the whole sample for easier recognition and discrimination between dead shells and freshly dead animals (e.g., in foraminiferans). For live fauna in unfixed samples, vital classical stains like Neutral Red or Carmine can be used. Fixed samples are typically bulk-stained with Rose Bengal, added to alkaline formalin (for details of concentrations, see Pfannkuche and Thiel 1988). Alkalinity (via sodium hydroxide) avoids the precipitation of Rose Bengal in aequeous formaldehyde (Riemann 1989). The stain is absorbed differently by the various meiofaunal groups. After 10 min, annelids and turbellarians stain dark red while nematodes may not even become pink; kinorhynchs may require >48 h because of their thick exoskeletons. As a general rule, staining for 1 h is sufficient for most meiofauna. Overstaining is sometimes difficult to remove (by acid ethanol) and impairs structural analysis under the microscope, but repeated rinsing often yields perfect and durable staining. The bright fluorescence of Rhodamine B is another convenient method of facilitating sorting (Hamilton 1969). Counterstaining is often helpful for differentiating between dead cells (detritus) and live animal tissue. Rose Bengal can be counterstained with Chlorazole Black or Phloxine B (Williams and Williams 1974; Mason and Yevich 1967). According to Haarløv and Weis-Fogh (1953), living or freshly dead cells are selectively stained by Violamin, which contrasts them against the detritus and sediment particles. The different fluorescence of living (active) and dead cells has been exploited by a method suggested by Williamson and Palframan (1989) originally for microbiology, but is probably also applicable for meiobenthology. Differentiation between dead and living cells is particularly relevant when evaluating microbes and protists. Crystal Violet, Sudan Black B, Coomassie Blue G, and various fluorescent dyes (e.g., FITC, CTC) have been found to be useful for bacteria and Foraminifera (see Epstein and Rossel 1995b; Bernhard 2000) and should also be tested for efficiency with metazoans.
3.2.2.2 Extraction Meiofauna must always be extracted from the sample because the total volume of the fauna is exceedingly small compared with that of the sample. Which of the numerous
74
3 Sampling and Processing Meiofauna
extraction methods should be applied depends on the nature of the sediment and on the aim of the study (see the detailed list in Somerfield et al. 2005, Table 6.1).
Qualitative extraction A convenient method of concentrating meiofauna for qualitative inspection from a large sample is to leave a jar or bucket with the moist sediment for about a day under normal room temperature. This will force most meiofauna to aggregate near the surface (mainly due to oxygen deficiency). Enriched subsamples can then be taken from the surface layers for analysis. For large sand samples, the most convenient and widely applicable method for quickly obtaining abundant numbers of meiofauna remains decantation by hand using a 45-µm or 63-µm sieve and a bucket. Decantation is most reliable for medium-to-coarse sand and should not be used for finer sediments. It is applicable to fixed or unfixed samples, directly in the field or from original sand brought into the lab. In live samples, it will yield sufficiently representative results if preceded by anesthetization of the often adhesive meiofauna by an isotonic MgCl2 solution (7% = 75 g MgCl2· 6H2O for sea water of 35 PSU). Another anaesthetizing substance that relaxes the animals very gently is a saturated solution of Mephenesin® (3-o-toloxy-1,2-propandiol). One method for obtaining even adhesive meiofauna is to rinse the sample briefly in freshwater, decant into a sieve and immediately reintroduce the decanted material into seawater. The freshwater shock will cause the animals to detach from the substrate. Soft, fragile taxa may be destroyed but much of the meiofauna, especially the armored or shelled taxa (e.g., crustaceans, many nematodes, kinorhynchs), will remain in a suitable condition. Large numbers of debris-free, fully viable animals, as required for physiological or toxicological analyses or for experimental purposes, e.g., bioassays, can be conveniently obtained from various sediments by “mass extraction” methods. Derived and modified from soil ecology, they are based on the avoidance reactions of many meiofauna (especially nematodes) to light and/or desiccation or warming of the sediment sample from above. This forces the animals to migrate downward through the sediment and eventually through a bottom sieve/gauze/ply paper tissue into a waterfilled collecting jar (Couch 1988; Forster 1998; Mangubhai and Greenwood 2004; Rzeznik-Orignac et al. 2004). However, it should be noted that these methods are not representative: sensitive taxa deteriorate quickly, while “stationary taxa” remain where they are and do not migrate (e.g., oligochaetes, nemerteans). Special procedures for specified animal groups (e.g., the bubble-and-blot method) will be dealt with in context with the respective systematic overview (Chap. 5).
Quantitative extraction Decantation: Medium-to-coarse sands are decanted as described above. If seawater decantation is repeated twice, followed by freshwater extraction, and it is done in small portions, enumeration can be considered quantitative (Ankar and Elmgren
3.2 Processing of Meiofaunal Samples
75
1976), particularly after staining with Rose Bengal. This has been successfully applied even for “soft meiofauna” (Noldt and Wehrenberg 1984). “Difficult” muddy and clayey sediments can be gently sieved and the suspended fauna effectively decanted if the sample is sonicated for 10 s (Murrell and Fleeger 1989). Several repetitions of the procedure are recommended for improved efficiency. Addition of a drop of detergent has also been proven helpful (see below). Elutr(i)ation: Developed by Boisseau (1957), this method has been modified by McIntyre. His “Aberdeen elutriation apparatus” was shown to work both reliably and quickly (Uhlig et al. 1973). Several more complicated versions exist, but all are based on the same principle: a water jet is used to separate fauna and sediment particles by specific gravity. When elutriating samples with a high content of fine or detrital material, it is advisable to pour the water through two sieves mounted in series to minimize the risk of overflow due to mesh clogging and thus the loss of animals. In fresh samples, pretreatment with MgCl2 (7% solution) is recommended for quantitative evaluation in fixed samples stained with Rose Bengal. Thorough inspection of the sediment residue enables the necessary extraction time and number of runs to be estimated. Careful elutriation is usually more exact and convenient than decantation, although, for soft meiofauna, elutriation is a rather rough method and may cause damage. Elutriation is restricted to sandy samples, considering its working principle. The seawater ice method (Uhlig 1964), although applicable to a wide spectrum of sediments, is not quantitative for all meiofaunal taxa (Fig. 3.4a, b). As much as the method is suitable for soft micro- and meiofauna (e.g., flagellates, ciliates, turbellarians), it is not reliable for the quantitative extraction of nematodes, halacarids, tardigrades, annelids, and some harpacticoids. However, the most important feature is the restriction of the method to really fresh samples with active animals. Only animals from freshly taken samples will react adequately through avoidance migration. Frozen full-strength seawater may have initial thawing concentrations of up to 80 PSU and more, but this decreases during the perfusion, resulting in water that is oligohaline (below 5 PSU). This gradient causes a massive salinity shock to the fauna and the subsequent downward migration of the animals. The avoidance reaction of (many) meiofauna to the extreme temperature gradients caused by the melting seawater ice supplements the effect of this method. In its various modified forms, which essentially enhance the throughput of samples per unit time and facilitate the handling, the seawater ice method is simple enough to be used in any lab. Care must be taken to ensure that the thawing process continues long enough (about 3 h), and that the bottom gauze is kept exactly in flush contact with the seawater in the collecting dish (there must be no trapped air). A suction corer specifically designed for the subsequent use of the seawater ice method has been designed by Ruppert (1972); the lateral drill holes were adjusted for the insertion of special tubes that could later directly be used as thawing tubes. For quantitative work with fine sand and mud, the tedious hand sorting of samples has been largely replaced by flotation methods. The sample is suspended in a medium that has a specific gravity close to that of the fauna. Thus, the animals become neutrally buoyant, remain in suspension and are separated from the sediment by centrifugation.
76
3 Sampling and Processing Meiofauna
tube seawater ice clamp
cotton wool
sample rubber band gauze, 100-150 µm Petri dish with seawater
a
overflow dish
b Fig. 3.4a–b The seawater ice extraction method. a One cylinder with seawater ice. b A setup for series extraction
Introduced into soil biology in the 1950s, the traditional flotation medium is a sugar solution, whose general availability and (locally) low costs make it useful even today (Esteves and Da Silva 1998). However, this technique the use of sugar has largely been replaced by Ludox® (Du Pont), a colloidal silica sol with a specific gravity that is tailored to different specifications. When introducing Ludox into meiobenthology, De Jonge and Bouwman (1977) used it in the “TM” specification (specific gravity 1.39). Since the direct addition of Ludox to seawater causes the precipitation of the suspended silica, seawater samples should be rinsed in freshwater before treatment. Dilution to the desired specific gravity of 1.13 must be done with freshwater before adding this solution to the sample. Nichols (1979) found that using the AM version of Ludox avoided gel formation. For economic reasons, Furstenberg and Wet (1982) point out that Ludox HS-40%® (specific gravity 1.28) can be mixed with water with very good results. Also, Burgess (2001) recommends Ludox HS 40® as the best formula in terms of interaction with the specimens, time and extraction efficiency. One should also test the efficiencies of Ludox LS® or Ludox HS-30%® (dilutable in seawater), which are available in smaller containers.
3.2 Processing of Meiofaunal Samples
77
The extraction efficiency usually is >90%, hardly varying with meiofaunal group, but it is advisable to repeat the Ludox treatment several times. The number of repetitions needed for quantitative evaluation has to be assessed via test runs. If the sediment can only be treated once without any repetition, a correction factor to improve the reliability of the data should be used (Heip et al. 1974). Even in soft muds, two repetitions and thorough centrifugation usually yield 95% retrieval of fauna. With some practice, this can be done in about 30 min per sample. The size of each subsample in the centrifuge depends on the size of the centrifugation beaker, but for sufficient accuracy the addition of at least four (ten is better) times the amount of Ludox (or a Ludox–seawater mixture) to the sediment sample and thorough suspension by careful mixing is required in each case. It is advantageous to remove pebbles and other heavy particles prior to the addition of Ludox. In clay sediments which tend to form stable clumps, the addition of a water-softening detergent like Calgon® (Barnett 1980) accompanied by prolonged stirring has been shown to aid the extraction of fauna without excessive damage (add one part to five parts of sample volume). In addition to their disaggregating effects, the ammonia contained in some detergents eliminates the smell of formalin (Cedhagen 1989). After the Ludox treatment great care must be taken to quickly flush the animals from the sieves into a seawater dish and to rinse the sieves immediately to ensure that the animals and sieves are not destroyed by gelling. In samples with very high silt contents (deep-sea samples), the addition of Kaolin powder after fixation and centrifugation in a Levasil® solution (Bayer) can help to bind the finest, most easily suspended particles and thus yields a higher extraction efficiency (Heiner and Neuhaus 2007). In a simplified yet effective version of the Ludox flotation which does not require centrifugation, Somerfield et al. (2005) suggested the repeated decantation of the sample in the Ludox solution. Ludox is meant for the processing of fixed samples, although most animals extracted from live samples will remain moving and only slightly malformed if they are quickly reintroduced into seawater. The detoxification of Ludox by dialysis is cumbersome, but possible (De Jonge and Bouwman 1977). Percoll® or a Percoll®–Sorbitol mixture allow for flotation without causing physiological harm to the animals (Schwinghamer 1981b), but the high costs of these substances make this approach prohibitive for routine processing. For protists, Starink et al. (1994) successfully used a modified Percoll® method. Meiofaunal groups with heavy shells, like foraminiferans or even small molluscs, can be extracted from the sediment using higher density decantation/flotation media, such as the nontoxic but expensive sodium polytungstate SPT (Robinson and Chandler 1993).
3.2.3
Fixation and Preservation
While some meiofaunal taxa are better preserved in alcohol (halacarids) or Bouin’s fluid (turbellarians) (see detailed list in Somerfield et al. 2005, Table 6.2), the usual fixative used for (marine) meiofauna is (buffered) formalin. Here, some hints and caveats are given for optimizing the fixation procedure:
78
3 Sampling and Processing Meiofauna
- Formalin should be added such that its end concentration is an ~4% formaldehyde solution (= about 10% saturated formalin, which is a ~38% aqueous solution of formaldehyde). Membrane-filtered seawater is preferable for the dilution of saturated formalin, since it has an excellent buffer capacity and it enhances the osmolarity of the fixative with favorable results for preservation. An excess of CaCO3 (chips of natural chalk, limestone or marble) in the stock solution will help to increase the alkalinity. For longer preservation without damage to calcified structures, the required careful buffering of the acidic formalin can be done with any alkaline buffer; most commonly hexamethylene tetramine (“Borax”) buffer tablets 7.0 for use in hematology or a 1% mixture of sodium dihydrogenphosphate + disodium hydrogenphophate are used. Another option after 24 h in formalin is refixation in 70% ethanol. Decantation of formalin-fixed samples should be performed under a hood (it is carcinogenic and irritates the skin!). - To preserve many contractile, soft-bodied meiofauna, it is advisable to relax the animals by anesthetization previous to fixation. For seawater fauna, the compound most commonly used for this is a 7% solution of the nontoxic substance MgCl2. 6 H2O, isotonic with seawater. The MgCl2 crystals should be kept dry so that the right concentration can be obtained, since this hygroscopic compound tends to become wet after repeated use. A concentrated stock solution can avoid these problems. Another recommended anesthetic is Menthol (25 g dissolved in 100 ml of 96% ethanol). One drop of this concentrate in a Petri dish of water will relax the animals. - A better fixative than formalin, especially for problematic soft-bodied fauna, is glutaraldehyde. However, its high price and associated handling risks usually restrict this fixative to ultrastructural studies. When an ultrastructural investigation of the animals in addition to the routine treatment is desirable, the use of Trump’s fixative (McDowell 1978) is recommended. The advantage of this fixative is its durability and versatility; specimens or samples can be stored for years at room temperature, well fixed for both light and electron-microscopic inspections. With careful handling, Trump’s can be directly used (even in the field) without any subsequent change of buffer and further handling. Recipe: for 100 ml of Trump’s solution
- Dissolve 1.16 g NaH2PO4 · H2O and 0.27 g NaOH in 86 ml distilled H2O or -
in 86 ml Na cacodylate, 0.1 M (see below) Add 10 ml saturated formaline (»37%, F-79 grade) Add 4 ml 25% glutaraldehyde to make a mixture of final pH 7.2
The highly bactericidal sodium cacodylate buffer, which is used as a solvent for the phosphate buffer, enhances resistance to bacterial deterioration. Recipe: for 100 ml of 0.1 M Na cacodylate (Na dimethylarsenic acid trihydrate) - 2.14 g Na cacodylate trihydrate - Add 0.145 g CaCl2 dihydrate - Fill up to 100 ml with H2O dest.
3.2 Processing of Meiofaunal Samples
79
Sodium cacodylate buffer is also recommended for the bulk-fixation of a sample containing much seawater, because it has been observed that phosphate buffer in seawater causes precipitation. - Microwave fixation is a method described by Berg and Adams (1984) for histological purposes. However, for durable preservation of the animals, microwave treatment must be followed by the application of formalin or some other fixative/preservative. The suitability of this technique remains doubtful, since there is little evidence that it is superior to the traditional, easily available fixation and preservation methods. - In samples retrieved from the deep sea, the distribution pattern of the meiofauna in the core may change considerably during the considerable time needed to retrieve the sampling gear. This can be avoided by using devices (e.g., bell jars) that allow the injection of the fixative directly after sampling while the instrument is still at the bottom. - The appropriate method for long-term storage of selected (museum) specimens depends very much on the meiofaunal group studied and is difficult to generalize (for information, see Part 3 in Higgins and Thiel 1988; Table 6.2 in Somerfield et al. 2005). In general, for museum purposes, formalin-fixed specimens should be refixed in ethanol. Polychaetes tend to break apart when fixed directly in ethanol. - By shock freezing with liquid nitrogen, the predatory attacks of meiofauna can be visualized (Kennedy 1994b). This process is not accessible through conventional fixing or sorting. - A lysis method that makes the rich material archived in many collections available for molecular studies has been developed for nematodes (Bhadury et al. 2006c). Its applicability to other taxa remains to be tested. - Ethanol (70%) is the standard fixative for fauna with calcareous parts. With the progress made in molecular studies it has regained importance, since undamaged DNA is classically extracted only from material fixed in ethanol, preferably in pure ethanol (70% or 96%), while formalin destroys the DNA material. Technique refinement has allowed the extraction and amplification of DNA, even from single meiobenthos individuals (Schizas et al. 1997). Dimethyl sulfoxide (DMSO) is another highly penetrative fixative that is useful for molecular analyses. - The problem of enabling both reliable fixation and preparation for molecular processing has been solved—at least for nematodes—by short-term formalin fixation (Bhadury et al. 2005). It is contended here that general molecular scrutiny preceding storage in museum collections is the approach that should be adopted for diversity assessment in the future, and this can connect traditional morphological research with advanced molecular taxonomy. - There is an interesting resin technique (after fixation in glutaraldehyde fixative) that allows the fixation and permanent storage of large numbers of animals on one microscopic slide (Rieger and Ruppert 1978). The animals are embedded in resin cast in the form of microscopic slides, (optionally) together with some sediment particles. The resulting resin slide is applicable for taxonomic as well as morphological and ecological purposes and allows the storage of significant parts of the
80
3 Sampling and Processing Meiofauna
samples as “ecotype material.” Depending on the taxonomic group, one disadvantage of this is the permanent nature of the enclosure; there is no option to change the animals’ positions at a later date if this is required during inspection. - Double-sided microscopic slides are a valuable tool for the close inspection and/or storage of specimens (Shirayama et al. 1993). Special storage slides are also commonly used in studies of foraminiferans and ostracods. For nematodes that are permanently mounted, Bates (1997) provides a recipe for sealing with Glyceel, a traditional embedding compound. - In cases where a sample contains too many animals for easy enumeration of all specimens, the total extracted meiofauna should be subdivided with a sample splitter to facilitate quantitative evaluation. A precision-made and statistically tested instrument allows subdivision into exactly equal parts. Today, the most reliable type, specially designed for meiofauna, is the “Jensen splitter” (Jensen 1982), which can be constructed in a good workshop. There are also some modifications of this very useful instrument for meiofaunal work. Quantitative drainage of the chambers can be improved by giving the bottoms of the chambers a slight slope towards the stopper holes. When mounted on somewhat longer supporting stands than those suggested by Jensen, the rinsing of the subsamples into dishes becomes easier. For other, simpler subdividing techniques, see Somerfield et al. (2005).
3.2.4 Processing and Identifying Meiofaunal Organisms A few instruments that are needed for the various procedures involved in handling and studying meiofaunal organisms will be mentioned here. They have proven very helpful in daily routine work with meiofauna, but are usually omitted from descriptions. Some hints about new identification techniques are also given at the end of this chapter. - The quantitative sorting of large numbers of organisms is much easier and more reliably done in specially made rectangular perspex trays called “Bogorov trays,” named after a Russian plankton specialist (Fig. 3.5). The inner surfaces of these trays are subdivided by a system of leveled bars, resulting in a meandering system of stripes for counting the extracted organisms. The widths of these stripes should be slightly less than the diameter of the eye field at the normal magnification of the dissecting microscope used for routine scanning. The advantage of this device when used for quantitative enumeration over any kind of Petri dish is its square shape and the bars. In the case of an accidental push or when working on board a ship, these bars largely prevent swashing and dislocation of the specimens and so they enable precise counting to continue. Straight lines, 1 cm apart and scratched across the bottom of the tray, support orientation. The overall size of the dish should be conveniently adjusted to the size of the microscopic stage. Slats on the underside of the dish prevent undesirable adhesion if the microscopic stage surface should become wet. A recent development, derived from the typical Bogorov tray, is a sorting tray where the
3.2 Processing of Meiofaunal Samples
81
Fig. 3.5 Sorting tray for meiofauna (a modified Bogorov tray); size 8 × 14 cm, height: 1.5 cm
meandering trough system has been milled by a cutter from a solid perspex plate (Fig. 3.5). Its precise manufacturing avoids the inevitable crevices and imperfections resulting from the gluing of perspex strips. For quantitative evaluation of the complete meiofaunal assemblage in a sample, the extracted contents should be inspected once at 10–15× magnification and another time at 30–50×. - For the removal of single live specimens from water samples, the use of a “mouth pipette” allows work to be completed far more rapidly than when using normal pipetting. We use the rimmed plastic protective tubes of syringe needles cut to a convenient length as mouth pieces for the pipettes. They allow a good yet relaxed hold between the teeth and are easily and inexpensively exchanged. However, each pipetting bears the risk of losing specimens in the glass tip and sucking up other unwanted particles. Coating with silicone can partly solve this problem. Animals clinging to the walls and the bottom of the dish are difficult to remove with a (mouth) pipette. - A far “cleaner” and more selective approach, but also one that is far more tedious, is “fishing” out specimens with thin needles or loops, formerly known as “irwin loops.” Stainless steel needle holders that are heavy enough to effectively reduce the natural slight vibrations (the resting tone) of the hand can be purchased from medical instrument suppliers. Their grip should firmly hold even the minute stainless steel pins used in entomology for small insects. The author found the long, flexible, yet sufficiently stiff needles used in acupuncture to be the optimal instruments for inspecting meiofauna; these are usable even without being clamped in a holder. The thinner types are extremely fine at the tip yet are durable and easier to handle than minute insect needles. The use of loops is recommended for removing more spherical organisms (ostracods, nauplii, cladocerans, eggs). They can be manufactured from very thin tungsten or nickel wire, which is available from suppliers of electronic equipment or laboratory equipment. The wire is then
82
3 Sampling and Processing Meiofauna
tightly twisted from both ends around a hooked needle to produce a loop. Fine adjustment to the desired loop diameter is achieved by further twisting it with another hooked handle. After slightly flattening the resulting loop by pressing it between the blades of forceps (combination pliers) and bending it into a convenient angle, the remaining ends of the wire are tightly twisted to form the shaft of the instrument, which is then inserted or glued into the grip of a needle holder. - Satisfying microphotographs of meiofauna are rare. For a professional standard of quality with sufficient depth of focus, special and expensive photomicroscopes (photostereomicroscopes are often even better) are required. When combined with microflashes they enable live photography. The problem of maintaining the animal in a desired position without risking smashes or losses is elegantly, though expensively, solved by the “compression chamber” devised by Uhlig and Heimberg (1981), which is commercially available. The use of video and digital cameras adapted to microscopes has considerably broadened the possibilities for the optical documentation of meiofauna. However, creating a didactically useful film on meiofauna requires that a great deal of professional effort be directed into “cutting” and editing. Two fully edited videofilms with commentary in English are available that aptly present the wide array of groups of organisms and their characteristics: (a)
(b)
Meiofauna of Marine Sediments (21 min, optionally PAL and NRSC systems) by J.A. Ott and A. Bochdansky, Vienna, Austria. This may be purchased from Verein Pro Mare, Biozentrum, Althahnstr. 14, A 1090, Vienna, Austria. Cryptic Fauna of Marine Sand (20 min) by R.P. Higgins, which can be obtained through the Society of Integrative and Comparative Biology. Digitized copies may become available soon.
- Today, software for computer-aided optometric studies and image analyses are widely available and can be conveniently used on the monitor screen. Using semi-automated image analysis software and conversion factors from the literature (see Sect. 9.3.2), Thomsen (1991) calculated the number, size, volume, and biomass of bacteria and meiofauna. Other novel techniques in microscopy (image EELS, laser scanning, 4D, cLS) can greatly increase our knowledge of analytical details, developmental processes, and structural complexes in meiofauna, contributing to novel interpretations of old problems (see Sect. 10). - Protocols for the computer-assisted, semi-automated calculation of meiofaunal biomass have been developed using digital microphotography (Baguley et al. 2004). This method combines a relatively large throughput with nondestructive handling, which enables subsequent taxonomic work. Traditional methods are calculations of volumes on the basis of a dot-counting overlay on top of microphotographs (e.g., Kaye 1993), a method derived from microbiology (bacteria volumes). - Electronic media and new software programs have increased the development of computer-assisted, partly interactive, identification keys for meiofaunal groups (e.g., Diederich et al. 2000 for nematodes; Wells 2007 for harpacticoids). Although an ever-growing number of pictures of meiofauna can be found on
3.3 Extraction of Pore Water
83
various websites and homepages, the “identification” of the depicted forms, when not controlled by a specialist, should be met with skepticism. This pathway to publication is by no means scientifically warranted, and so these often impressive illustrations should only be taken as helpful indications. Electronic support for the design of illustrations (Bouck and Thistle 1998) can facilitate scientific drawing, a technique that is necessary for taxonomic descriptions but is often found to be cumbersome and difficult.
Box 3.1 Sampling and Evaluating Meiofauna The statistical precautions, sampling strategies and general devices used in benthic research are also relevant for the sampling of meiofauna and are wellcovered in pertinent methodological works, the latest being that by Eleftheriou and McIntyre (2005). The kind of sample processing applied is related to the type of sediment and goal of the study. Provided below is a summary of some practical hints for reliable treatment. - The correct sample size and number per study area to use depends on the number of specimens present. Therefore, it is important to pre-sample, evaluate subsamples and inspect living fauna. - Tube corers of varying size and numbers per sampler (Kajak corer, Craib corer, multicorer) are the most common and, if used correctly, the most reliable tools that have been tested for different sediment types and depths. Grab samplers (e.g., the van Veen grab) do not generally work quantitatively. - Live meiofauna are usually concentrated by decantation or by the “seawater ice method,” especially for soft-bodied animals. - For easier recognition of meiofauna the sample should be pre-stained (Rose Bengal). - The most effective quantitative extraction methods for fixed meiofauna are decanting or elutriation (for sands), and flotation with Ludox® for fine sand and mud. - For most studies, meiofauna are fixed in buffered formalin. For taxonomic purposes, soft-bodied taxa should be relaxed beforehand (MgCl2). Re-fixation in ethanol is best for long-term storage. Trump’s fixative allows for ultramicroscopic sections. - Fixation for molecular work is usually by alcohol; preferably by pure ethanol (96%). - To speed up evaluations of samples with abundant meiofauna, the use of a sample splitter is advised. - The use of a Bogorov tray is a more reliable and less cumbersome enumeration technique than sorting in Petri dishes. - Fine acupuncture needles are perfect instruments for sorting and dissecting meiofauna, as are tiny twisted loops of thin tungsten wire (“IRWIN loops”) for spherical objects.
84
3.3
3 Sampling and Processing Meiofauna
Extraction of Pore Water
The importance of pore water for meiobenthic animals requires analyses with methods that are mostly based on the extraction of free pore water. The inherent problem with this analytical procedure is the destruction of fine-scaled physical and chemical gradients which often characterize the meiobenthic habitat. Suction sampling. In accessible intertidal areas, the easiest way to extract pore water is to gently push a disposable glass pipette or a hypodermic syringe directly into the sediment and to obtain the pore water by suction. Water at the different horizons can be collected from a sediment core through lateral openings in the corer that are initially sealed. Pressing the core with a tightly fitting piston facilitates the extraction of the pore water (Jahnke 1988). A piece of fine nylon gauze (20 µm mesh size) or a commercially available copper grid used for transmission electron microscopy, sealed onto the tip of the pipette, prevents the clogging of the small opening by sediment particles. An effective and easily built device for obtaining pore water is a modified glass capillary (Fig. 3.6a). It has a tightly fitting syringe cap into which many holes have
a
b
Fig. 3.6a–b Pore water samplers. a Capillary syringe (after Howes and Wakeham 1985). b Pore water lance (after Giere et al. 1988a)
3.3 Extraction of Pore Water
85
been drilled with a microdrill. The opposite end of the glass capillary is connected to thin gas chromatography tubing (important: teflon tubing is not gas-tight and cannot be used to determine oxygen and hydrogen sulfide in pore water), connected to the glass tube via a tightly sealed syringe stub (Luer stub). Another tight syringe stub at the other end of the flexible tubing is connected to a small syringe (1 ml). Drawing the piston of the syringe with some care enables pore water to be obtained without air bubbles, which tend to penetrate into the tubes at their connecting ends. The insertion of a disposable microfilter in the tubing will filter turbid pore water. A pore water sampler particularly designed for flushing with nitrogen and which is therefore well suited to measurements of oxygen or sulfide content has been constructed by Zimmermann et al. (1978). Its considerable size renders it more suitable for permanent installation in the sediment to be investigated. Dye (1978) constructed a pointed metal tube with a sample port covered with fine gauze that, during insertion into the sediment, is tightly capped with a PVC sleeve. When the desired depth is attained, the sleeve is slid up and the pore water can enter the inner chamber from a connected syringe. For sublittoral sediments, a diver-operated “pore water lance” allows the extraction of pore water from a series of horizons. A modified version of the pore water lance, originally constructed by Balzer (Kiel), is described in detail in Giere et al. (1988) and is depicted in Fig. 3.6b. Through the series of samples extracted with this instrument, differentiated gradient profiles of pore water can be obtained for measurements. Its versatility and sturdiness make the pore water lance a reliable and effective tool at depths accessible to diving. Similar designs are now in use by several working groups. For soft bottoms in deep water, more complicated “harpoon samplers” with triggering devices have been constructed. As a spring-loaded piston moves up in the sampling chamber, pore water is sampled and filtered (Barnes 1973). Another suction corer for use in deep water has been designed by Sayles et al. (1976) for geochemical studies, but it can be modified for the smaller dimensions associated with meiobenthic studies. A damped piston moving in a cylinder is mounted on top of a pointed metal tube with a series of ports capped with some filtering device. Through the action of the triggered piston, pore water is sucked through the ports into a series of chambers mounted in the inner lumen of the tube. When hauled on deck, the water can be extracted from the chambers by syringes. Squeeze sampling. Reeburgh (1967) used a plastic squeezer equipped with filter screens on top of a drain tube. Mounted on a glass bottle, the squeezer is subjected to gas pressure and the pore water is efficiently squeezed and filtered out of the sediment. Centrifuge sampling. In coarse sand, and particularly in sediments of low porosity, the extraction of pore water by suction or squeezing becomes problematic and inefficient. Here, centrifuging yields much better results. After only a short centrifugation time and without further contamination, pore water can be extracted in quantities larger than those obtained by squeezing methods. Saager et al. (1990) developed specially manufactured centrifuge tubes with built-in filter
86
3 Sampling and Processing Meiofauna
units that filter the samples without any contamination. A further step towards maintaining an undisturbed pore water sample is to use special corers adapted to function directly as centrifuge tubes. More detailed reading: Giere et al. (1988); Pfannkuche and Thiel (1988); Somerfield et al. (2005).
Chapter 4
Biological Characteristics of Meiofauna
4.1
Adaptations to the Biotope
The heterogeneity of meiofaunal habitats is so large and meiobenthic taxa so diverse that there are only a few general trends in morphological adaptations, and these mostly apply to mesopsammic fauna. The most widespread and clearest adaptive features to a particular mesopsammic environment have evolved in the meiofauna of medium and coarse sands. The formative constraints on and the premises for entering the world of narrow voids in this interstitial environment will be detailed in the following sections.
4.1.1
Adaptations to Narrow Spaces: Miniaturization, Elongation, Flexibility
The prime requirement for all meiofauna to be successful is to be small, at least in one dimension (e.g., body width). This adaptive constraint becomes evident when comparing body sizes of related animals from various habitats (Fig. 4.1) or the proportion of small animals within a taxon for various habitats (Fig. 4.2) Particularly in those meiobenthic animals that predominantly belong to macrobenthic groups (Table 4.1), the decrease in body size is striking and believed to have intrinsic lower limits for the various animal groups: 0.5–1 mm in many taxa (Swedmark 1964), 0.3 mm in copepods (Serban 1960). Dwarfism is mainly accomplished by reducing the number of cells while keeping the average cell size fairly constant. Rotifers consist of only about 1,000 cells, and the nematode Caenorhabditis elegans in the male phase of 959 somatic cells (and in the subsequent female phase of 1,031 cells), of which every lineage is thoroughly known. Yet, as an exception, loriciferan species (sometimes only 50 µm long; perhaps the smallest known metazoans) reportedly consist of as many as 10,000 or more cells (Kristensen 1991a). Dwarfism often leads to a simplification of body organization or to a loss of organs (number of segments, legs, gonads, loss of eyes), and, along the lines of regressive evolution, it can phylogenetically lead to new taxa that are restricted to interstitial refuges (see Chap. 6). O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
87
88
4 Biological Characteristics of Meiofauna 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 mm
open water
muds
phytal
sand
Fig. 4.1 A comparison of copepod sizes (in mm) in various habitats. (After Kunz 1935)
80
70
60
50
40
30
20
10
0 %
in phytal
in muds
in coarse sand
in medium sand
Fig. 4.2 The proportions of small copepods (< 0.5 mm length) in various habitats. (After Lang 1948)
4.1 Adaptations to the Biotope
89
Table 4.1 Examples of dwarfism in interstitial animals belonging to large taxa (after Remane 1933, extended) Animal group (genus, species) Body size [mm] 0.4 Cnidaria (Halammohydra ) 0.4 Polychaeta (Nerillidium, Diurodrilus ) 1.0 Gastropoda (Microhedyle ) 1.5 Priapulida (Meiopriapulus ) 0.8 Crustacea, Tanaidacea (Pseudotanais mortenseni) 1.2 Crustacea, Isopoda (Microcharon sp.) 1.0 Crustacea, Amphipoda (Ingolfiella petkovskii) 3.0 Echinodermata, Holothuria (Labidoplax) 3–4 Ascidiacea (Psammostyela )
Adaptation to the interstitial habitat is often achieved by reducing the width only, and slender bodies can be surprisingly long (see, e.g., among ciliates the Trachelocercidae, among turbellarians the Coelogynoporidae, among nematodes the Stilbonematinae, among polychaetes the “archiannelid” Polygordius, among oligochaetes the gutless phallodriline genera, e.g., Olavius, and among the Acari the oribatid Nematalycidae). A filiform body shape, sometimes with a corresponding reorganization of the internal organs, offers advantages that are only indirectly related to the small habitat dimensions. The locomotory active surface (via ciliation or body musculature) becomes enlarged, effects of adhesion and anchoring become improved, and the extreme ratio of body surface to body diameter supports transepidermal uptake and diffusion of dissolved organic substances and enables a complex pattern of tactile stimuli. While the length-to-width ratio normally ranges from 3:1 to 10:1, it can reach 100:1 in, for example, interstitial nemerteans, some polychaetes and oligochaetes (compare the “width index” of Remane 1933). Some specialized nematodes, crustaceans and other groups attain similar size relations (Jensen 1986; 1987b). This convergent adaptive advantage of body elongation is particularly apparent in representatives that belong to groups that normally have a more ovalto-round shape (Fig. 4.3). Extreme flexibility is an important adaptation for life in the labyrinth of the interstitial system of sand grains. In soft-bodied, vermiform animals this is rather easily accomplished, but in animal groups with a chitinous cuticle, which often have compact body sections, the body must become more articulated, resulting in small, uniform segments (e.g., in interstitial harpacticoids, tanaidaceans, isopods, amphipods). Through this modification, the body attains an overall vermiform shape, flexible enough to easily allow U-turns. In those interstitial worms where the number of segments is not constantly fixed, the number of segments is often greatly enhanced (some oligochaetes have >150 segments, and the polychaete Polygordius has up to 185 segments). Perhaps linked to the trend for elongating the body and clinging to the substratum is the frequent development of a tapering body end, forming a tail, which is convergently present in numerous interstitial animal groups (Fig. 4.4). It is doubtful
Avelia (Ciliata)
Vannuccia (Turbellaria) Haplognathia (Gnathostomulida)
Polygordius (Polychaeta)
Cylindropsyllus (Harpacticoida)
Nematotanais (Tanaidacea)
Nematalycus (Acari)
Fig. 4.3 Convergent trend for body elongation in interstitial meiofauna. All figures are depicted at the same scale. (After Remane 1933, extended)
Marenda nematoides (Foraminifera)
90 4 Biological Characteristics of Meiofauna
Spirostomum filum Thylacorhynchus caudatus
PLATHELMINTHES
Boreocelis urodasyoides
GASTROTRICHA
Urodasys viviparus
NEMATODES
Trefusia longicauda
Fig. 4.4 Interstitial animals with a tail. All figures are depicted at the same scale. (After Ax 1963, extended)
PRIAPULIDA CILIATA
Tubiluchus troglodytes
ASCIDIACEA
Heterostigma fagei
4.1 Adaptations to the Biotope 91
92
4 Biological Characteristics of Meiofauna
that there is a common function for this structure. While the tails in animals like Urodasys (Gastrotricha), Tubiluchus troglodytes (Priapulida), Trefusia longicauda (Nematoda), or Batillipes bullacaudatus (Tardigrada, with an inflatable terminal bubble) appear to function as a device anchoring against displacement, in others it has instead been interpreted as a tactile organ (Ax 1963). Another frequent adaptation to the interstitial environment is flattening of the body. This appears to enable the animals to squeeze through narrow crevices and to increase body flexibility. Also, it enhances the forces of friction against the substratum, important for effective movement by wriggling. As in body elongation, there are also physiological and behavioral advantages to becoming flat (oxygen diffusion, transepidermal nutrient uptake, enlargement of the contact zone with the sand grains, development of a “creeping sole”). In groups with a roundish shape, a dorsoventral depression is an exceptional phenomenon that occurs mainly in interstitial forms (e.g., in the ciliates Remanella, Kentrophoros and Loxophyllum, in some chromadorid nematodes, in the oligochaete Olavius planus, and even in some crustaceans like mystacocarids and specialized ostracods).
4.1.2
Adaptations to the Mobile Environment: Adhesion, Special Locomotion, Reinforcing Structures
For many meiobenthic animals, adhesion is an important adaptation that allows them to live in a mobile environment. Solid external cuticular plates, shells and rings can serve as anchoring devices that increase adhesion to the sediment. These characterize some interstitial halacarids, ostracods and harpacticoid species as well as epsilonematid and desmoscolecid nematodes. Combined with scales, spicules and thorns, these structures are also frequently found in gastrotrichs, kinorhynchs, loriciferans, priapulids (particularly the priapulid larva), and in aplacophoran molluscs. The development of adhesive or “haptic” structures is particularly frequent in the interstitial fauna of mobile sands (Fig. 4.5). Aside from these mechanical/structural means, adhesion to sand grains is also achieved through a variety of physiological and behavioral adaptations (Fig. 4.6). Functionally, these organs are mostly based on the “dual gland system” (e.g., many nematodes, gastrotrichs and rotifers; Tyler 1976; Hermans 1983), where one gland contains the adhesive mucus and the neighboring gland produces the releasing compound. – Secretion of viscid mucus on the whole body surface: many nematodes, turbellarians, gastropods, annelids – Excretion of sticky coelomic fluid through caudal coelomopores in interstitial oligochaetes (Marionina) – Adhesive glands concentrated in particular regions, forming special structures: an adhesive “girdle” in the turbellarian Rhinepera and the polychaete Dinophilus, posterior haptic “toes” in many rotifers, gastrotrichs and in the annnelids Diurodrilus, Saccocirrus, Hesionides, Protodrilus,
4.1 Adaptations to the Biotope
93
60
50
40
30
20
10
0 %
In mud *
In mud *
Among plants
In sand
* Different sites
Fig. 4.5 Percentage of haptic animals among meiofauna from various habitats. (After Remane 1933)
– Haptic setae: in ciliates (cirri), gastrotrichs, many nematodes (Draconematidae, Epsilonematidae), kinorhynchs and in many ostracods – Strongly adhesive flagellae in interstitial flagellates (e.g., Amphidinium testudo) Besides glandular adhesion, there are examples of mechanical adhesion, such as the viscid disks on the toes of the tardigrade Batillipes (see Fig. 5.27) or the flattened body of the harpacticoid Porcellidium (see Figs. 5.34 and 8.10a). Parallel to the structural and morphological features mentioned above, an effective adhesion process requires behavioral adaptations. The animals must react to perilous water currents, sediment displacement or other sudden disturbances with an immediate adhesion impulse. After a while, contact through glandular excretions will be chemically released or will become physically ruptured by the movements of the animal (the oligochaete Marionina). In the phytal, where larger and more complex substrates are present for clinging, typical adaptations of the meiofauna include clinging legs, often equipped with long bristles and spines (Halacarida, Tardigrada, Harpacticoida, Ostracoda; see Sect. 8.5), or the suckerlike protrusion in the harpacticoid copepod Porcellidium (see Fig. 8.10a). Many meiobenthic animals have developed specialized locomotive organs and characteristic modes of locomotion, often in conjunction with adhesion or elongation of the body.
Thaumastoderma (Gastrotricha)
Batillipes (Tardigrada) Diurodrilus (Polychaeta)
Rhinepera (Plathelminthes)
Cicerina (Plathelminthes)
Encentrum (Rotifera)
Fig. 4.6 Interstitial animals with characteristic adhesive structures. All figures are depicted at the same scale (After Remane 1933, extended)
Turbanella (Gastrotricha)
94 4 Biological Characteristics of Meiofauna
4.1 Adaptations to the Biotope
95
– Gliding by means of ciliation (Fig. 4.7). An unusually large number of ciliated animals live interstitially in the sand. Here, gliding is not only typical of ciliates, turbellarians and gastrotrichs, where it is a group characteristic, but also of many meiobenthic polychaetes (archiannelids), molluscs, hydroids (Halammohydra) and, in a modified way, rotifers that move with their wheel organ. Many ciliated animals move very quickly, particularly members of the proseriate platyhelminth family Otoplanidae, and some gastrotrichs seem to “swim” through the coarse sand, although they are animals that show positive thigmotactic behavior and so in fact remain in contact with the substratum. – Wriggling, i.e., undulatory propulsion by alternate pushing and bending, is typically used by nematodes that possess only longitudinal musculature. However, wriggling also occurs in specialized oligochaetes like the enchytraeid Grania, which has a rather solid cuticle and a thick longitudinal musculature but only a thin circular muscle layer. Also, some gastrotrichs with a thick cuticle and some turbellarians exhibit writhing and kicking movements. Some psammobiotic harpacticoids do not move through leg strokes but with a general wriggling of the vermiform body, which pushes the animal through the sand. – Crawling on the sand grains occurs relatively rarely in some slow-moving groups. It is typical of halacarids, ostracodes and tardigrades. – Burrowing is common only in the meiobenthos of soft muds, where it is achieved through peristaltic contractions of the body musculature (as in some annelids) or
Fig. 4.7 Percentage of ciliated meiobenthos among total fauna. (After Remane 1933)
96
4 Biological Characteristics of Meiofauna
by the typical protrusion–eversion of an introvert (sipunculids, priapulids, loriciferans). Kinorhynchs derive their name from this kind of motion. In crustaceans inhabiting soft muds, digging is realized by powerful leg movements (some harpacticoids, ostracods, tanaids and cladocerans). – Climbing through the thicket of sand grains or plants by extending long, elastic and often sticky appendages and drawing the body behind is a typical movement of some exceptional interstitial animals whose “normal” relatives are pelagic swimmers or sessile benthic forms: the mesopsammic genera of cnidarians (Psammohydra, Otohydra), the meiobenthic anthozoan Sphenotrochus, the slow-moving brachiopod Gwynia, the mobile bryozoon Monobryozoon, the small holothurian Leptosynapta, or the sand-living ascidian Psammostyela. – Somewhat related to climbing is the iterative contraction–elongation of the whole body with alternating fixation of the body ends (looping). Characteristic of inch worms (the larvae of geometrid butterflies) and hirudinids, this type of movement occurs fairly frequently in meiobenthic animals, mostly in conjunction with rhythmic body elongation and subsequent adhesion to the substratum (Fig. 4.8). It characterizes the movements of very different meiobenthos such as the hydrozoans Psammohydra nanna, Cryptohydra thieli, and Protohydra leuckarti, epsilonematid and draconematid nematodes, the gastrotrich Macrodasys, several turbellarian platyhelminths, synaptid holothurians, the tiny gastropod Unela, rotifers adhering alternatively with their “toes” and their wheel organs (such as Philodina), and the annelid Rheomorpha.
Macrodasys (Gastrotricha)
Draconema (Nematoda)
Rheomorpha (Annelida)
Epsilonema (Nematoda)
Fig. 4.8 “Looping” movements in various meiobenthic taxa. All figures are depicted at the same scale. (Various authors)
4.1 Adaptations to the Biotope
97
Members of meiobenthic groups living in the interstitials of sand are frequently “armored” with structures interpreted as reinforcements and protection from mechanical stress (pressure, agitation of sediment; Fig. 4.9). However, in cases where the animals live in rarely exposed soft mud, the mechanically protective character of these structures is doubtful and they probably instead provide a means of increased adhesion (see above). Internal crystals, fibrils and spicules occur repeatedly in interstitial animals. Some ciliates have conspicuous fibrils (e.g., Diophrys, Aspidisca) and spicules (Remanella) in their cytoplasm. The turbellarian Acanthomacrostomum spiculiferum derives its species name from its armature of internal spicules in “sagittocysts” (Gschwentner et al. 2002). The family of acoel plathelminths Sagittiferidae is named after their numerous internal spicules. Other meiobenthic forms belonging to soft-bodied groups like molluscs and holothurians contain a lattice of characteristic spicules: for example the snails Rhodope and Hedylopsis as well as the minute sea cucumber Leptosynapta (see Rieger and Sterrer 1975). Another way of stiffening the body is the vasculariza---tion and turgescence of cells in tissues or intercellular spaces. Cushion-like structures occur in some polychaetes, many gastrotrichs (Rieger et al. 1974; Teuchert 1978), some monhysterid nematodes (van de Velde and Coomans 1989), and turbellarians. In the turbellarians Nematoplana and Coelogynopora, intestinal cells form a stiffening, “chordoid” tissue extending through the longitudinal axis of the body. Also, the characteristic accumulation of turgescent cells in the prostomium of the interstitial oligochaete genus Aktedrilus (Tubificidae) probably has a stiffening function that adapts the worm to burrowing through the sand. Alveolar external tissue has also developed in the interstitial chaetognath Spadella interstitialis (Kapp and Giere 2005). The layer of vacuolized chambers in the egg capsules of some turbellarians from exposed sands should also be mentioned in this context.
4.1.3
Adaptations to the Three-Dimensional Dark Environment: Static Organs, Reduction of Pigment and Eyes
The presence of static organs in many meiofauna is considered to result from the need for orientation in a wide three-dimensional sediment system—comparable to the corresponding phenomenon seen for many members living in the water column. As an exception from the general body structure of the group, static organs (organelles) occur in some meiobenthic members of ciliates (Remanella) and hydroids (Halammohydra, Pinushydra), turbellarians (Acoela and otoplanid Proseriata), enoplid Nemertinea (Ototyphlonemertes), marine enchytraeid oligochaetes (Grania), the isopod group Anthuridea, and the meiobenthic synaptid holothurians. In the latter, they have developed in not only the interstitial forms but also among the larger relatives. Body pigmentation is reduced in many meiobenthic animals that live in a dark environment. They are whitish-opaque or transparent, which enables microscopic
Acanthiella (Plathelminthes)
Rhodope (Gastropoda)
Falcidens (Aculifera)
Lepidodasys (Gastrotricha)
Fig. 4.9 Structures that reinforce the bodies of meiobenthic animals. All figures are depicted at the same scale. (Various authors)
Nematoplana (Plathelminthes)
98 4 Biological Characteristics of Meiofauna
4.1 Adaptations to the Biotope
99
scrutiny of their internal organs for identification without dissection. The lack of body pigments and prevalence of transparent tissues is particularly striking in comparison with their larger relatives from the epibenthos. However, there are some exceptions to the rule of general pigment reduction in meiobenthos. Particularly in the calcareous sands of warm-water regions, white pigmentation of meiofauna is common. Symbiotic relationships with white sulfur bacteria make gutless tubificids, stilbonematid nematodes and some ciliates shiny white. Sometimes, even an orange-to-pink or yellowish color (Meiopriapulus fijiensis, polychaetes, ostracods, some harpacticoids) blends in well with the pink-to-orange shine of some coralline sands. In phytal meiofauna we find some colorful halacarids (Copidognathus) and turbellarians (Procerodes). Whether this coloration actually represents a camouflage adaptation aimed at avoiding motion-oriented or nonselective macrobenthic predators is unknown. In any case, it sometimes makes them difficult for the researcher to discern. Endobenthic meiofauna are mostly blind; only a few epibenthic forms such as naidid oligochaetes, ostracods, harpacticoids, and tardigrades have retained simple eyes. The reduction of eyes is particularly striking in animal groups where eyes usually are present, e.g., in polychaetes.
4.1.4
Adaptations Related to Reproduction and Development
The minute body sizes of the meiobenthos largely determine their developmental modes, since they do not support the shedding of abundant gametes. Especially in the interstitial, specialized approaches to sperm transfer, fertilization and development have been developed. Due to their reduced sizes, many interstitial animals have only a single ovary. Most of them produce just one or only a few eggs (Table 4.2). In ostracods, where paired penes are common, the meiobenthically small species often develop just one penis. The formation of loose sperm bundles (spermatozeugmata) or more complicated spermatophores considerably enhances the chances of Table 4.2 Reduced egg number in some typical meiobenthos (from various authors) Taxon Egg number 1–4 Halammohydra (cnidarians) Kalyptorhynchia (turbellarians) 1–3 2–5 Trilobodrilus (polychaetes) Microdriline oligochaetes 1–4 1 Bathynella (syncarid crustaceans) Microparasellidae (isopods) 1–4 1–2 Angeliera (isopods) 1–2 Bogidiella (amphipods) 1–2 Uncinotarsus (amphipods) 2–3 Leptosynapta (holothurians)
100
4 Biological Characteristics of Meiofauna
fertilization and reduces the risk of sperm loss in a world where spermatozoa released into the pore water would not be carried any great distance by water currents. Sometimes sperm bundles are deposited on the substratum, e.g., in the polychaete Nerilla, where they are subsequently taken up by the partner. More frequently they are transferred directly onto the skin of the partner, but the subsequent process of their transfer to the eggs often remains unknown (histolysis?): this occurs in some tardigrades, hesionid polychaetes, tubificid oligochaetes (Aktedrilus monospermathecus), and kinorhynchs. Hypodermal injection occurs in the polychaete Trilobodrilus, the snail Hedylopsis, in some gastrotrichs, and (particularly well studied) in many rotifers, where the initially produced sterile sperm histolytically “clears” a route towards the eggs for the subsequently released fertile sperm. The most direct way of sperm transfer ensuring fertilization is copulation, which frequently occurs in meiobenthic animals (e.g., in polychaetes, see 2efv Fig. 5.43). This not only entails the “construction” of organs that directly participate in sperm transmittance (cirrus, penis) and often sperm storage (vesicula seminalis, spermatheca), but it also necessitates sense organs to find the partner from a distance and behavioral cooperation in order to obtain the correct positioning needed to complete copulation successfully. In groups such as turbellarians or oligochaetes, the whole taxon is characterized by this most evolved approach to reproduction; however, meiobenthic representatives from taxa which normally simply shed their spermatozoa into the ambient water, like polychaetes, have also developed penislike organs, complicated modes of copulation (e.g., Questidae) and elaborate sexual behavior. The sexual foreplay that has developed in some turbellarians, gastrotrichs and polychaetes is astonishingly complex for these “lower invertebrates.” One important way to further enhance the chances of successful reproduction in a world with a limited individual range of distribution is hermaphroditism. Most macrodasyoid gastrotrichs (e.g., Turbanella) and small meiobenthic polychaetes are hermaphrodites, including Ophryotrocha, Microphthalmus, and some tardigrades (some Echiniscidae; many Eutradigrada), hydroids (Halammohydra, Otohydra), and even some meiobenthic crustaceans (some Tanaidacea) and holothurians (Leptosynapta). Groups such as gnathostomulids, turbellarians and oligochaetes exhibit hermaphroditism as a general feature of the taxon, and so are preadapted to a meiobenthic life. The low chances of survival of many meiobenthic species, aggravated by their reduced mobility and low number of embryos, are compensated for by direct development without any free-swimming larvae, by relatively short generation times, and frequently continuous reproductive periods. Like the sperm, the fertilized eggs do not freely float in the pore water. Instead, with their sticky surfaces they adhere to sediment particles or are ensheathed in a capsule (turbellarians) or cocoon (oligochaetes, some polychaetes), which becomes glued to the surface of a sand grain. Loss of offspring is further reduced by hatching well-developed and self-sufficient juveniles. Viviparity is not uncommon (e.g., the gastrotrich Urodasys viviparus, the rotifer Rotaria, the hydroids Armorhydra and Otohydra vagans). Parental brood protection of the embryos occurs frequently in special capsules or pouches of/on the body. Nerillid polychaetes attach their offspring to the posterior part of
4.1 Adaptations to the Biotope
101
the body, and ovisacs (in harpacticoids) or “marsupia” (in peracarid crustaceans) provide protection to the developing embryos. Rapid asexual multiplication (budding or fissure) occurs in some turbellarians (Macrostomida), some polychaetes (Syllidae), in most Naididae (Oligochaeta), and in aeolosomatid annelids. Parthenogenetic reproduction is common in rotifers, chaetonotoid gastrotrichs, darwinuloid ostracods and in cladocerans. In meiobenthic animal groups, the tendency to abbreviate ontogeny has often led to neotenic or progenetic development (Westheide 1984, 1987a). Progenesis characterizes many polychaetes (e.g., Ophryotrocha, Dinophilus, Apodotrocha, Trilobodrilus), which have retained larval traits such as ciliary tufts and bands, and a lack of coelomic compartmentation. However, progenesis also occurs in meiobenthic opisthobranch molluscs, in many gastrotrichs and interstitial Cnidaria, and has also been found in some loriciferans. In some (freshwater) crustacean groups, neoteny has led to phylogenetically interesting new groups (see Chap. 6). Encystment and dormancy often help animals to survive periods of hostile life conditions and are particularly frequent in freshwater meiofauna (Alekseev et al. 2007). Turbellarians, cladocerans, rotifers and some harpacticoids have stages of “dormancy.” In freshwater biotopes, tardigrades, some ostracods and some nematodes are perfectly adapted to survival under extreme conditions such as dryness through the development of “cryptobiosis.” In the marine realm, encystment occurs in tardigrades, in many turbellarians from salt marshes, in Dinophilus taeniatus (Polychaeta), and in the harpacticoid Heteropsyllus nunni (Coull and Grant 1981; Williams-Howze and Coull 1992). Some antarctic harpacticoids seem to have a diapause stage during their copepodite phase (Dahms 1991). Interestingly, many resting eggs of planktonic copepods occur in sediment and hatch under more favorable conditions (e.g., increasing spring temperatures). Dormant copepod eggs were regularly found in sediment samples of anoxic mud from the Black Sea (Luth et al. 1999; Sergeeva and Gulin 2007). Larval stages of some deep-sea species of Rugiloricus (Loricifera) retard their hatching through encystation (Kristensen 1991b). In all of these developmental adaptations, the respiration is massively reduced and metabolism becomes deeply quiescent. Despite the diversity of the numerous animal groups found in the meiobenthos, and despite differences in organization, structure, taxonomic rank and phylogenetic age, they have all experienced the constraints and dynamics of the interstitial habitat. They have all have been subject to “integrating adaptations” (Remane 1952) which can form analogous specializations with often surprisingly uniform convergent traits in groups of different systematic position. Here, the meiofauna exhibits parallels with troglobitic fauna, edaphic assemblages, and partly also phytal communities (e.g., Remane 1952). More detailed reading: Ax (1966, 1969); Swedmark (1964); Schwoerbel (1967).
102
4 Biological Characteristics of Meiofauna
Box 4.1 Adaptations of a Typical Meiobenthic Animal A small body size is the characteristic feature that overrides all other adaptations of a meiobenthic animal. It is indispensable when squeezing through the narrow crevices between sand grains or pushing through thickets of debris and mud. Distinctive adaptive features have evolved in sand meiofauna (interstitial fauna). Filiform or flattened bodies, often equipped with long tails, enable flexibility and contact with sediment particles, facilitating movement through the void system. Locomotion in sand is often achieved by body ciliation or by wriggling movements, while burrowing occurs mostly in soft muds and climbing in phytal habitats. When erosive disturbances occur in the mobile substrate, adhesion reduces the risk of being washed away. This is accomplished by utilizing complex glands on toes or spines that produce sticky mucus. Reinforcing external or internal body structures (armor, spicules, alveolar cushions) are conspicuous features of many meiofauna and are believed to physically protect the body in mobile habitats. Static organs frequently help animals to orientate in the dark, three-dimensional system where eyes and pigments are useless. The minute sizes of meiobenthos as well as their mazelike habitats have prompted parsimonious modes of reproduction: the production of only a few eggs, direct sperm transfer or internal fertilization, brood protection, abbreviated larval life, and restricted propagation complete the array of integrating adaptations common to many meiofauna.
Chapter 5
Meiofauna Taxa: A Systematic Account
Many of the zoological groups belonging to the meiobenthos are omitted in zoology textbooks, or they are commented upon as “small and isolated groups” in a few lines of small print. However, they often represent anatomically fascinating and phylogenetically important taxa. Particularly in the “new animal phylogeny” based on the analysis of gene sequences or whole genomes, small meiobenthic groups such as rotifers, gnathostomulids or tardigrades represent important and much disputed bridging taxa, whose links with larger phylogenetic groups are much disputed (see Jenner 2004; Philippe et al. 2005). In this account, the reader will find freshwater taxa more briefly presented than the marine groups (see Rundle et al. 2002 for more details on the freshwater taxa). In this ecologically oriented overview, phylogenetic discussions will be presented in general features only. My descriptions of meiobenthic taxa (including gross diagnostic aspects and omitting many anatomical details) cannot substitute for reading textbooks or the primary literature. The figures given herein depict only some selected forms and should not be used for more detailed identification. Each paragraph begins with taxonomic data, continues with biological, ecological and distributional comments, and ends with some aspects of relationships and phylogeny.
5.1
Protista (Protoctista)
Though often neglected by zoologists, the sizes and ecological relevances of many unicellular benthic heterotrophs mean that they merit consideration in meiobenthological studies. Since the natural classification of protists is undergoing a great deal of change, the traditional names and groupings are retained here.
5.1.1
Foraminifera (Rhizaria: Granuloreticulosa)
Most foraminiferans belong within the size range of the meiobenthos, although there are numerous species with larger, often solid shells that can attain sizes of up O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
103
104
5 Meiofauna Taxa: A Systematic Account
to 12 cm or more. In total about 34,000 species have been described, and about 4,000 of these recently (Lee 1980b). In both deep-sea bottoms and in shallow reaches, a single sampling campaign can collect 50–60 foraminiferan taxa (Hannah and Rogerson 1997). Some are small symmetrical or snail-shaped, some plait- or ribbon-shaped, and some are disk-like. Others are completely asymmetrical and are hardly recognizable as foraminiferans, and so they have often been overlooked. The multitude of shell shapes is bewildering (see Fig. 5.1). The original organic substance of the shell secreted by the cytoplasm will mostly be more or less completely substituted for calcareous material. The Gromiida and Allogromiida (“atestate foraminiferans”) have delicate or agglutinated soft shells of organic material that are often not recognized to house protists. One of the more conspicuous, filiform foraminiferans that crawls through sand is Marenda (Fig. 4.3), formerly assigned to the Amoebozoa. The affiliation of these groups to the traditional Foraminifera and the derivation of more detailed interspecific phylogenetic relationships are achieved through molecular methods rather than the structural characters of the shells (for a review see Pawlowski 2000). The large species often are heterokaryotic, similar to the Ciliophora (see below), i.e., they have a macro- and
D 1 1
2
1
A
2 1
3
6
4 5
7 8
A
4 2
B
3
5
C 1
3 2 C
Fig. 5.1a–d “Morphogroups” of Foraminifera in their typical habitats (based on Jones and Charnock 1985). Morphogroup a (hyperbenthic suspension feeders): 1, Halyphysema; 2, Dendrophyra; 3,5,8, Pelosina; 4,7, Jaculella; 6, Bathysiphon; Morphogroup b (epibenthic, surface-dwelling herbivores, detritivores, omnivores): 1,2, Psammosphaera; 3, Hippocrepina; 4, Ammodiscus; 5, Saccamina; Morphogroup c (endobenthic herbivores and detritivores): 1, Ammobaculites; 2, Miliammina and Quinqueloculina; 3, Textularia; Morphogroup d (phytal herbivores): 1, Trochammina inflata; 2, T. labiosa
5.1 Protista (Protoctista)
105
a micronucleus. Many of the disk-shaped epibenthic forms live symbiotically with unicellular algae (Symbiodinium group). They cover the bottoms of shallow waters, where they utilize the sunlight. The biology and ecology of benthic foraminiferans have been recently compiled and reviewed (Sen Gupta 2002). Since their eminent role in the deep-sea benthos (see below) was discovered, foraminiferans have received increasing research attention. Common planktonic drifters are far surpassed in terms of species number by epi- and endobenthic forms in soft sediments. Here they cling to the substratum using their long, often branching pseudopodia (reticulopodia, rhizopodia, see Fig. 5.1). The nonparasitic foraminiferans have been assigned to four “morphogroups” on the basis of life position and feeding habits (see Fig. 5.1; Jones and Charnock 1985). These include the tubular and branching suspension feeders that are semisessile and anchored in an erect position, preferably in more exposed sites of the sediment (Type A), the deposit feeding or (passively) herbivorous epibenthic Type B that occur on more lentic, detritus-rich sediments, the more vagile, endobenthic sand-based detritivores and herbivores (Type C), and the phytal forms that live on phytodetritus or plants (Type D). Foraminifera move slowly by contracting their amazingly large nets of cytoplasmatic appendages that they use to collect algal cells, phytodetritus, bacteria or even small animals for ingestion. Many species are bacterivorous or even ingest small animals (Goldstein and Corliss 1994). They are considered heavy grazers on bacteria and their biofilms (Bernhard and Bowser 1992). In some cases, animals as large as 2–3 cm in size, like cumaceans, caprellids and newly metamorphosed echinoderms, have been found in their cytoplasmic nets. Apparently the prey is immobilized (toxins?) before it is rapidly digested (digestion lasts only one day, even for large prey). The uptake of nutrition seems to be a rather selective process, with some species preferring only certain benthic microalgae (e.g., diatoms), while other epibenthic species are suspension feeders. Most benthic species live in the surface layers, especially the diatom eaters. However, most of them are versatile in microhabitat selection and react to food supply, current exposure or oxygen availability. Many foraminiferans can occur down to 30 cm depth in the sediment, even in oxygen-depleted sediments, where they cluster in the micro-oxic zones around animal burrows (Thomson and Altenbach 1993). They can also live facultatively anaerobic for extended periods of time (Bernhard 1996; Moodley1997 et al. 1997, 1998a,b; Bernhard and Sen Gupta 2002), which may become increasingly relevant as oxygen-depleted areas continue to expand. Both soft-shelled and hard-shelled foraminiferans have been found under permanent anoxia in the Black Sea. Under low-oxygen organic enrichment conditions, a few opportunistic species form a low-diversity, high-abundance community, often with tests that deviate from the common patterns. In near shore, organic-rich sediments the seasonally varying dissolution of the tests due to undersaturation of calcite might influence the population dynamics and impair ecological success of foraminiferans (Green et al. 1998). Many dominant species of foraminiferans are widely distributed and often cross biogeographical barriers. A few species are adapted to brackish water conditions.
106
5 Meiofauna Taxa: A Systematic Account
The considerable ecological role of foraminiferans was not assessed until recently because of the difficulties involved in counting them and particularly in reliably separating dead and live forms (see Chap. 3). In the tidal flats of the North Sea, they usually occur at densities of about 700 individuals per 10 cm2 (Ammonia beccari, Chandler 1989) or >1,000 ind. cm−3 sediment (Ellison 1984). Their significance was further underlined by 13C-uptake experiments with foraminiferans (mainly Ammonia) that were fed with diatoms in an intertidal flat (Moodley et al. 2000). They documented the rapid incorporation of algal carbon by foraminifera and their important contribution to the meiofauna. Sandy bottoms on the continental shelf harbored more than 400 individuals per cm3, and 150–200 ind. cm−3 sediment were even found in the deep-sea bottom (Gooday 1986). Gabel (1971) counted >1,000 ind. per g sediment (dry wt) in the northern North Sea, with a decreasing trend towards the south. Although Foraminifera with their net of pseudopodia can densely cover the sea bottom, their share of the productivity and benthic turnover has rarely been quantified. They can dominate the meiofauna, yet in most studies on meiobenthos, which usually only deal with Metazoa, they are neglected. Particularly in the deep-sea (see Sect. 8.3), the overall trophic and productive processes must be reconsidered, since 50–80% of the meiofaunal species and 30% of their biomass can consist of foraminiferans (see Fig. 8.3; Shirayama and Horikoshi 1989). They make a central contribution to the cycling of organic matter in the deep-sea bottom (Larkin et al. 2006). Especially in polar regions, allogromiid and gromiid Foraminifera seem to be common taxa, but their ecological roles are poorly known (Pawlowski et al. 2005). Another group of testate protists with filose pseudopodia that are grouped within Foraminifera or related to them are the Komokiacea (Gooday et al. 2007). They occur in relevant numbers and exhibit species richness on and in deep-sea bottoms, but are widely overlooked during benthos sampling because of their irregular shapes. Small komokiacean species may fall into the meiofauna category (Fig. 5.2), while in others the complex system of branching tubes often coagulates with the sediment such that they resemble larger pebbles or mud balls (“komochki” Russian for a stone or pebble). Their relationship to the Foraminifera is still to be uncovered by molecular methods, their general ecological importance in the deep-sea is probably much higher than acknowledged so far. Another group of strange rhizopod (?) protozoans, the Xenophyophorea of the deep-sea, are sometimes mentioned in context with meiofauna and drew attention due to the accumulation of radionuclides in
Fig. 5.2 Typical komokiacean protist (Edgertonia?) from the deep sea of the North Pacific (4,430 m). (Drawing from photograph, courtesy of T. Radziejewska)
250 µm
5.1 Protista (Protoctista)
107
their cells (Swinbanks and Shirayama 1986). However, since in many species their multinuclear cells attain up to several centimeters in size, they are not detailed here. Foraminiferans may even play a key role in the production of manganese nodules, where they contribute 10% of the outer layer. Granules of manganese and iron are quite common in the cytoplasm of Foraminifera (Mullineaux 1987). Simple Foraminifera flourish even in the deepest oceanic troughs (Todo and Kitazato 2005). More detailed reading: tidal flats, Hofker (1977), Lee and Anderson (1991); North Sea, Rhumbler (1938), Gabel (1971), deep-sea, Gooday (1986), identification (North Sea), Murray (1979); reviews, Murray (1973); Lee (1980b); Sen Gupta (2002).
5.1.2
Heliozoa (Actinopodia)
The traditional taxon Heliozoa is still described in most textbooks as a planktonic group of protozoa. However, the majority of heliozoans (size range with axopodia: 0.1—1 mm) are now considered to be a component of the freshwater micro- and meiobenthos. The brackish and marine heliozoans are morphologically close to the limnetic ones (Mikryukov 2001). Especially in the warmer season, intermittent buoyant (planktonic) phases can be observed, often resulting in blooms—a good example of benthopelagic coupling. Heliozoans live as passive “contact predators,” especially in the unconsolidated surface layers of lentic sands, temporarily attached by their sticky axopodia or by stalks to detritus particles, surficial sand grains and often also to macrophytes. The heliozoan body shape is spherical with prey-capturing axopodia (pseudopodia stiffened by a central axis of microtubules) extending like sunrays from the spherical central body. Specialists consider the group polyphyletic and subdivide it into seven subtaxa with a similar appearance. 80% of all Heliozoa are assigned to the Centroheliozoa, while the other subtaxa are permanently or temporarily stalked. The number of species identified so far is low; many freshwater species seem morphologically identical to marine forms. There are around 15 in the White Sea and 27 in Australia (Mikryukov 2001). Micropredators, they feed on microphytobenthos, ciliates and rarely on small metazoans like gastrotrichs and rotifers. Their species composition seems to depend on the water depth and the presence of macrophytes. In contrast to the prevailing limnetic forms, our knowledge of the marine heliozoa is very limited and based mostly on studies from the White Sea, the Black Sea (Mikryukov 1994, 2001) and from Mediterranean shores (Golemansky 1976; Febvre-Chevalier 1985).
5.1.3
Amoebozoa (“Rhizopoda”): Gymnamoebea, Testacea
Many Amoebozoa fall within the meiobenthic size range and their shells (“testa”) exhibit considerable morphological diversity (15 morphospecies have been found in one sample). Marine species usually have a chitinous organic shell with only a few, fine particles embedded in it (see Golemansky and Todorov 2004), while more sand
108
5 Meiofauna Taxa: A Systematic Account
Fig. 5.3 A typical interstitial testate amoebae from upper marine littoral sands. (After Golemansky and Todorov 2004)
20 µm
and detritus particles are embedded in the shell in limnetic forms (e.g., Centropyxis). The psammobiotic species are usually particularly small (mostly around 50 µm), and the shell is often flattened. An anchoring device, the sac-shaped shell has a wide circular collar from which the pseudopodia extend. Some lateral or caudal spines (see Fig. 5.3) may help to reduce the risk of suspension. These characters are interpreted as adaptations to interstitial life (Golemansky 1986). About 120 benthic species in 16 genera are described (Todorov and Golemansky 2007). In the warmer season between 10 and 21 (on average) species with 70–175 individuals were encountered per cm3 of sand (maximum 32 spp with 242 ind. cm−3). In these dense populations, psammobiotic testate amoebas represent effective herbivores or passive carnivores. However, they are often disregarded in accounts of meiofauna due to inappropriate sample processing. Since they are mostly euryhaline species they occur preferably in the moist sand above the water line, but others can be encountered subtidally as well. Golemansky (1994) even reported rich populations with about 90 species of interstitial testate amoebae in the brackish shoreline (the “hygropsammal”) of the Black Sea, with some of them also occurring in suboxic sands. More detailed reading: Golemansky (1978, 1994).
5.1.4
Ciliophora (Ciliata)
These heterokaryotic protists (the genome is separated into micro- and macronucleus) are traditionally divided into Holotrichia, Heterotrichia, Hypotrichia etc. These artificial groupings have gradually been substituted for more phylogenetically natural taxa. However, many references still refer to the old classification, with the Karyorelictida being the most primitive ciliate group. However, their primitive position is in doubt (the fact that their 2 n macronucleus is not capable of division appears to be a derived feature). The identification of ciliates is initially based on the inspection of live animals. The diagnostic features of many forms, which are characteristic enough to allow relatively easy identification at the genus level (body shape, ciliation, position of mouth opening), disappear immediately after fixation. However, species determination usually requires that the nuclear apparatus and infraciliation are scrutinized via
5.1 Protista (Protoctista)
109
25 µm
staining procedures. An illustrated key has been developed for the interstitial marine forms (Carey 1992). While estimates of the total number of ciliate species vary (between 3,000 and 8,000), it is assumed that about 1,000 belong to the meiobenthos. About 40 species have been found in 1 cm3 of sandy sediment (Wickham et al. 2000). Benthic habitats harbor more ciliate species than the open water. There are often clear structural correlations between general body shape and size of the pore volume (FauréFremiet 1950). While “euporal” ciliates are apparently rather independent of grain size and occur in both fine and in coarse sediments, “microporal” ciliates populate the interstitial system of fine to medium sands (120–400 µm). Their rather thin pellicles give them high flexibility and their slender, often thread-like or flattened bodies allow them to quickly glide forward despite considerable body length (>3,000 µm). The microporal species avoid sediment agitation. The Karyorelictida are characteristic of this microporal ciliate community with the common family Trachelocercidae (Fig. 5.4). The genera Trachelonema, Tracheloraphis, Geleia, and Remanella occur frequently in the fine-to-medium coastal sands (see Patterson et al. 1989). However, convergent adaptations to an interstitial life (body elongation, flattening, head-like anterior end, formation of a tail shaft, see Sect. 4.1.1) have also developed in the noncaryorelictid genera Condylostoma, Spirostomum, Blepharisma (Heterotrichia). “Mesoporal” species live in coarse sands with pore diameters of 400–1,800 µm, where a thick and sometimes armored pellicle (plates, spines) helps them to survive strong wave exposure. The wide void system allows the small-to-medium-sized, mostly oval-to-slightly-flattened forms (often Hypotrichia) to move in the interstitial water with
Prorodon Frontonia
20 µm
Euplotes 100 µm
100 µm
50 µm
50 µm
Blepharisma
Kentrophorus
Tracheloraphis
50 µm
Geleia
50 µm
Condylostoma
300 µm
100 µm
Remanella
Spirostomum filum
Fig. 5.4 Some benthic representatives of characteristic Ciliata. (Various authors)
110
5 Meiofauna Taxa: A Systematic Account
sudden jerks of their ventral cirri. They are also able to cling to the grains by adhesive cirri. These mostly hypotrichous ciliates are frequently represented by the genera Aspidisca, Frontonia, Diophrys, Prorodon, Euplotes, Coleps, etc. (Figs. 5.4, 5.5). In sandy sediments several improved enumeration methods (sonication, centrifugation after flotation with Percoll) have been developed which demonstrate that the abundance of ciliates has been greatly underestimated (Epstein 1995; Wickham et al. 2000). Most ciliates occur in medium to fine sands where they can numerically exceed all other animals, even in biomass: tens to hundreds are observed per ml sand (with 1–20 µg fresh weight) or 20 ind. per g sediment (dry weight). With modern methods and under favorable conditions, as many as 2,300–2,500 marine ciliate cells per ml of sand have been recorded in the upper sediment layers (McLachlan 1985; Wickham et al. 2000). Fenchel (1969, 1978) found up to 5,000 ind. under 1 cm2. Figures from Australian sites were clearly less: 12 ind. under 1 cm2 in mangrove mud, 240 ind. under 1 cm2 in fine beach sand, 100 ind. per ml coral reef sand (Moriarty et al. 1985; Alongi 1986). The highest abundance is usually found in sediments with <1 wt% organic substance (Sich 1990), while silty sediments whose voids become clogged by rich organic matter are avoided by most ciliates. In brackish water sediments, up to 750 ind. with 27 mg biomass per ml were reported by Dietrich (1999). Freshwater
300 µm
Fig. 5.5 A typical microhabitat of ciliates in microbial mats: various ciliates—Tracheloraphis, Frontonia, Diophrys, Trochiloides; cyanobacteria, euglenoid flagellates, one nematode. (Fenchel 1969)
5.1 Protista (Protoctista)
111
Depth (cm)
sediments were populated by up to 11,000 benthic ciliates per 1 cm3 (Gasol 1993). In contrast, in exposed coarse sands and in sediments of <100 µm particle size, ciliates decrease rapidly in abundance, but in calm periods these differences may disappear (Lucchesi and Santangelo 1997). In lotic sands, species with a thick pellicle prevail, often armored with fibrils (e.g., Aspidisca, Coleps). The interstitial of sea ice is populated by species that crawl on their cilia (Agatha et al. 1993, see Sect. 8.1.1). As with other meiofauna, the microdistribution of ciliates is patchy (see Patterson et al. 1989). Studies by Santangelo and Lucchesi (1995) suggested a positive correlation between the occurrence of benthic flagellates and ciliates, suggesting that both groups respond to the same environmental factors. Microbial mats are a favorite habitat of many ciliates, as depicted in Fig. 5.5. In general, the herbivorous and omnivorous ciliates live in more superficial horizons, while the bacterivores are encountered deeper down (Fig. 5.6). In winter, many ciliates are found deeper down in the sediment than in summer (e.g., Coleps, Condylostoma). All trophic types are represented among ciliates. Although most of them are bacterivores, there are also detritivores—often even selective algivorous grazers (e.g., Loxodes and Remanella spp.) feeding on diatoms of selected size (Fig. 5.7), histophagous scavengers, and up to about 10% carnivorous predators (Fenchel 1968a, 1969). Sich (1990) grouped the ciliate fauna from the brackish
METAZOA (TOTAL)..................................
Metopus vestitus .....
CILIATA (TOTAL)
Remanella margaritifera............ ....... Plagiopogon loricatus Parablepharisma............ chlamydophorum Urostrongylum caudatum.................. Mesodinium pupula......................
T.kahli.................................................
Sondaria vorax..................
Tracheloraphis sp...............
Diophrys scutum..............................................
Strombidium sauerbrayae...
Aspidisca sp......................
D
Myelostoma bipartitum....... Caenomorpha levandri
C
Parablepharisma pellitum........................
B
Metopus contortus............
A
Sonderia cyclostoma..........
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Eh
−100 0 100 300 mV
Fig. 5.6 Vertical distribution of ciliates in a Scandinavian beach in relation to the redox potential. Broken line indicates position of redox potential discontinuity. Sediment layers A–D stand for: (A) light, oxic surface sand; (B) grayish sand; (C) decomposing Zostera; (D) blackish anoxic sand. (After Fenchel 1969)
112
5 Meiofauna Taxa: A Systematic Account
% 50 Remanella margaretifera
25
0
10
20
30 40
50
60
70 µm
% 50
Remanella rugosa
25
0
10
20 30
40
50 60 70 µm
% Remanella brunnea 25
0
10
20
30 40
50 60 70 µm
% 50 Remanella gigas
25
0
10
20 30
40
50
60 70 µm
Fig. 5.7 Selective feeding of different Remanella species (Ciliata) on diatoms of different sizes. (After Fenchel 1968)
5.1 Protista (Protoctista)
113
Kiel Bight (Baltic Sea) into bacterivores (50%), herbivores (30%), carnivores (1%) and omnivores (15%). He found experimentally that only bacteria (10–15%) which swim freely in the pore water are ingested by ciliates, while those remaining on the sand grains are not utilized. Most ciliates are found in the oxidized upper two centimeters of the sediment and cell numbers usually decrease with depth. However, many bacterivorous species tolerate periodically hypoxic conditions and aggregate around or below the RPD layer (Geleia nigriceps, Peritromus, Cardiostomatella, Paraspathidium, Remanella sp.), where they feed on the rich sulfur bacteria (Fig. 5.6). Berninger and Epstein (1995) even contend that the majority of the ciliates in North Sea intertidal sediments live in the anoxic realm, migrating up and down. The capacity to live even under permanently anoxic conditions mostly coincides with the presence of “hydrogenosomes” instead of mitochondria (Fenchel and Finlay 1991, Hackstein 1999). Many of the more specialized ciliates like Uronychia transfuga, Kentrophoros spp. harbor bacterial endosymbionts (methanogens; Fenchel and Finlay 1989; Van Hoek et al. 2000) which enable them to live in sulfidic and methanic sediment (see Sect. 8.4). The annual population growth of ciliates closely follows the cycles of temperature and food supply, with maxima in spring–summer. In fall, food availability seems to determine population sizes, which decrease rapidly towards autumn–winter. In boreal latitudes, minimal populations are found in February and March. Generation time (sexual reproduction by conjugation) is variable, usually about ten days, but ciliates can multiply asexually by division in just a few hours (Sich 1990). Annual P/B values (see Sect. 9.3.2) of about 250 are not unusual for ciliates (compare with 10–20 for metazoan meiobenthos and 2–3 for macrofauna!). It is this rapid turnover which underlines the productive relevance of ciliates even when their standing stock remains relatively small, particularly since many carnivorous ciliates feed intensively on their bacterivorous relatives. The ecological relevance of bacterivorous ciliates has been documented in experiments on the degradation of algal debris from wrack (Rieper-Kirchner 1989). The addition of ciliates to the experiments caused a considerable increase in bacterial activity, resulting in a faster degradation of plant remnants (see also Alkemade et al. 1992). The microbial populations apparently remain in their exponential growth phases through ciliate grazing. However, detritivorous ciliates also contribute directly to the mineralization of organic matter through the regeneration of nutrients. For pore water bacteria, Sich (1990) found that ciliates reduce the annual bacterial production by about 50%, so that Epstein (1997a) hypothesized that the bacterial stock might be size-controlled by the grazing of ciliates (together with nematodes). However, considering the overwhelming numerical dominance and production of total bacteria, a direct influence is certainly not exerted by ciliates on the overall bacterial community (Epstein and Shiaris 1992; Kemp 1990; for freshwater conditions see Gasol 1993). Regarding the high metabolic rates of ciliates, the contention arose that the bulk of ciliate production represents an energetic dead end that was not transferred to higher trophic levels and was often ecologically overestimated (Kemp 1988). Early observations noted that rotifers and turbellarians ingest ciliates (Fenchel 1969).
114
5 Meiofauna Taxa: A Systematic Account
Species- and size-specific interactions between meiofauna and various ciliates in a sandy tidal flat were also documented experimentally (Epstein and Gallagher 1992). Ciliate-feeding nematodes in fine shallow sands consumed a considerable proportion of the (larger) ciliate production (Moens and Vincx 1997; Hamels et al. 2002). On the other hand, large karyorelictid ciliates have been found to devour meiofauna such as small nematodes and cladocerans. These recent data characterize the ecological role of ciliates in the food web as mediating between bacteria (their food) and metazoa (their predators). However, even in fine-to-medium sands, where they can become particularly important, they are not likely to restrict bacterial production. More detailed reading: ecology, Fauré-Fremiet (1950); Dragesco 1960; Fenchel (1968a,b, 1969); Hartwig (1973b); polar regions, Scott and Marchant (2005); interstitial ciliates, Dragesco (1960); Hartwig (1973a); systematic and phylogenetic aspects, Corliss (1974, 1975, 1979); illustrated key, Carey (1992); productive role, Epstein (1997a,b); reviews on marine ciliates, Fenchel (1978); Patterson et al. (1989); freshwater, Gasol (1993); Patterson (1996).
Box 5.1 Protozoan Meiobenthos: A Link Between Bacteria and Metazoa Larger protozoans, well within the size range of meiobenthic fauna, are often disregarded because their sampling and study require specific methods. While Ciliata, with their typical structural adaptations to the sediment pores, are considered the “classical” meiobenthic representatives, taxa such as benthic Foraminifera, Heliozoa and Amoebozoa are known predominantly from studies of particular biotopes: foraminiferans in the deep sea, heliozoans mostly in the freshwater phytal, and amoebozoans in the moist upper shores. Here these groups are often speciose, abundant and of ecological relevance. Now, with improved methods, and a better eye for these often inconspicuously formed protists, we have evidence that they are probably ubiquitous, present in great abundance, and have fast turnover rates and high productivity. Foraminifera can dominate sediment life in the deep sea; ciliates can become so numerous that some authors consider that they control bacterial production. On the other hand, they are an important food for many nematodes. The study of heliozoans is just beginning and yet, as so often happens in meiobenthology, it depends on the activities of just a few specialists.
5.2
Cnidaria
Representatives from most cnidarian classes have been found in the meiobenthos, although their general organisation and size do not seem to make them particularly preadapted to interstitial life. Since almost all polyps of cnidarians can detach from
5.2 Cnidaria
115
their substrata and slowly change their positions, it is often arbitrary as to whether we should regard small polyps as true meiobenthos or just as temporary meiofauna, or even to omit them in a treatise on meiobenthos altogether due to their prevalent sessility. In any case, there are interstitial forms that are regularly found in sandy bottoms, and they account for about 20 species of hydrozoans, and a few meiobenthic scyphozoans and anthozoans. The first report of a “classical” meiobenthic medusa described the tiny (only 300–1,200 µm) hydromedusa Halammohydra from the littoral sands of the North Sea (Remane 1927). At the time it was a small zoological sensation (Fig. 5.8), and its lasting reputation is reflected in the emblem of the International Association of Meiobenthologists. The systematic position and the metagenetic characterization of meiobenthic medusae have often been disputed and vary among specialists because of the frequent reduction in diagnostic features. Progenesis and reduction of the medusoid
500 µm
500 µm
Halammodydra schulzei
H. octopodides H. vermiformis
Fig. 5.8 The well-known interstitial genus Halammohydra (Hydrozoa). (Various authors)
116
5 Meiofauna Taxa: A Systematic Account
stage often blur the situation. Today, a combination of morphological characters with features of the cnidome is used as the basis for assigning taxonomical affinities. A selection of some better known as well as some recently discovered meiobenthic cnidarians will be presented here.
5.2.1
Hydroida (Medusae)
Halammohydra has an ovoid-to-slender body sack that is reminiscent of a polyp, with 1–2 rows of tentacles bearing large cnidocysts. The umbrella is completely reduced, although the statocysts, in normal medusae located peripherally at the umbrella, remain well developed near the basis of the tentacles. At the aboral end there is a groove serving as an adhesive organ. The whole body is strongly ciliated, which enables Halammohydra to swim gently for a short while. However, its usual movement is climbing between sand grains. The genus is hermaphroditic. Today, Halammohydra is considered a much derived if not neotenic hydromedusa (Fig. 5.8). The genus is known to occur worldwide, with about ten species identified. In favorable habitats Halammohydra can be quite common. The species vary slightly in the shapes of their body sacks, their movements and their preferred biotopes. The best-known Halammohydra schulzei is clumsy, strongly haptosessile, and slow-moving. It is found in sublittoral coarse sand. Halammohydra octopodides from sublittoral fine sand is more slender and active, a trend further developed in H. vermiformis, which populates the upper eulittoral of beaches. Otohydra vagans and O. tremulans (Fig. 5.9) bear only one whirl of tentacles and are only 350 µm in length. Lacking an adhesive groove, the animals are often found swimming with their body ciliations in the interstitial voids. They are hermaphroditic and viviparous, brooding their Actinula-like larvae in their wide body sacks. Although their appearance is rather polypoid, they are considered to be apomorphic medusae grouped with the Halammohydrida. Armorhydra janowiczi, discovered together with Otohydra at Roscoff, France (Swedmark and Teissier 1958), has more medusoid features, with its small umbrella and a tiny velum (Fig. 5.9). It lacks statocysts and has two kinds of tentacles, filiform ones with cnidocysts and capitate ones with adhesive desmonemes. Based on the set of various cnidocyst types, the cnidome, Armorhydra was taxonomically designated a representative of the Limnohydrina, whose corresponding polyp has also been found.
5.2.2
Hydroida (Polyps)
The best-known representative of these hemisessile or haptosessile meiobenthic forms only a few mm in length is Protohydra leuckarti. This has no tentacles and cnidocysts cover the whole body (Fig. 5.9). Gonads are rarely developed, asexual
200 µm
500 µm Otohydra
Fig. 5.9 Some meiobenthic hydroids. (Various authors)
Meiorhopalon arenicolum
1 mm Armorhydra
Psammohydra
1 mm Euphysa
Protohydra
Pinushydra
5.2 Cnidaria 117
100 µm
1 mm 100 µm
118
5 Meiofauna Taxa: A Systematic Account
multiplication is by transverse division. Discovered in Belgian brackish waters, it occurs worldwide in muddy, rarely sandy and often brackish sediments where it preys intensively on a wide food spectrum (oligochaetes, insect larvae, nematodes and harpacticoids). Hydrozoan polyps assumed to belong to protohydrids have even been found in the void system of Arctic coastal ice (Bluhm et al. 2007), where they probably live on the rich “sympagic” meiofauna. Psammohydra nanna from a sandy bottom in Kiel Bight (Baltic Sea) was described as an interstitial polyp with four short tentacles (Schulz 1950). Today it is also known to occur in the North Sea and the Atlantic and Mediterranean oceans. Gonads have never been observed; only asexual reproduction by division is reported. Euphysa ruthae (Norenburg and Morse 1983) is an exceptional interstitial polyp through its mode of reproduction: budding with subsequent polarity reversal of the buds. Cryptohydra thieli, a polyp with an aboral adhesive button, lives in the sublittoral shell hash interstitial east of the Florida coast (Thomas et al. 1995); it seems to be related to the mud-dwelling Acaulidae. Pinushydra chiquitita and Nannocoryne mammylia (Bouillon and Grohmann 1990, 1994) are hydropolyps from interstitial sands of Brazilian beaches. Pinushydra is remarkable through its basal ring of statocysts, sessile gonophores and an adhesive disc at its base. Meiorhopalon was found adhering with filamentous mucous sheaths to coarse sand grains in the Plymouth Sound in the UK. Its four bulges at the basal hydranth are considered static organs (Salvini-Plawen 1987).
5.2.3
Scyphozoa
Stylocoronella riedli, described by Salvini-Plawen (1966) from Rovinj (Adriatic Sea), was probably seen first in 1962 by Monniot in the western Mediterranean (Banyuls-sur-Mer). It is a tiny interstitial scyphopolyp that has barely changed its body organization (Fig. 5.10a): the mouth cone is expanded into a proboscis and surrounded by 24 tentacles with ocelli at their base. Sexual reproduction is unknown; the species multiplies by budding planuloids that develop directly into young polyps. Strobilation has not been observed.
5.2.4
Anthozoa
The strange stony coral Sphenotrochus (Rossi 1961) (Madreporaria, Caryophyllidae) has been found in coarse sand of the Mediterranean and Atlantic oceans, where it moves around using its numerous (up to 24) tentacles (Fig. 5.10b). Its maximum size (about 10 mm) certainly exceeds the range of the meiobenthos, however, at a length of 1–2 mm (in the range of many meiobenthos), it starts to divide asexually. During fission it develops (unlike its sessile relatives) bipolar whirls of tentacles (as well as calcareous septa) before the separation of the resulting individuals.
a
119
250 µm
100 µm
5.3 Free-Living Platyhelminthes: Turbellarians
Stylocoronella
b
Sphenotrochus
Fig. 5.10 a–b Interstitial representatives of Scyphomedusae (a) and Anthozoa (b). (a SalviniPlawen 1966; b Rossi 1961)
Box 5.2 Medusae in the Interstitial: A Zoological Novelty Limited in species number and abundance, meiobenthic Cnidaria were considered a sensation when the first interstitial medusae of the genus Halammohydra were discovered in coarse sand and described by Remane (1927). Since then, isolated meiobenthic representatives of all cnidarian classes have been found, some of them exhibiting typical adaptations for meiobenthic life, like ciliation or static organs. In the polyp generation, it is difficult to delineate the more mobile meiobenthic forms from those species that are mostly attached.
5.3
Free-Living Platyhelminthes: Turbellarians
Turbellarians, which are among the most primitive Metazoa, have been important research objects for over 100 years (Rieger 1998). The basal phylogenetic position of the Platyhelminthes within the spiralian Eubilateria has been supported by molecular biological methods (Riutort et al. 1993). The “Turbellaria” has been recognized as a paraphyletic taxon within the free-living Platyhelminthes, and it is not based on the principles of phylogenetic systematics (Ehlers 1985). However, the term “Turbellaria,” which is well established in the scientific literature and
120
5 Meiofauna Taxa: A Systematic Account
textbooks, will be maintained here, as it is colloquial and ecologically well defined (compare Rieger 1998). The Acoela (+ Nemertodermatida) seem to be an early, monophyletic branch, distant from other turbellarians, if not bilaterian metazoans (Jondelius et al. 2002). In other traditional families (see below) monophyly is not always supported by molecular methods (Rohde et al. 1993; Litvaitis et al. 1994). The majority of the speciose group of flatworms is small-sized. Most meiobenthic turbellarians are oval-to-round in cross-section; only the large species are flat. Their vermiform bodies, extreme flexibility, ciliation, hermaphroditic genital organs and internal fertilization by copulation render them a classical meiobenthic group, preadapted for an interstitial life. As in ciliates, identification of the softbodied animals mostly requires live specimens. To ascertain further details, adequate preservation, sectioning and staining is required. In many studies this tedious procedure has contributed to a certain neglect of the group, although in meiofaunal samples turbellarians often rank third in abundance after nematodes and harpacticoids. At least the identification of orders and some characteristic families is possible at least on the basis of well-recognizable features, such as shape of the pharynx, the gonads, intestine, presence of statocysts, etc. More than 3,000 meiobenthic species of turbellarians from marine and freshwater habitats have been described; most of them from the marine littoral. Also for freshwater species, regional comparisons of plathelminth faunas suggest that species richness may be an order of magnitude higher than currently estimated (Schockaert et al. 2008). Yet, our knowledge of this group is inadequate and many more new species will be discovered. For example, in a relatively small area around the island of Sylt, about 500 species have been found, and in a recent sampling campaign performed in the Mediterranean Sea, 50– 95% of the species were new to science (Delogu, pers. commun.). Based mostly on thorough studies from the North Sea coasts, this section characterizes some of the major groups, discusses some major recognizable features, and provides some additional ecological remarks (Fig. 5.11).
5.3.1
Major Turbellarian Groups
Acoela: This order, the basis for platyhelminths (Litvaitis et al. 1994) or even bilaterian metazoans (Katayama et al. 1993; Jondelius et al. 2002), contains about 250 species that are small and mostly roundish. A remarkable concordance between their morphological and molecular classification has been assessed by Hooge and Tyler (2006). They are characterized by an often inconspicuous mouth opening in the epidermis and the absence of a true intestine. Instead, the intestinal tissue is represented by a solid “digestive parenchyma.” Usually a statocyst is visible near the anterior end. In this group the ovary is not differentiated as discrete ovarian and vitelline tissue; the mature oocytes are in an anterior position. The diagnostically important male gonocytes are in a posterior position. Acoel turbellarians occur preferably in marine (fine) sand rich in detritus, and are also frequently found in polar sea ice. Many feed on diatoms, and some species
Macrostomida
Typhloplanoida
Kalyptorhynchia
Seriata
Otoplanidae Coelogynoporidae Tricladida
Fig. 5.11 Representatives of various meiobenthic turbellarian groups. (Various authors)
Acoela
Detail of jaw apparatus
5.3 Free-Living Platyhelminthes: Turbellarians 121
122
5 Meiofauna Taxa: A Systematic Account
are specialized to live in the sulfide biome. Significant genera: Haplogonaria, Pseudaphanostoma, Oligofilomorpha; Convoluta (this genus depends on the photosynthesis of numerous microalgal endosymbionts, Douglas 1988). Macrostomida: Both marine and limnetic species occur. They have a pharynx simplex, a simple, unspecialized pharynx, in the anterior part of their body. Like in the Acoela, the ovaries are uniform and not differentiated into various organs. Some species are characterized by numerous internal spicules (Fig. 4.9), while a few have “clepto-cnidocysts;” functional cnidocystes (originating from interstitial cnidarians?) used as defence weapons in their dermis. Numerous Macrostomida have eyes, but there are no statocysts. The mostly limnetic Microstomidae commonly exhibit asexual reproduction by fission. The resulting chains of zooids eventually separate into single individuals. Rhabdocoela(= Neorhabdocoela): In this order, the female gonads are differentiated into a germarium and a vitellarium, and statocysts are not developed. Sub-Ord. “Dalyelloida”: Probably not a natural systematic taxon, it contains mainly freshwater forms characterized by a barrel-shaped pharynx (pharynx doliiformis) with an anterior opening. Many of them have eyes and associated symbiotic zoochlorellae in their tissues. Dalyelloids with a small mouth often feed on diatoms; those with a wide mouth are predacious. Sub-Ord. “Typhloplanoida”: An artificial group of marine and freshwater species (e.g., Mesostoma) characterized by a pharynx rosulatus. Their male gonads are equipped with posterior “stylets;” the vitellarium and germarium form longitudinal tubes. Sub-Ord. Kalyptorhynchia: In analogy to the Nemertinea (see Sect. 5.5), the more than 100 marine species have an eversible proboscis, independent of mouth and pharynx, at their anterior end. This proboscis is a suction apparatus or is armed with adhesive or toxic glands, chitinous teeth or elaborate grasping hooks or “jaws” (see Fig. 5.11). In contrast to Gnathostomulida (see Sect. 5.4.1), these jaws always occur at the front end of the animals, at a certain distance from the mouth opening. Numerous kalyptorhynchs feed on harpacticoids. Once grasped, the prey is then conveyed, bending over, to the posterior mouth opening. Seriata: The germarium and vitellarium are follicular in shape; the pharynx is tubular and highly folded (Pharynx plicatus), and its opening is directed posteriorly. Sub-Ord. Proseriata: The numerous species of this marine group are mostly adapted to the interstitial system. Some of them are rather slender, and their bodies very long (Otoplanidae are often 10 mm long!). The pharynx opening is located in the posterior part of the body. The gonads are formed as longitudinal, diffuse follicles. There are various families of proseriates, and only some conspicuous ones are mentioned here. Otoplanidae: In these typical interstitial turbellarians, the head is set slightly off and equipped with long sensory cilia. It contains a refractory statocyst (plus an accessory statocyst) and the brain and posterior adhesive cells are conspicuous. The pharynx is short and lies horizontally. Highly effective ventral ciliation allows the animal to “rush” through the sand as if swimming. Characteristic genera are Bulbotoplana, Parotoplana, Otoplanella.
5.3 Free-Living Platyhelminthes: Turbellarians
123
Coelogynoporidae: These interstitial turbellarians are also very long and slender and have a marked “head,” but, in contrast to the otoplanids, the body ciliation is uniform, not just ventral. Posterior to the statocyst the brain is visible in a conspicuous capsule. The pharynx is long and directed vertically. Coelogynoporids appear to prefer coarse sand. Nematoplanidae: As indicated by the scientific name, these turbellarians are filiform, long and flexible. As an adaptation to their rigid interstitial habitat, they often have a vacuolar and turgescent intestinal tissue or parenchyma (see Sect. 4.1.2; Fig. 4.9). Nematoplanid turbellarians have no statocyst. Sub-Ord. Tricladida: These platyhelminths have a clearly tripartite intestine. While the bulk of the taxon—the limnetic “Tricladida paludicola”—represents a monophyletic unit (Riutort et al. 1993), the status of the few marine triclads (“Tricladida maricola”) has not yet been ascertained. Procerodes ulvae (Fig. 5.11) is common in boreal coasts under pebbles and in sand. Some triclad genera (e.g., Atrioplanaria) live exclusively in subterranean habitats; they are stygobiotic (see Sect. 8.7.2). Sub-Ord. Polycladida: Within this group of mostly large flatworms, some few representatives of the Theamatidae live interstitially in marine sand (CuriniGalletti et al. 2008).
5.3.2
Distributional and Ecological Aspects
The well-developed diversification of turbellarian groups permits the allocation of the major taxa to certain biotopes, resulting in a characteristic distribution pattern. Acoela (particularly the smaller forms), Macrostomida and most Neorhabdocoela prefer fine sand and mud. Kalyptorhynchia dominate in the sheltered tidal flats; when they occur higher up the beach they are restricted to the groundwater zone. In coarser sands, Proseriata prevail in the lower beach slope, while the Otoplanidae are found subtidally and in the agitated swash zone. In the brine channel systems of polar sea ice, acoel turbellarians feeding on diatoms represent the dominant group of the “sympagic” meiofauna (Gradinger et al. 1999; see Sect. 8.1.1). For notes on the biology and ecology of freshwater turbellarians see Kolasa (2000). The highest species numbers of turbellarians have been reported from fine sands and tidal flats, where their richness can match the entire macrofauna. Although species richness can vary depending on the tidal currents, it has been estimated to be as high as or even higher than that of nematodes and twice that of harpacticoids. In one North Sea tidal flat 83 species occurred with as many as 20 congeneric species per 100 m2. Small habitat structures like Arenicola tubes represent “centers of population development” with no less than five different spatial niches and numerous co-occurring turbellarian species (Reise 1984, 1987, 1988). In an Australian mangrove environment Turbellaria made up 60% of all meiofauna and represented the most important taxon (Alongi 1987a). In most biotopes, the bulk of the species are represented by the Kalyptorhynchia, followed by the Seriata. The Proseriata often dominate in individual abundance.
124
5 Meiofauna Taxa: A Systematic Account
Table 5.1 Turbellarian species number and density per 10 cm2 (brackets) in various habitats in the tidal zone around the Island of Sylt, North Sea (based on data from Reise 1988) Salt marshes
Beaches
Tidal flats
Subtidal
All habitats combined
High 102 (120) Low 104 (138) Creeks 116 (200)
Exposed 71 (202) Semi-exposed 136 (582) Sheltered 154 (293)
Sand 159 (339) Silty fine sand 142 (133) Mud 63 (135)
Sand 187 (80)
Total number of species 435 Average abundance per 10 cm2 252
In rigid environments like exposed sands and suboxic muds, the species number decreases but still remains relatively high (Reise 1988). The same holds true for the sublittoral, so the typical decrease in species number from the more stable sublittoral to the harsh supralittoral is not found in turbellarians (Table 5.1). However, they do become rare towards the deep-sea and are scarcely recorded in bottoms beyond depths of 400 m. In part, the relative scarcity of records of their presence in deep bottoms could be due to sampling/preserving methods that inadequately preserve the delicate body integrity of turbellarians. Turbellarians not only have a rich species number but they can also develop high abundances (Table 5.1). Ólafsson (1991) reported up to 700 individuals per 10 cm2 from Icelandic sandy beaches, numbers that dominated over all other meiofauna. On the Belgian coast, the presence of between 100 and 500 individuals per 10 cm2 resulted in a biomass that locally exceeded that of nematodes (Martens and Schockaert 1986); these figures were confirmed by Reise (1984) for a German tidal flat. In one m2 of Arctic sea ice lived 5,500 individuals of acoel turbellarians (Gradinger et al. 1999). It seems to be a characteristic of turbellarians that they can become the dominant meiofauna group in rather extreme biotopes (e.g., salt marshes, swash zone of beaches, polar sea ice; brackish water habitats). Among the abiotic factors, it is the sediment water content which greatly determines the occurrence of the soft-bodied turbellarians. This explains their high numbers in water-saturated sublittoral flats. In the drier upper beach, they prefer the moist deeper layers near the groundwater horizon. On the other hand, most turbellarians need an oxygenated environment, which restricts them in less aerated habitats such as tidal flats to the superficial horizons and to the narrow oxic zone around macrofaunal holes. Exceptions are, however, the acoel Solenofilomorphidae and the catenulid Retronectidae, as well as some other turbellarian groups that contain typical thiobiotic species (see Sect. 8.4.2). Exposure to water currents and wave action as well as general physical instability do not seem to adversely affect turbellarian settlement (Fegley 1987; Hellwig-Armonies 1988), which would explain their frequent emergence and hyberbenthic occurrence. Both in the field and in experiments (Armonies 1989a,b, 1990) a considerable number of turbellarian species, particularly the pigmented ones, regularly enter the overlying water column, and so they can colonize, after being distributed by water currents, new areas in a short while (sometimes just a matter of hours). This “hyperbenthic” behavior of many turbellarians is preferably nocturnal and may contribute to the wide geographic ranges of turbellarian species (Ax and Armonies 1990).
5.3 Free-Living Platyhelminthes: Turbellarians
125
Salinity seems to be a subordinate factor for most turbellarians: almost all species tolerate both brackish and fully saline conditions. The species occurring in the brackish Baltic Sea are also represented in the more haline North Sea (Armonies 1988d). The brackish water species are widespread and most of them are amphi-North Atlantic in distribution; many even occur along the Alaskan coast. This would corroborate the hypothesis of a boreal–subarctic brackish water community (Ax and Armonies 1990). Among the 340 euryhaline species of turbellarians living in brackish waters, 70 are classified as genuine brackwater species (Ax 2008). Life in the channel system of sea ice, a habitat preferred by many Turbellaria (Gradinger et al. 1999), also requires extreme tolerance to fluctuating salinities from brine to almost freshwater conditions. Armonies (1988c) showed that although salt marsh species of turbellarians preferred distinct ranges of mostly brackish salinity, they could endure adverse conditions, some of them in protective cysts or in dormancy. A rich food supply is the other determinant of the occurrence of voracious that about one third of the turbellarians. In a North Sea tidal flat, Reise (1988) found that almost 65% of all species (mostly the larger ones) were predators of other meiofauna, e.g., nematodes (Ott 1972a). Using a freezing technique, Kennedy (1994b) visualized turbellarian attacks on nematodes. Particularly in sandy habitats, turbellarians can also exert a considerable predatory pressure on the smaller macrofauna (Watzin 1985, 1986). In experiments, a few turbellarian species from North Sea tidal flats consumed 20% of their prey copepod population plus additional ostracod prey (Menn and Armonies 1999). In both freshwater and marine habitats, platyhelminth predation is believed to strongly influence the population dynamics of meiofaunal prey species. Hence, a high abundance of small fauna is a prerequisite for the presence of turbellarians. Selective predation has been observed in many turbellarian species (Watzin 1985, 1986). Otoplanids and coelogynoporids prefer oligochaetes; their absence in beaches of New Zealand was related to the scarcity of oligochaetes (Riser 1984). Also, Martens and Schockaert (1986) think the overwhelming majority of turbellarians are predacious, leaving hardly any bacterivorous species. Other authors contend that about one third of the turbellarians are diatom eaters, and about 10% prefer small flagellates and bacteria. As much as turbellarians are directly linked to a rich supply of suitable prey, they themselves seem to be rarely utilized as food, despite their abundance. Small syllid polychaetes and juvenile shrimps (Crangon) are among the few predators known to feed on them (Watzin 1986; Reise 1979). It is unclear whether the presence of defensive rhabdites in the dermis of many turbellarians or simply poor preservation in the predator’s gut contributes to the lack of reports on their trophic utilization. The differentiated distribution pattern of turbellarians is made even more complex by marked ontogenetic population dynamics with strong numerical fluctuations. About 4–5 annual generations, a restricted reproductive period, and a developmental time of 20–30 days are average values. However, many Acoela are “polyvoltine,” reproducing throughout the year with ten annual generations. Other turbellarians are bivoltine or just univoltine. Population dynamics are further influenced by seasonal and irregular temperature variations. Maximal abundance is usually in spring and fall (Acoela, Macrostomida); the extremes of summer and winter are apparently less favorable, although for the
126
5 Meiofauna Taxa: A Systematic Account
Table 5.2 Percent compositions of turbellarian fauna at the same sandy site (Island of Sylt, Germany) Sopott (1973)
Faubel (1976a)
Reise (1984)
Turbellarian taxa
Sandy beach
Sandy beach
Sandy tidal flat
Acoela Macrostomida Proseriata Kalyptorhynchia Typhloplanoida Dalyelloida
26.3 10.4 41.0
61.8 5.8 22.0 5.6 4.6
21.7 8.3 22.3 26.1 19.5
}
22.3
turbellarian inhabitants of supralittoral shores and salt marshes, population peaks have been recorded in the cold season (Faubel 1976b). The variability in the compositions of turbellarian groups, evident in different studies from comparable biotopes or even from the same area, may reflect the intricate local trophic and abiotic links (Table 5.2). In view of the dispersal that emergence and transport processes bring about, it is amazing to still find turbellarians in a differentiated biotopical and regional pattern (Reise 1988; Jouk et al. 1988). More detailed reading: ecology, Reise (1984, 1988); Martens and Schockaert (1986); Ax 2008; phylogenetic aspects, Ehlers (1985); Riutort et al. (1993); Jenner (2004); ultrastructure, Rieger (1981); Peter et al. (2001); illustrated key to relevant genera, Cannon (1986); review, Rieger (1998).
Box 5.3 Turbellarians: Primitive Yet Complex; Neglected Yet Important Although they belong to the most primitive bilaterian groups, turbellarians show a bewildering variety of complex anatomical features, especially in their hermaphroditic genital organs. A variable, nonterminal position of the mouth, complex sensory organs, effective and usually predacious feeding structures, often with elaborate mouth parts; all of these features are packed into a soft, extremely flexible parenchyma and are combined with intricate behavior, making turbellarians characteristic members of the meiobenthos. With their ciliated epidermis they can glide through narrow voids while hunting for nematodes or breaking up diatom shells. Due to their high abundance and grazing efficiency and their moderate utilization as prey, turbellarians play a key role in the meiobenthic food web. Without a detailed knowledge of their ecological niche, it remains uncertain as to why, in small areas of seemingly uniform structure (lower tidal flats, seagrass beds, brine channels in sea ice), so many species can coexist. Identification after bulk fixation is usually possible only to the family level. Since assignment to the species level requires tedious and special sample processing, sectioning and staining, turbellarians are often neglected in evaluations of meiobenthos. This can then lead to incomplete species evaluations, but more importantly, to a much-distorted ecological picture of food webs in which these important and often top predators go unassessed.
5.4 Gnathifera
5.4
127
Gnathifera
Recent morphological studies (Funch and Kristensen 2002; Kristensen and Funch 2000; Kristensen 2002) suggested a phylogenetic affiliation of Gnathostomulida, Rotifera and the newly discovered Micrognathozoa, which now are often combined in the taxon “Gnathifera.” The scarcity of parenchyma and details of the cuticular jaw apparatus were the major morphological criteria for this grouping, which is also supported by ultrastructural studies (Rieger and Tyler 1995; Sørensen et al. 2003). Cladistic analyses produced further evidence for the taxon “Gnathifera” (see review of Jenner 2004). However, there needs to be more conclusive molecular data to certify this grouping (Garey 2002).
5.4.1
Gnathostomulida
The Gnathostomulida are sometimes treated in the same context as the turbellarians because of the historical aspects of their discovery and early phylogenetic interpretation. It took 40 years to realize that gnathostomulids represent a separate, well-defined phylum and not just an order of turbellarians. Discovered by Remane in Kiel Bight (Baltic Sea), it was Remane’s student Ax (1956) who, after finding additional specimens from the North Sea and the Mediterranean, wrote an account of Gnathostomula paradoxa, where he considered it to belong to a primitive turbellarian group. Riedl (1969) finally classified the group as being a separate phylum on the basis of characters that had no parallels with structures known from turbellarians (see below). Today there are about 25 genera and almost 100 species known, six of them with a global distribution, but many more will probably be found with more study, particularly in thiobiotic habitats (see Sect. 8.5). All gnathostomulids are slender, vermiform, more or less cylindrical and soft-bodied animals, 500–1,000 µm in length (Fig. 5.12). In their ciliation and their gliding movements they are similar to turbellarians. However, they are characterized mainly by two unique features: (a) Their epidermal cells are uniciliated, i.e., each cell has just one, relatively long, slowly beating cilium that propels the animal in a slow, gliding movement (some species can even move backwards). (b) They have conspicuous cuticularized jaws protruding into the pharynx, which are inserted in a uniquely arranged bilateral–symmetrical musculature (Sørensen et al. 2003). A ventral mouth plate is flanked by two large jaws with a complicated pattern of teeth and grooves, providing important diagnostic features (etymological derivation of the group’s name: “animals with a small mouth armed with jaws”). The head, which is mostly studded with some stiff sensory setae, and the rear end are often somewhat set off. Like turbellarians, the intestine has an internal dead end, but some species have a periodically opening anal pore. A relationship with rotifers
128
5 Meiofauna Taxa: A Systematic Account
100 µm 100 µm
Austrognathia
100 µm
Gnathostomula
Haplognathia
Fig. 5.12 Some representative Gnathostomulida, with details of jaw apparatus; all brought to same scale (Various authors)
(Gnathifera, see below) would imply that the lack of an anus in many Gnathostomulids is a secondary, derived character. The genital apparatus is hermaphroditic, but functional males appear beside females and functional hermaphrodites. During their life spans, the gonads and accompanying copulatory organs (stylet for sperm injection; bursa, vagina) repeatedly develop fully and subsequently degenerate after the reproductive cycle has ended—a feature not yet understood in terms of its biological significance. In some species, the spermatozoa can be very large and conspicuous. In Gnathostomula, sperm bundled into spermatophores is injected into the subcuticular “bursa” of the partner (observed in cultures of Gnathostomula jenneri). Cleavage of the egg has been interpreted as following a spiral pattern. However, the parenchyma is poorly or not developed and there is practically no body cavity (acoelomate), since
5.4 Gnathifera
129
the internal organs fill the body completely. Fragmentation of the hind end occurs frequently, but asexual multiplication has not been observed. Gnathostomulids were discovered relatively recently, probably because their preferred habitats, suboxic and sulfidic fine sands, had not been investigated in much detail and they are not generally common animals. In fact, gnathostomulids can populate very narrow systems of interstices. Their unique arrangement of musculature enables them to squeeze through pores that are narrower than their diameter (Tyler and Hooge 2001). In favorable habitats (see below), when scrutinized with the right methods, one can find > 100 specimens per liter of sediment. Riedl (1969) reported > 6,000 specimens of G. jenneri per liter! Gnathostomulids are mostly associated with detritus-rich, hypoxic or slightly sulfidic fine sand like that typically found in sheltered tidal flats, seagrass beds, mangroves, and lower coral reefs. Here they feed on bacteria, fungal hyphae and perhaps diatoms, which they rasp off with their jaws. They reach maximum occurrence in the subsurface horizon near the oxic/sulfidic interface or along the tubes of burrowing macrofauna. However, they have also been encountered in anoxic layers at greater depths. In the permanently sulfidic bottom beneath brine seeps in the Gulf of Mexico they even represented the dominant animal group (Powell and Bright 1981). They were also regular inhabitants of the reduced sand core in pebble-shaped stromatolithic nodules from Bermuda (Westphalen, 1993). As Riedl (1969) put it, in their main environment, fine sediments smelling of hydrogen sulfide, gnathostomulids can “dominate all the other groups of the biotope, even the nematodes.” A remarkable feature in gnathostomulids is the frequent co-occurence of several species in virtually the same patch of sediment. Reise (1981b) found three species along a gradient of a few mm around Arenicola tubes, while Sterrer (1971) counted 13 species in one small sample. The basis of this high syntopic diversity with strong niche partitioning is not known and is contrasted by the amphiatlantic distribution of some common species. The systematic relations of the Gnathostomulida are still disputed (see above). While the jaw apparatus might superficially remind us of some kalyptorhynch turbellarians, the monociliated epidermal cells and the structure of the pharynx musculature clearly set them apart from the plathyhelminths. In any case, Gnathostomulida seems to be a group with a long, phylogenetically isolated history, indicated by the occurrence of cosmopolitan species despite the lack of any effective mechanisms of dispersal, and combined with a high degree of sympatry and a preference for “exotic” biotope conditions. More detailed reading: Riedl (1969); Sterrer (1972); Sterrer et al. (1985); Reise (1981b); Ax (1985); Sørensen et al. (2003, 2006); Jenner (2004).
5.4.2
Rotifera, Rotatoria
With about 1,000 benthic species, Rotifera are a dominant group of the freshwater meiobenthos, whe reas the taxon has been less frequently documented in marine samples. Greatest species richness is in Notammidae (Cephalodella), Lecanidae (Lecane)
130
5 Meiofauna Taxa: A Systematic Account
and Dicranophoridae (Dicranophorus) with many mesopsammic species (Segers 2008). There are some 60 marine meiobenthic rotifer genera; most species belong to the genera Encentrum (about 50 spp.), Lecane (about 20 spp.) and Proales (about 15 spp.). Species richness (and sampling effort) seems highest in the Northern hemisphere, while diversity culminates in the warmer regions. The structural diversity of rotifers is remarkable and is used for identification of live samples, at least to the generic level. While they can reach 2 mm in length, the bulk of the group is of meiobenthic size. The smallest rotifers (about 50 µm) easily reach the protozoan size range and represent some of the smallest metazoans. The wheel organ is the distinguishing structure from which the scientific group name is derived. Originally a ciliated plate for creeping and possibly sweeping particles from the substratum, its function has been modified and complicated by the formation of two separate ciliary whirls (resembling a rotating wheel), which produce strong incurrent water eddies for food ingestion. There are two important benthic subgroups (Fig. 5.13): Bdelloidea: Rotatoria, Mniobia, Philodina. These live mostly between mosses; reproduction is a diploid parthenogenesis; there are no males. Monogononta, Ploimida: Trichocerca, Lepadella, Lecane, Proales, Notomma, Notholca, Ploima. Common marine species are members of the genera Encentrum, Trichocerca, Proales. Most meiobenthic forms belong to the Monogononta, which has developed a complex alternation of diploid and haploid stages (see below). The second characteristic of Rotifera (beside the wheel organ) is the pharynx, with its complex cuticularized jaws (trophi) that form a grasping and chewing apparatus, the mastax. The trophi and their muscles are important features for rotifer identification and phylogenetic affiliation (see below). The body is covered by a structurally and chemically specific, flexible, syncytial integument, which can be locally reinforced to form plates (e.g., in Notholca), but is not shed during growth. The animals do not develop collagen. The head with the wheel organ is retractable into the trunk, so that, in fixed samples, this diagnostically important organ is often not visible. The body cavity is not lined with a mesothelium. The saccate body usually tapers into a flexible and eversible foot, which usually ends in a pair of toes. Since it carries adhesive glands, the tail with its toes serves as anchoring organ, but in some species it also participates in a looping locomotion where body contractions and extensions alternate while the wheel organ or the foot is fixed. Rotifers are strictly eutelic. They have about 1,000 cells; their cell number is definite and remains constant after the first 5 h of life when mitosis terminates. The males are short-lived and dwarfed, without intestine. Reproduction in the Monogononta Ploimida is by heterogonous alternation of parthenogenetic summer generations and one bisexual winter generation. Sperm transfer is by injection into the females, and the few eggs are frequently brooded. These diploid winter eggs represent resting stages for overwintering. Convergent with tardigrades and some nematodes, many rotifer species can develop cryptobiotic stages that tolerate desiccation. Ecological aspects. Oligotrophic to mesotrophic lakes with a well developed phytal littoral are favourable habitats for benthic rotifers; they can harbour between
Proales Lindia
Fig. 5.13 Some benthic Rotifera, lengths 0.5–1 mm. (Various authors)
Encentrum
Triphylus
Rattulus Philodina
5.4 Gnathifera 131
132
5 Meiofauna Taxa: A Systematic Account
150 to 250 species with a maximum in tropical regions (Segers 2008). A speciose rotifer fauna is also found in slightly acidic, soft waters. Among limnetic meiofauna, rotifers can exceed in abundance not only nematodes, but all other meiofauna (Schmid-Araya 1998; Ricci and Balsamo 2000). Resting stages, together with parthenogenesis, contribute to their ability to quickly colonize and populate new or ephemeral habitats with often drastic population fluctuations. This makes quantitative data such as average rotifer population density (about 100 specimens per 10 cm3 sediment) problematic. In optimal zones, such as the well-oxygenated phytal or detritus-rich and water-saturated sandy shores of lakes, samples contained up to 11,500 ind. per 10 cm3 sediment (Pennak 1940). “Psammobiotic” (Wiszniewski 1934) rotifers have morphological adaptations to an interstitial life: flattened bodies, enlarged adhesive glands, long toes, no pigment and eyes (e.g., many Dicranophoridae). Species less strictly bound to the mesopsammon were termed “psammophilous.” Stygobiotic rotifers live preferably in the groundwater horizons, e.g., Encentrum subterraneum (=Wierzejskiella subterranea), Paradicranophorus (see Sect. 8.7.2). Gravel streams also harbor a rich and diverse rotifer assemblage. In marine habitats there is a marked reduction in rotifer diversity and abundance. About 100 marine species have been recorded, belonging mostly to the genera Lecane, Trichocerca, Lindia, Proales, Lepadella, Testudinella and Colurella. Turner (1993) described between 20 and 790 specimens per 100 cm3 belonging to fifteen species of psammobiotic forms from a marine beach in Florida. Lecane inermis and Colurella salina occurred in particularly high numbers at the high tide mark and in the surface layers (down to about 5 cm). Remane (1949) and Tzschaschel (1979, 1980) often found densities of between 30 and 60 specimens per 100 cm3 in sandy North Sea shores. Also in the marine realm, rotifers seems to occur only in shallow, well-oxygenated sediments (coarse sand, shell hash) and sublittorally down to 300 m water depth. In tidal flats, a reduced oxygen supply can limit their distribution and may explain their preference for the sandy upper subsurface horizons near the highwater mark while the lower reaches and the deeper strata are avoided. In winter, the populations from shallow reaches seem to perform temperature-dependent vertical migrations in somewhat deeper layers. In the void system of Arctic sea ice, close to its lower surface, rotifers seem to be fairly frequent and accounted for > 20% of the overall meiofauna (Gradinger et al. 1999; see Chap. 8.1.1). Other brackish water habitats with fluctuating conditions such as estuaries are also regularly populated by benthic Rotifera. Considering the oxygen demands of most rotifers, the colonization of anoxic and sulfidic deep-sea sediments around gas hydrates and between bacterial mats by new Lecane species is amazing (Sommer et al. 2003). Cyclomorphosis, the regular changes of body form and structure exhibited by some planktonic Rotifera, also occur in some meiobenthic forms (e.g., Mniobia). The adaptive significance of this interesting phenomenon, also described for tardigrades (see Sect. 5.7), remains unclear. Little is known about the nutrition of benthic rotifers. They probably feed on algal cells and protozoans, which they scrape off or grasp using their wheel organ and jaws. Their presence on sea ice suggests that the rich diatom populations are
5.4 Gnathifera
133
utilized. Carnivory has been observed in some planktonic forms. Rotifers are important members of the small food web, particularly as food items for cyclopoid copepods and juvenile fish. Derived, often extreme biological adaptations (cryptobiosis, parthenogenesis) have contributed to the ecological success of rotifers. A phylogenetic relationship to the newly discovered Micrognathozoa seems well founded on the basis of morphological characters (see below), and an affiliation with gnathostomulids is probable. Molecular analyses to scrutinize the validity of the resulting taxon Gnathifera are urgently required. More detailed reading: ecology, Remane (1949); Tzschaschel (1980); Pennak (1951); Ricci and Balsamo (2000) (freshwater); Wiszniewski (1934) (freshwater); taxonomy, Tzschaschel (1979); Voigt and Koste (1978); monograph, mainly anatomy, Clément and Wurdak (1991); phylogeny, Clément (1993); Rieger and Tyler (1995); Kristensen and Funch (2000); identification key, Fontaneto et al. (2008).
5.4.3
Micrognathozoa
Another meiobenthic animal group new to science, the monospecific Micrognathozoa comprising the species Limnognathia maerski was described in 2000 (Kristensen and Funch), although it had been found in Greenland in the late 1970s. The rich moss vegetation of some cold freshwater springs (frozen for most of the year) in a moor outflow harbored many of these tiny animals (about 150 µm long and 55 µm wide), which were initially thought to be rotifers. Some Limnognathia spp. have also been found on the subantarctic Crozet Islands (De Smet 2002). Another discovery of Limnognathia maerski in a cold spring in Wales (Kristensen, pers. comm.) confirms a relationship to cold springs as a preferred habitat. Their characteristic structure is the highly complicated cuticular jaw apparatus, which is ultrastructurally homologous to the jaws of gnathostomulids and rotifers (it consists of microtubular rods). The body is dorsally protected by a set of plates but is ventrally “naked.” The epidermis is cellular, not syncytial. A ventral ciliary field serves as a locomotor organ with which the animals can swim in slow spirals in the moss thicket. Limnognathia frequently adheres to the substrate with its posterior ventral side by means of a sticky pad of cilia that is different in fine structure from the “duo-gland pattern” found in the adhesive toes of rotifers and other adhesive organs. The bipartite head is studded with sensory cilia; long bristles are arranged along the length of the body. The animal bends its thorax like an accordion while moving, and the sack-like abdomen ends in a small tail. Only females have been found, even after long-term rearing in moss cultures, which points to possible parthenogenesis. The two eggs visible in mature specimens have different sizes and structures (winter and summer eggs, like in rotifers). Apart from the complex pharynx and jaw apparatus the digestive tract is rather simple, ending in a periodically opening dorsal anus. The food seems to consist of diatoms and probably also bacteria.
134
5 Meiofauna Taxa: A Systematic Account
Kristensen and Funch (2000) suggested from structural evaluation that the Micrognathozoa should be considered the sister group of the Gnathostomulida, an affiliation that was later confirmed by molecular analysis (Kristensen 2002). On the other hand, there are also some affinities with the Rotifera. Altogether Micrognathozoa have been considered a missing link between these groups; their discovery supports the monophyly of the taxon Gnathifera and rejects the assumed sister-group relationship of Gnathostomulida to Plathyhelminthes (Funch and Kristensen 2002; Kristensen 2002).
Box 5.4 Gnathifera: Elusive, Exotic, Extreme Two of the “youngest” animal groups, the Gnathostomulida and the Micrognathozoa, are meiobenthic and assigned to the Gnathifera. They eluded discovery for a long time because of their exotic biotopes: sulfidic sediments (Gnathostomulida) and moor springs in Greenland (Micrognathozoa). In contrast, the third taxon of the Gnathifera, the Rotifera, has been well known since the beginning of microscopy in the seventeenth century. Their extreme character is only revealed by life history studies. Desiccation? No problem, try resistant resting stages. Remote habitats without a sexual partner? Colonize them via air transport and reproduce parthenogenetically. The early cessation of cell cleavage with a defined number and fate of cells adds to these peculiarities. The prime character that unifies these three, rather diverse-looking phyla are their elaborate and conspicuous jaw apparatuses, which are probably homologous. Many questions about these structurally and biologically aberrant meiobenthic forms remain to be answered: Males in Micrognathozoa? Their survival during eight months of frost? Relation of Gnathostomulida to sulphide, their metabolism in sulphidic bottoms that are permanently under a thick brine seep? Complexity of reproductive cycles in Rotifera, triggers that induce meiosis and the change from the diploid to the haploid phase? Adaptive significance of cyclomorphosis in benthic rotifers? - We could therefore add the connotation “enigmatic” to the terms listed in the caption.
5.5
Nemertinea
The whole phylum Nemertinea seems perfectly preadapted to a life as meiobenthos when the size is reduced. This may be why the smallest, meiobenthic representatives among them (about 50 species) have essentially the same organizational pattern as the huge ribbon worms (Fig. 5.14). Many of the typical mesopsammic nemerteans are relatively long (up to several centimeters in length), but are extremely flexible and thin. They are totally ciliated, have a unique ectodermal proboscis that protrudes from a cavity that is involved in locomotion and is lined by mesoderm, a “rhynchocoel.” The proboscis is armed
5.5 Nemertinea
135
Oerstedia dorsalis (8 mm) Protostomatella arenicola (8 mm)
Micrura fasciolata (15 mm)
Ototyphlonemertes sp. (8 mm)
Fig. 5.14 Some meiobenthic Nemertinea. (Various authors)
with viscid glands, and, in the order Hoplonemertinea, with stylets used to catch prey. It works independent of the oral opening to grasp prey, similar to kalyptorhynch turbellarians. In some species it is also used as a locomotor organ. In some heteronemertines the head is flanked by two deep lateral slits that form sensory organs. In many meiobenthic nemertines, the head has a pair of anterior statocysts associated with the brain. Nemerteans have a four-lobed cerebral center with a large ventral part. The blood vascular system is closed and numerous pseudosegmental protonephridia and gonads run in long rows along each side of the body. The complicated body musculature allows for extreme flexibility and ever-changing body form and width: nemerteans can contract their body to 1/12 of their length and squeeze through tiny voids. The majority of the meiobenthic species live interstitially in marine sand (30 species) and belong to the class Enopla (i.e., “armed nemerteans”), of which the
136
5 Meiofauna Taxa: A Systematic Account
Hoplonemertinea are the main subgroup; these carry stylets at the tip of the proboscis. Characteristic genera from sandy bottoms like Ototyphlonemertes and Arenonemertes clearly demonstrate the adaptations to their interstitial habitat: gliding ciliary movement, sticky areas achieved through numerous epidermal glands, sensory cirri, paired statocysts, formation of a tail (in some species), turgescent chordoid tissues (O. antipai), suppression of the planktonic “Pilidium” larva. There are reductive trends in the cerebral organs, in eye development, in the intestinal diverticula, and in the number of gonads and eggs. Ingole et al. (2005) reported nemertines ranking second after nematodes in muddy deep-sea bottoms with an amazing 11 ind. 10 cm−2. Some nemerteans, most of which belong to the Anopla, lack ocelli, statocysts and any proboscis armature. Their mouths are ventral and they all are of meiobenthic size (the most common genus is Cephalothrix, which has a mouth far posterior to the cephalic tip; other genera are Procephalothrix, Carinina). The rather small species of the Anopla often live epibenthically on/in muddy bottoms. While Ototyphlonemertes (see Fig. 5.14) has a worldwide distribution, other genera like Arenonemertes are only known from a few species of restricted occurrence. Members of Prostoma and Potamonemertes can be considered stygobiotic freshwater forms. Nemerteans are rather insensitive to factors like grain size, oxygen and even food supply. They are voracious predators and also eat carrion, but they can survive long phases of starvation while reducing their body size. Their sluggish, extremely delicate bodies must be handled with great care during investigation. Even the gentlest method, extraction through deterioration of the (oxic) environment when a sediment sample is kept for some time, is not quantitativily effective for this group. Nemertinea is an isolated animal group whose relationships are rather ambiguously discussed, especially its connection with Plathyhelminthes (Garey 2002). Independent molecular studies support a position near the annelids sensu lato (Trochozoa, Spiralia) (Turbeville 2002; Jenner 2004; Petrov and Vladychenskaya 2005). More detailed reading: interstitial species, Gerner (1969); taxonomy, key for interstitial species, Norenburg (1988); phylogeny, Riser 1985, 1989; Jenner (2004).
5.6
Nemathelminthes: A Valid Taxon?
The heterogeneous combination of various animal classes into the phylum Nemathelminthes is probably not a natural, monophyletic unit. Therefore, nemathelminths are often considered to comprise independent phyla, which are probably unified through a common evolutionary trend for dwarfism, but differ in terms of varying reductive phases of coelom formation. In some nematodes, priapulids and gastrotrichs, the body cavity is completely lined with a mesodermal coelothel, i.e., a coelom is present in the cases studied. In other forms this mesothelium is absent, incomplete or reduced. Subsequently, the body cavity has been characterized as a “pseudocoelom” and nemathelminths have also been termed “pseudocoelomates”
5.6 Nemathelminthes: A Valid Taxon?
137
in some textbooks. In many groups, the body cavity is reduced and is more or less densely packed with the inner organs (numerous nematodes and gastrotrichs); in others it is spacious and filled only with body fluids and some mesenchymatous cells (kinorhynchs, priapulids). A new phylogenetic concept is to group together large taxa that have an acellular, often chitinous cuticle that is molted: The Arthropoda and some of the former nemathelminth phyla (nematodes, kinorhynchs, loriciferans, priapulids) represent the supertaxon “Ecdysozoa.” This grouping, which breaks up the traditional taxa Nemathelminthes and Articulata, is based on molecular analyses (18S rRNA and Hox gene studies) as well as chemical signatures. It has received increasing molecular support recently (Zrzavy 2001, Garey 2002, Giribet 2003, Philippe et al. 2005, Mallat and Giribet 2006, Petrov and Vladychenskaya 2006). Accordingly, the gastrotrichs, which lack chitin, would be separated from the Ecdysozoa, a position again supported by cladistic and DNA analyses. However, the Ecdysozoa hypothesis has been rejected by some zoologists, who contend that it lacks support from homologous morphological characters and paleontological evidence (see Wägele and Misof 2001). In other (morphologically based) concepts, nemathelminths are split with regard to the presence of a retractable anterior body part (introvert) or vestiges thereof (“Introverta:” Nematoda, Kinorhyncha, Loricifera, Priapulida), or with regard to the body armature of spiny plates or scales (“Scalidophora:” Priapulida, Kinorhyncha, Loricifera). The latter grouping has recently been shown to be doubtful on the basis of detailed molecular analyses of several loriciferan species (Sørensen et al. 2008). Since the positioning is so controversial and the molecular information is still ambiguous, “Nemathelminthes” is retained in this nonphylogenetic treatise, but the reader should be open for this exciting new phylogenetic debate in which the role of progenetic trends should be scrutinized. The free-living nemathelminths are nonsegmented, bilateral animals, mostly with a mouth and an anus, and often with a complex cuticular lining covering the epidermis. In the Scalidophora the cuticle contains chitin. Direct development after internal fertilization of the gonochoristic species prevails. There is a trend towards a defined and fixed cell number (eutelic development) with little regenerative potential. Asexual development is not known, but parthenogenesis occurs in some taxa (e.g., in Loricifera).
5.6.1
Nematoda (Free-Living)
The most frequent metazoans, nematodes usually dominate each meiofaunal sample both in abundance and biomass. Many free-living nematodes are of meiobenthic size; however, numerous undescribed nematodes in the deep-sea are macrobenthic reach the macrobenthic size range (Lambshead, pers. commun.). Most meiobenthic nematodes are 0.5–3 mm long, the smallest being 0.2 mm; their lengths are 20–40 times their width. In terms of their general body structure many features pre-adapt them to living in sands and muds. Nematodes occur in all substrates,
138
5 Meiofauna Taxa: A Systematic Account
sediments, climatic zones and water depths. 80–95% of the individuals and 50–90% of the biomass of meiobenthos usually consists of nematodes (Fig. 5.15a,b). Each square meter of sea floor is populated by 1–12 million nematodes (Heip et al. 1985a; Soetaert et al. 1995). Correspondingly high densities are found in freshwater biotopes, especially in the littorals of lakes (Traunspurger 2002). About 4,000–5,000 species of free-living marine and about 11,000–20,000 nematode species have been described in total so far. Records of nominal nematode species differ much; about 20,000 free-living species have been described, the vast majority (90%) living in marine habitats (Eyualem-Abebe et al. 2008). Some researchers consider nematodes a “hyperdiverse” taxon with > 1 million species, while others doubt this estimate (Lambshead and Boucher 2003). Even if about 100,000 species is assumed to be a realistic guess (Coomans 2000), completely
% others ⬙Turbellaria Harpacticoida Nematoda
80
60
40 20
Feb.
a
Jul. 1978
Dec.
% others Gastrotrichs Kinorhynchs Ostracods Polychaetes Nauplii Copepods Nematodes
80 60 40 20
Dec. Feb.
b
Jun. 1997
Dec.
Apr. 1998
Fig. 5.15 a–b Dominance of nematodes among meiofaunal groups, as indicated by their relative abundance. a From the shallow sublittoral of the Belgian North Sea coast; b from deep-sea mud. (a After Herman et al. 1985; b compiled from Shimanaga et al. 2000)
5.6 Nemathelminthes: A Valid Taxon?
139
describing them all would be an utopian task considering the number of specialists required. The marine biotopes around the British Isles alone harbor 41 nematode families with 450 species. Another estimate for the North Sea sediments is 800 species, and 1,625 species have been recorded in European seas (Costello et al. 2006). In meiofaunal samples, the number of species of nematodes often exceeds that of all the other groups put together by an order of magnitude. Therefore, 100 cm3 of shallow-water sand may contain 20 species. In deep-sea sediments, where the species richness is spectacularly high, this number can even be higher. “Contrary to popular opinion, marine nematodes do not all look the same and it is time that this myth was finally put to rest” (Platt and Warwick 1980). The eminent and diverse ecological importance of nematodes was only realized when their undifferentiated treatment as a huge but hardly identifiable bulk taxon was abandoned. The recent increase in meiofaunal studies with more detailed nematode analyses led to a more realistic assessment of their ecological role. This is in part the result of better identification media, techniques and training workshops. Today, valuable illustrated keys and databases are available (Platt and Warwick 1983, 1988, Warwick et al. 1998; Vandepitte et al. 2008). Several electronic keys have been designed specifically for nematodes (Tarjan and Keppner 1999; Deprez 2006. These demonstrate that it is not so much a lack of valid and conspicuous diagnostic features that deters meiobenthologists from dealing with nematodes, but rather their exorbitant species richness and often tiny sizes. Identification of meiobenthic nematodes is based mainly on characters that are visible without dissection, but this requires a good compound microscope (“Nomarski” interference contrast is very helpful). The general body and tail forms of nematodes are by no means uniform (which is often supposed; see Fig. 5.16). Together with marked buccal cavity differences (see below), this allows the spectacular species richness to be broken down into large groups as a first step towards taxonomic ordination.
5.6.1.1 Taxonomy and Systematics Taxonomically relevant external features are: tail shape and spicular apparatus, numbers and arrangements of sensory setae (particularly around the head), of caudal glands and of gonads, the positions and shapes of amphids (paired anterior chemical sense organs), and epicuticular structures (e.g., annulation). The most important internal features are the shape and cuticularization of the buccal cavity, the structure of the genital organs, and sperm morphology. In some taxa the general body shape and its external cuticle is of diagnostic value (e.g., some Chromadoridae, Desmoscolecida, Draconematidae and Epsilonematinae; see Fig. 5.16), while in most nematode groups the body is rather uniform, with the anterior end more truncated and the posterior end slender; in males this merges into an inflected tip. Thistle et al. (1995) assert that for deep-sea nematodes the tail shape in combination with the buccal armature yields a promising analytical tool for future work. The scope of this book only allows to mention a few more comprehensive higher taxa and some extraordinary species. For more details, perusal of the original
140
5 Meiofauna Taxa: A Systematic Account
50 µm
200 µm 500 µm
Eubostrichus 100 µm
Desmoscolex
50 µm Paramonohystera
50 µm
Eleutherolaimus (front end)
200 µm
Draconema Echinotheristus
Fig. 5.16
Metepsilonema
Various Nematoda of different appearance. (Various authors)
literature is recommended. An earlier taxonomic division into two subgroups, the Adenophorea and the Secernentea were confirmed in the system of Lorenzen (1994). Since most of the pertinent (ecological) meiobenthic literature is still based on this classification, these categories are used here too. While Kampfer et al. (1998) maintain that Adenophorea is monophyletic, other specialists reject the monophyly of this taxon and suggest a division of Nematoda into two classes, Enoplea and Chromadorea (Coomans 2000). With increasing molecular studies (> 200 taxa now analyzed) this general division was confirmed (De Ley and Blaxter 2004) and maintained after the addition of numerous marine species to the analysis (Meldal et al. 2007). This would suggest that the long-standing and widely adopted divison of nematodes into “Adenophorea” and “Secernentea” should be abandoned. Yet the systematics of the Nematoda have undergone further changes and even the major taxa are still not stable or generally recognized. Among the Chromadorea, a cluster termed the “Rhabditida” has been resolved (this was previously the “Secernentea”). At least Monhysterida, Chromadorida and Desmodorida seem monophyletic (Litvaitis et al. 2000). Recent overviews of
5.6 Nemathelminthes: A Valid Taxon?
141
the phylogenetic relationships of nematodes are given by De Ley et al. (2006) and Meldal et al. (2007). Since even nematode orders are far from stable, we need more genetic, morphological and ultrastructural details to further resolve the natural phylogenetic units in nematodes. Complete genomic deciphering of two Caenorhabditis species demonstrated the high variability of the genetic background, even in closely related nematodes. Only 35–70% of the genome is similar to other nematodes (Parkinson et al. 2004), underlining the exhaustive genetic diversity in this phylum. It also underlines the urgency with which further genetic analyses for use in the systematics of lower nematode taxa are required. As the work of Derycke et al. (2005, 2007a, 2008) has illustrated, there is high intraspecific allelic and genetic variance with considerable local differences. These may be valuable for ecological population studies but they are misleading in taxonomic research using molecular methods. As yet, there is a huge gap in this promising field that can hopefully be tackled using new fixation methods that yield morphological descriptions and genetic identification for the same specimen (Bhadury et al. 2005). Adenophorea: Setae and adhesive glands are present, amphids are conspicuous. Most meiobenthic nematodes belong to this subgroup. Mostly from marine and brackish water sites, but some 400 species of various families are also limnetic and even stygobiotic (mainly Dorylaimida, Tobrilidae, Monhysteridae, Desmoscolecidae, Xyalidae: Theristus). Important orders with numerous genera and species include the Monhysterida, Chromadorida, Enoplida and Trefusiida. The family Desmoscolecidae sometimes also takes ordinal rank as Desmoscolecina. Trichodorus, Tripyla and Onchulidae, with their soft cuticles and aberrant spicular apparatuses, are considered primitive taxa. Secernentea: No setae and adhesive glands, amphids are much reduced. Within the scope of this book, this taxon is of less relevance, since it contains the bulk of terrestrial, or parasitic and pathogenic forms. Among the meiobenthic forms there are those limnic groups that prefer a certain degree of organic pollution (Rhabditidae, Diplogasteridae). Only a few species of “halophilic” Secernentea occur in biotopes of marine shores, albeit often in large numbers (e.g., Pellioditis, Rhabditis marina).
5.6.1.2
Biological Aspects
Nematodes lie normally on their side and move, in the absence of circular muscles, in their natural environment by alternating contractions of their longitudinal musculature in a characteristic dorso-ventral bending action. In conjunction with the surrounding sediment this results in a somewhat snake-like, writhing motion performed at a relatively high speed (Cullen 1973). Despite the fact that they are mostly small in size, this motion pushes them along at some 15 cm per minute. The contractions increase the internal turgor and cause pressure waves to run along the body. In flocculent sediments biostabilization through mucus secretion may render wriggling movements more efficient (Riemann 1995). Some species
142
5 Meiofauna Taxa: A Systematic Account
can jerkily “jump” by rapidly bending their bodies with subsequent sudden relaxation (e.g., in Theristus). Only the structurally aberrant Epsilonematidae and Draconematidae move by “looping,” which involves alternatively adhering and detaching their mouths and tail regions (see Fig. 4.8). Desmoscolecidae move by “contractive waves.” Under artificial conditions, when performed in a drop of water under the microscope, nematode movement is rather awkward and helpless wriggling up and down. Although about 45% of all adenophoran nematodes are interstitial, there are not many morphological features that occur regularly in mesopsammic species and can be ascribed to the mesopsammic life, although their functions are often not clear (Lorenzen 1986): a strongly bent tail with adhesive glands, additional adhesive organs (e.g., in Epsilonematidae), a flexible, tapering “neck” region, extremely long setae (Thrichotheristus), aberrant positions of amphids (Epsilonematidae), flattened body (rare, e.g., Neochromadora angelica). Life history. Complete life tables of nematodes are known for just a few species. The following data refer mainly to the ecological review of Heip et al. (1985): Within a small sample, species with a generation cycle of a few days can co-occur with others that need an entire year to complete their generation cycle, making generalizations about “the typical nematode” problematic. For a realistic assessment of their production and ecological relevance, complete life history studies on more nematodes species (e.g., on Chromadorina germanica by Tietjen and Lee 1977; Chromadorita tenuis by Jensen 1983) are needed and can be expected with the increasing ability to cultivate nematode species (Moens and Vincx 1998). Adult females usually produce between 10 and 50 eggs, and after hatching the juveniles undergo four molts before reaching maturity. All cuticular structures (and the armature of the buccal cavity) are shed with each molt. Growth continues after the last molt. The average nematode has a fresh weight of 1 µg. Generation time is mostly between 13 and 60 days, resulting in 4–10 annual generations (however, there is also a species with a generation time of just 3 days—Pellioditis marina—and others with only one reproductive period per year!). The high production of nematodes contrasts with their rather low biomass; the annual P/B ratio is about 8–10 (see Sect. 9.3.2). While marine nematodes reproduce by bisexual amphimixis, in freshwater habitats obligate or facultative parthenogenetic or hermaphroditic development of nematodes has repeatedly been reported (for details see Nicholas 1984), with populations consisting either exclusively (e.g., Chronogaster spp.) or predominantly (e.g. Eumonhystera spp.) of females. For more detailed life history data on freshwater nematodes, see Traunspurger (2002) and Bergtold and Traunspurger (2006). For those who can read Russian, the monograph on the biology of marine nematodes by Chesunov (2006) is recommended.
5.6.1.3
Ecological Aspects
Wieser (1953, 1959), in his seminal papers on the ecology of nematodes from European and American littoral coasts, was the first who found a relation between community structure, sediment granulometry and trophic guilds expressed in the cuticular armatures of the nematode’s buccal cavity (Fig. 5.17).
5.6 Nemathelminthes: A Valid Taxon?
143
FINE SAND, RICH ORGANIC CONTENT
COARSE SAND, LOW ORGANIC CONTENT
Carnivores, omnivores, epistrate feeders
SHELTERED OR EXPOSED PHYTAL
Epistrate feeders, omnivores, selective deposit feeders
Selective and non-selective deposit feeders, carnivores
Deposit feeders, often non-selective MUD, SILTY SAND
Fig. 5.17 Trophic guilds of nematodes from various tidal habitats based on different mouth structures. (After Wieser 1953 and other authors)
1. In mostly homogeneous muds and fine sand, nonselective deposit feeders with a well developed but weakly cuticularized buccal cavity prevail. Food particles, often larger bacteria, detritus and diatom cells, are taken up using the lips and the anterior buccal cavity. 2. In more heterogeneous (fine) sandy substrates, selective or nonselective deposit feeders with a small or vestigial non-cuticularized buccal cavity are found. Their food particles (bacteria, small detritus) are soft and mostly obtained by suction of the muscular pharynx. 3. In sand with more microhabitats epigrowth feeders prevail. They scrape off the (algal) surfaces of grains or pierce single cells. Hard cuticular ridges for scooping or pointed tips for piercing are well developed in the narrow buccal cavity. Coarser sandy sediments contain mainly predators and omnivores with large and powerful pointed teeth and lancets as buccal armature; their buccal cavities are wide. 4. The exposed phytal is dominated by algivores and predators/omnivores. In more sheltered algal sites an increasing number of epigrowth feeders and selective deposit feeders scrape off and pick up particles (epigrowth and detritus) from the thalli of the plants.
144
5 Meiofauna Taxa: A Systematic Account
Given the huge number of nematode species, it is not surprising that all trophic resources are utilized, but it has become increasingly clear that many species originally considered generalists are actually specialized feeders with a high rate of trophic niche partitioning. Wieser’s (1953, 1959) allocation of (marine) nematodes into four trophic groups was developed under distributional aspects and was intended to serve as a practical way of defining communities. It is this simplification that makes Wieser’s classification applicable to nematode studies from many other areas (e.g., Ott 1972a; Kennedy 1994a). Of course, this grouping had to be refined to meet the requirements of local nematode populations (Joint et al. 1982; Jensen 1987a; Romeyn and Bouwman 1983). A modified classification with ciliate feeders as a separate group was introduced by Moens and Vincx (1997a) for the nematode fauna of estuarine tidal flats (Fig. 5.18). To complicate the situation, the variability of buccal structures also depends on ontogenetic age (Lorenzen 2000). Moreover, when assigning nematodes to feeding groups it is important to consider
Facultative predators
Microvores
Ciliate feeders
Predators
Deposit feeders
Epigrowth feeders
Fig. 5.18 A modified categorization of the nematode feeding groups from estuarine tidal flats. (After Moens and Vincx 1997a)
5.6 Nemathelminthes: A Valid Taxon?
145
that many opportunistic nematodes can switch their feeding mode in response to the food available. Today, there is substantial evidence that, due to their restricted capacity for proteolytic nutrient digestion, nematodes are mostly selective feeders. The high syntopic diversity of nematode species has been interpreted as being a consequence of this highly developed food partitioning. They can selectively differentiate between prey organisms, and even between bacteria (Moens et al. 1999b), which seem to be the main nutritive source for the group, especially in the deep-sea. The worms accelerate bacterial digestion, secreting bactericidal lysozymes (beta-glucoronidase) which dissolve the thick cell walls (Tietjen and Lee 1977). The bacterial symbiotic Stilbonematinae live exclusively on their own epicuticular “bacterial kitchengardens” (review: Ott et al. 2004). Aggregations of bacteria on decaying carrion or plant debris strongly attract nematodes. When (bacterivorous) nematodes were added to decomposing wrack, degradation times were significantly reduced. This detritus was then the preferred food of macrobenthic polychaetes (Tenore et al. 1977b; Rieper-Kirchner 1989). The mucus secretions that nematodes leave in their trails stimulate and/or modify bacterial growth (the gardening hypothesis, Riemann and Schrage 1978; Jensen 1996; Moens et al. 2005; freshwater: Traunspurger et al. 1997). The secretions seem to have enzymatic effects on refractory polysaccharides such as cellulose (Riemann and Helmke 2002). The mucus-lined fine nematode burrows modify the microtexture and thus the microclimate in the sediment (Cullen 1973; Fenchel 1996) and in biofilms (Mathieu et al. 2007). The rather stable tubes of Ptycholaimellus increased the aeration of the deeper layers, modified the depth of the oxic/anoxic interface, and influenced the occurrence of other meiofauna (Nehring et al. 1990). It has repeatedly been pointed out that nematodes take up considerable amounts of dissolved/lysed substances that are not discernible in the intestine (e.g., Adoncholaimus fuscus, Moens et al. 1999c; see also Moens et al. 2006). This mode of nutrition is of obvious relevance in the gutless genera Astomonema and Rhabtothyreus. Odor compounds produced by biofilms have also been shown to act attractively on nematodes (Höckelmann et al. 2004). For freshwater nematodes Traunspurger (1997b) developed a tropho-ecologically based classification which is probably also applicable to marine and terrestrial species. He discerned four categories: 1. Swallowers (deposit feeders) without teeth feeding on whole protists 2. Epistrate feeders that use a small tooth to tear and swallow protists and microalgae 3. Chewers with a wide and sclerotized buccal cavity (the traditional omnivores/ predators) 4. Suction feeders with a stylet apparatus that suck plants, fungi and animals Studying several trophically different lakes in Germany demonstrated the wide applicability of this classification (Moens et al. 2006, Fig. 5.19 ) In consumption experiments with tidal flat nematodes, microalgae were preferred over bacteria (Pascal et al. 2008). Especially in tidal flats, many diatom-eating nematodes
146
5 Meiofauna Taxa: A Systematic Account
graze on the rich stock of microphytobenthos (Montagna 1995; Santos and SouzaSantos; 1995; Riera et al. 1996; Moens and Vincx 1997a; Middelburg et al. 2000), crushing the silica shells of diatoms with specialized mouth parts (Fig. 5.20). Here, more than 30% of the nematodes represented microalgal grazers that actively migrated to their favorite food patches, creating a dynamic and irregular distribution pattern (Moens et al. 1999a) that is different from the small-scale distribution created by DOM users (Blome et al. 1999). Not only buccal structures, but also the general morphometrics of nematodes such as the width/length ratio and the general body mass, seem to be influenced by the availability and type of food and to be related to the feeding type. A granulometric influence on nematode morphometrics is also widely discussed, as are other factors such as hydrodynamics (erosion) or oxygen availability. Microvores tend to have a small width/length ratio (slender worms), while greater ratios defined predators and epigrowth feeders (Tita et al. 1999). The predatory taxa prevailing in medium-to-coarse sands are usually robust (e.g., Desmodoridae) and often have well-sculptured cuticles and long setae (Draconematidae, Epsilonematidae and Desmoscolecida, Fig. 5.16). Thus, stout, bulky nematodes seem to be characteristic of shallow sandy sites and the uppermost centimeters with good oxygen availability. However, on sand banks with high wave exposure Vanaverbecke et al. (2007) noticed a trend towards longer bodies, which perhaps prevents suspension. Similarly ambiguous are the trends in finer sediments. The nematodes generally become smaller and more slender in barely exposed finer sediments (Udalov et al. 2005).
100
Königssee (oligotrophic)
80
Obersee (eutrophic) 60
40
20
0 %
Swallowers
Epistrate feeders
Chewers
Suction feeders
Fig. 5.19 Relative distributions of nematode feeding types (Traunspurger classification) in the littoral of two ecologically different lakes. (After Moens et al. 2006)
5.6 Nemathelminthes: A Valid Taxon?
147
Hypodontolaimus balticus Fig. 5.20 a–b Ingestion of benthic diatoms by specialized nematodes. a Mouthparts breaking the shells open; b Mouthparts piercing the shells and sucking the contents out. (After v. Thun 1968)
The few typical mud inhabitants are mostly small and have short setae (e.g., Sabatieria pulchra). On the other hand, Tita et al. (1999) recorded that the average body width and biomass of nematodes were higher in mud than in sand, resulting in a higher calculated respiration. Soetaert et al. (2002) contended that nematode body length increases when the sediments become little permeable and more cohesive. This is perhaps related to organic matter, but probably also to oxygen supply. The typical thiobiotic species from “sulfide layers” (see Sect. 8.4.2) has a long, slender and sluggish body with a thin cuticle (Linhomoeidae, Molgolaimidae). This longer and more slender shape has been interpreted as enabling higher mobility, facilitating effective vertical migrations; other advantages of a slender body may be the enhanced transepidermal uptake of oxygen and dissolved food. Referring to oxygen and hydrogen sulfide, Ott separated already in 1972 the nematode community of an American salt marsh into four associations. With increasing water depth, food limitation seems to become the decisive factor and causes a decrease in the size spectrum in deep-sea nematodes compared to shallow-water species (Thiel 1975; Udalov et al. 2005). In contrast, a shift towards larger sizes has been recorded in deep-sea nematode communities from Antarctic regions. The prevailing low temperatures have been put forward as the reason for this contradictory evidence, and so it may be only a local phenomenon (De Broyer
148
5 Meiofauna Taxa: A Systematic Account
et al. 2001, see Sect. 8.3). In the deep-sea, tail shape (rounded/conical/long, clavate) has been found to be another convenient parameter that significantly indicates differences between nematode communities from energetically different sites (Thistle et al. 1995a). Summarizing, a combination of feeding group with some additional morphometric and granulometric characteristics can probably be related to distinct nematode communities from different biotopes, depths and geographic ranges (Warwick and Gee 1984; Vanreusel 1990; Vincx et al. 1990; Gheskiere et al. 2005; Schratzberger et al. 2004). Schratzberger et al. (2007a) analyzed nematode community functions and combined a set of selected morphological features (body size, and shape, buccal structure, tail shape) with known biological traits. Assigning these functional groups is perhaps the most informative system used to connect the diverse biological requirements of nematodes with the functional dynamics of the community. Median grain size, silt content, and water depth were found to be attributes relevant to functional grouping. Nicholas et al. (1991) related the pattern of nutritive guilds of mangrove nematodes to the shore profile and the emersion gradient: epistrate and diatom feeders prevailed in the low tide zone, mostly selective microbial feeders were found in the higher mangrove zone, while the sediments above high water neap were dominated by omnivores, predators and plant root feeders. Kennedy (1994b) compared the relative roles of the four main nematode feeding types in a sandy and a muddy site respectively from the Exe Estuary in England (Fig. 5.21): omnivores/predators, although of subordinate abundance, dominated in biomass and production in the sandy site, while deposit feeders prevailed in density and in biomass or production in the muddy site. Focusing on the carbon production (Fig. 5.22, more than 50% was consumed by nonselective deposit feeders in the muddy site, while omnivores/predators assumed this prevailing role in the sandy site. There was even a clear differentiation among deposit feeders, with surface species preferring fresh organic matter while deeper dwelling species devoured older material.
5.6.1.4 Biodiversity The higher the degree of habitat microstructure, the richer the nematode community. Heterogeneous, fine sands in shallow sea bottoms with a rich food supply and an interstitial system that provides enough solute and oxygen transport harbor the highest number of species; about 100 species per investigation area are not unusual. As the amount of silt and organic content increases the occurrence of many stenotopic nematode species in the sediment is limited, and so the diversity tends to decrease. More eurytopic species show an affinity for silty or inhomogeneous sediments. As with other taxa, nematode biodiversity in the deep-sea is amazingly high and has not been satisfactorily explained (Packer et al. 2006; see Sect. 8.3). Extreme habitats such as wave-beaten beaches, hadal troughs or sulfidic muds with more homogeneous sediment structures and low organic contents are often popu-
5.6 Nemathelminthes: A Valid Taxon? 80 70 60 50 40 30 20 10 0 %
149
Abundance Mud site Sand site
Deposit feeders
60
Epigrowth feeders Omnivores/Predators
Biomass
50 40 30 20 10 0 %
Deposit feeders
Epigrowth feeders Omnivores/Predators
60
Production
50 40 30 20 10 0 %
Deposit feeders
Epigrowth feeders Omnivores/Predators
Fig. 5.21 Relationship of nematode feeding type to mud and sand sites in the Exe Estuary. (After Kennedy 1994b)
lated by a rather monotonous community of characteristic specialists (Vincx et al. 1990). Strangely enough, some specialists from shallow sandy sediments (Epsilonematidae, Draconematidae) are often also characteristic of deep-sea sites. In a study from the sublittoral of subantarctic regions, deposit feeders prevailed while epistrate feeders were always frequent (Vanhove et al. 2000). As predicted by trophodynamic models, the diversity decreases upon nutrient depletion as a result of increased competition. Conversely, both density and diversity of selective deposit- and epistrate-feeding nematodes increased in the Belgian shelf area after phytoplankton blooms (Vanaverbeke et al. 2004). Boucher (1990) searched for factors determining the diversity patterns of sublittoral nematode assemblages, and concluded from a comprehensive evaluation of literature data that sediment
150
5 Meiofauna Taxa: A Systematic Account
MUD SITE
1.5 % Selective deposit feeders
1.1 %
SAND SITE
7.04 g C x m-2 yr-1
55.9 % Non-selective deposit feeders
16.1 %
14.8 %
27.8 %
Epigrowth feeders
Omnivores / Predators
25.1 %
57.4 %
5.42 g C x m-2 yr-1
Fig. 5.22 Partitioning of carbon consumption among nematode feeding groups in mud and sand sites of the Exe Estuary, England. (After Kennedy 1994b).
granulometry or trophic differences alone could not be the only factors responsible, since samples from muddy and sandy sediments did not differ significantly in their species richness and Shannon–Weaner index. It seems that local or large-scale biotopical factors, e.g., salinity gradients or pollution, are additionally modulating the distributional pattern (Vanreusel 1990, 1991; Bouchet and Lambshead 1995). Nematode abundance and biomass is largely correlated with food supply, and often with bacterial density. However, trophic studies in a tidal mudflat revealed that nematodes selectively preferred microalgae to bacterial food (Pascal et al. 2008). This would explain why shallow flats rich in organics seem to attain highest abundances. In North Sea flats among others, abundances of >15×106 m−2 have been found,
5.6 Nemathelminthes: A Valid Taxon?
151
decreasing towards the sublittoral to about 1 × 106 m−2. In the deep-sea, considerably lower values (around 1 × 104 m−2) have been recorded (Sect. 8.3). Densities in favorable freshwater habitats reach the highest values reported from marine habitats (see Traunspurger 2002; Michiels and Traunspurger 2005). In many limnetic ecosystems (e.g., in lakes), nematodes can contribute 50% or more to the total meiobenthic production, providing more evidence of their salient ecological role. The intensive taxonomic radiation and trophic differentiation of nematodes already indicates a distribution pattern with many local and seasonal variations. From the shallow sublittoral towards the deeper bottoms of the continental shelf, nematodes decrease only slightly in abundance. There is also only a slight reduction in nematode dominance towards the eulittoral. The same is true for the nematode occurrence along the salinity profile in estuarine waters (Soetaert et al. 1995). Apparently, among the high number of species there are enough generalists and specialists for each combination of factors to compensate for those nematodes that drop out (cf. Riemann 1966).
5.6.1.5 Vertical Distribution A community of “epibenthic” nematodes can be found on coral fragments or plant blades that consists of highly specialized and rare taxa. However, in sediments the majority of nematodes aggregate in the uppermost few centimeters, with diatom feeders (Spilophorella, Ptycholaimellus) at the very surface. The vertical decrease in abundance/diversity is highly dependent on the oxygen supply: well-oxygenated sands of exposed beaches can be populated by nematodes down to a depth of 1 m! In muds, which have steep oxygen gradients, the bulk of nematodes live in the uppermost few centimeters. However, specialized nematodes can be found even beneath a well-developed oxycline. In deeper shelf and canyon bottoms, more than 50% of all nematodes occurred in the anoxic layers (Soetaert et al. 2002). The presence of a distinct nematode community in these layers (Ott 1972), specialized in many ways to living under sulfide-dominated conditions, has been widely reported (see Sect. 8.4.2). Stilbonematinae (Chromadoridae) and the gutless Astomonema live in symbiosis with “sulfur bacteria” (Ott 1993; Musat et al. 2007); but the aposymbiotic Linhomoeidae, Siphonolaimus, Cyatholaimus, Terschellingia longicaudata, Monhystera disjuncta, Sabatieria pulchra and some deep-sea Epsilonematidae also have their highest abundances around or beneath this chemocline. Also, the bacterivorous limnetic taxa (Tobrilus spp., Chronogaster troglodytes, Poikilolaimus sp.) seem to thrive preferentially at the food-rich oxic/sulfidic interface. Paramonohystera wieseri from anoxic depths has been experimentally shown to survive extremes of temperature better under anoxia than in normoxic conditions, and is termed “obligate anaerobic” (Wieser et al. 1974). Various deep-sea epsilonematid species populate the muds of the suboxic upwelling zone off Peru (Decraemer et al. 2005). Eudorylaimus andrassy lives in oxygen-free zones of Lake Tiberias (Israel) for eight months each year and can survive being placed in a sealed jar
152
5 Meiofauna Taxa: A Systematic Account
with completely anoxic sediment for six months (Por and Masry 1968). Theristus anoxybioticus performs a unique biotopic change from the oxic surface to deep anoxic layers during its lifetime: adults live in the upper centimeters of muds of methanic seepage sediments in the Kattegatt (Denmark), while their juveniles migrate into the anoxic subsurface layers (Jensen 1995). Vertical distributions may vary, and many nematodes migrate diurnally or seasonally, often depending on the oxygen availability or tidal cycles (Hendelberg and Jensen 1993). Jensen (1984a) suggested that in Finnish waters low temperatures indirectly influence the vertical distribution of epigrowth-feeding phytal nematodes: the animals change to a bottom-dwelling lifestyle when their habitat, the algal canopy, is destroyed by ice. They leave the sea bottom again when the vegetation starts growing in spring. A substrate-mediated and food-influenced annual growth cycle was also found by Novak (1989) for the nematode fauna of a Mediterranean seagrass area, reflecting the growth cycle of the plants. Alongi (1990a) confirmed a food-mediated seasonal change in nematode dominance: deposit feeders and predators dominated during the cooler months of fall and winter, while epigrowth feeders were most frequent in spring and summer. Steyaert et al. (2001) found a species-specific and highly dynamic vertical migration pattern controlled by the tidal rhythm and requiring particular sampling methods and species analyses. Phytal nematodes living on macrophytes in the Black Sea demonstrated a diurnal vertical cycle: during the daytime they lived on the thalli, and they entered the bottom at night (Kolesnikova et al. 1995). Reports on the role of nematodes as a nutritional source for macrofauna are rare. In tidal flats, young Carcinus feed intensively on the larger nematodes, e.g., Enoploides and Adoncholaimus. The latter, in turn, is necrophagous and is attracted by decomposing macrofauna. Small crustaceans, fish and carnivorous polychaetes feed on nematodes; for some small fish they are even the dominant prey (Feller and Coull 1995; Colombini et al. 1996) (see Sect. 9.4.2). Typically, however, a good proportion of the nematode biomass does not seem to be transferred to macrofaunal levels. Nematodes are linked to other meiobenthos through the omnivore/predator group, but the bulk of the detritivores seem more integrated into the short trophic loops of decomposers (see Sect. 9.4.1). The high species number of nematodes, even within small samples, makes them good indicators of disturbance- and pollution-induced changes, provided that detailed identification is feasible. Heip and Decraemer (1974) could relate a local decrease in diversity with the efflux of polluted river water. Conversely, after recovery from years of pollution, abundance, particularly that of species indicating organic enrichment, decreased, while diversity increased (Essink and Romeyn 1994). However, natural changes like sediment structure, oxygen supply, temperature, etc., may interfere and thus complicate such an interpretation of diversity indices. To further scrutinize this interrelationship, Schratzberger and Warwick (1998a,b) performed microcosm experiments with whole meiofaunal communities, comparing the reactions of the nematode assemblage to physical and eutrophication. A thorough statistical evaluation discriminated the community reaction: that in exposed sands was more resilient to physical disturbance than that from sheltered
5.6 Nemathelminthes: A Valid Taxon?
153
muds. Following the “intermediate disturbance hypothesis” (Connell and Slatyer 1977), the nematodes from mobile sands were said to be more adapted to physical disturbances (of limited intensity) than those from muds. On the other hand, the nematode assemblage in muddy sediments reacted less sensitively to a sudden organic enrichment than that from the sandy sites. To assess community changes or functional ecosystem responses without the tedious species identification, summative parameters have been introduced. Bongers (1990) and Bongers et al. (1991) contend that a genus- or family-level relation between “colonizers” and “persisters,” the “maturity index,” is indicative of disturbance- or pollution-induced changes. Schratzberger et al. (2007b; Recife) used a combined analysis of some independent data sets arising from large-scale field surveys and small-scale laboratory experiments to investigate the effects of seabed disturbance on nematode communities. Disturbance response was documented as a function of disturbance type (coastal development, dredged material disposal, bottom trawling, glacial fjord), origin (man-made, natural) and intensity (low, medium, high). Natural and human-induced seabed disturbance could not be clearly differentiated. Both generated changes in the taxonomic and functional diversities of their assemblages. The magnitude and direction of effects depended on the origin and nature of the stress-generating factors. The application of multivariate statistics often helps to reveal changes in community structure, even above the species level, and facilitates discriminating the effects of frequently combined factors, e.g., disturbance/pollution (see Schratzberger and Warwick 1999a; Somerfield and Clarke 1995). Far more simple is the nematode–copepod index (Raffaeli and Mason 1981), which is based on the observation that in general the more robust nature of nematodes compared to harpacticoids, especially the interstitial ones, leads to superior persistence in gradients with increasing pollution. Refinements to correct for changes caused by factors others than pollution, such as mere physical perturbation, have since been made and have limited the general range of applicability as an indicator of pollution (see Sect. 8.8). Modern genetic methods provide a new approach to tackling the problem of mass identification of nematodes. A DNA barcoding technique, based on nematode-specific primers of the 18S rRNA gene, enabled Bhadury et al. (2006a,b) to correctly assign over 97% of the specimens present. Such a rapid genetic identification method may also prove valuable (perhaps indispensable) to future ecological and environmental studies. The perils of DNA barcoding based on just a few genes (Will et al. 2005) can be attenuated by supplementing this approach with traditional taxonomic work. A second aspect that lends nematode studies importance in future research is their physiological potential. Since they strongly react to endocrine disrupting (ED) chemicals known from vertebrates, nematodes might become promising organisms for biomonitoring (Höss and Weltje 2007). Among the often exotic physiological pathways mainly those concerning anoxia have been briefly addressed above. A regulative capacity for salinity tolerance exists in tidal species (Forster 1998). Some species can produce unusual substances and use metabolic processes whose adaptive value is seldom initially obvious, e.g., the production of potent neurotoxins
154
5 Meiofauna Taxa: A Systematic Account
(Kogure et al. 1996), the glyoxylate cycle, the synthesis of cellulases and polyunsaturated fatty acids, amino acid synthesis from acetate and glucose, and other metabolic pathways that are barely known from metazoans. Owing to their species richness, in the future and with the aid of modern (molecular) methods (Badhury et al. 2006a,b), nematodes may play a prominent role in addressing questions about the impacts of pollution or biodiversity changes resulting from the impacts of humans on ecosystem processes. A tool capable of coping with the enormous diversity of species considering the restricted capacity of the few experienced taxonomists might be a summative DNA barcode database that can be aligned with known barcode sequences of identified and video-captured species, as suggested by Lambshead and his group for the deep-sea (Cook et al. 2005; Bhadury et al. 2006a,b; Lambshead and Packer 2006). One problem with this approach might, however, be the high metapopulation diversity known for some nematodes (Derycke et al. 2007a). Of particular relevance for coupling taxonomy with genetic identification is the newly developed short-term formalin fixation method (Bhadury et al. 2005), which allows both approaches to be applied to the same individual. Accessibility to molecular genetic analyses, even those of archived nematode material achieved by a new hot-lysis method (Bhadury et al. 2006c), might gain considerable custodial relevance in the future. Some nematode species with short generation times and high reproductive rates allow for inexpensive axenic culture and mass breeding, making them a potentially powerful tool in environmental and toxicological studies (Williams and Dusenbery 1990; Moens and Vincx 1998; for freshwater: Traunspurger and Drews 1996; Höss et al. 2006). So far, about 30 free-living species can be cultivated in the laboratory. The fact that the freshwater nematode Caenorhabditis elegans (Rhabditidae) belongs to this group gives it far-reaching potential, since this is, so far, one of the few invertebrate metazoans for which not only the genome but also the entire cell lineage has been determined. The extreme nematode diversity between species, even among populations, would be grossly underestimated by classical taxonomy alone, but is revealed by genetic analyses or an integrative approach that combines molecular data with multivariate morphometric methods and, where possible, interbreeding experiments (Derycke et al. 2007a; Derycke et al. 2008). The high number of unique genes and the thousands of novel protein families present in nematodes (Parkinson et al. 2004) argue for further exploration of their complex genetic background. Considering their ecological success (pollution!) and their impact on human and agricultural ecosystems (parasites!), this search could bring meiobenthic animals to the forefront of general awareness and thus yield unprecedented success. The aspects addressed here bring a whole new meaning to the statement of Heip et al. (1982), which was originally related to field ecology: free-living “nematodes are …… the most important” taxon in all marine sediments. More detailed reading: taxonomy and systematics, Heip et al. (1982); Keppner and Tarjan (1989); Platt and Warwick (1983, 1988), Warwick et al. (1998); computer-assisted key, Diederich et al. (2000); anatomy, Chitwood and Chitwood
5.6 Nemathelminthes: A Valid Taxon?
155
(1974); Bird and Bird (1991); biology, life history, Wharton (1986); Hopper and Meyers (1966); Gerlach and Schrage (1971); Tietjen and Lee (1977); Herman and Vranken (1988); ecology, Wieser (1959); Alongi and Tietjen (1980) Platt and Warwick (1980); Heip et al. (1982, 1985a); Jensen (1987a); Warwick (1989); Moens and Vincx (1997a); books and general reviews, Ferris and Ferris (1979); Nicholas (1984); Heip et al. (1982, 1985a); Gal’tsova (1991); Wharton (1986); Malakhov (1994); Lee (2002); monographs on freshwater nematodes, EyualemAbebe et al. 2006; Traunspurger (2000, 2002).
Box 5.5 Nematodes: Exuberant, Exotic, Everywhere Meiobenthic nematodes, the animal taxon with the greatest species richness in the benthic zone, have exploited every biotope in the aquatic regions of the Earth, from polar ice to deep-sea mud, from mountain streams to jungle phytotelmata. Nematodes also represent the majority of the meiobenthos in suboxic muds and even in hydrothermal vents with their extreme conditions. In the seemingly deserted beach sand, many thousand nematodes belonging to numerous species can live under each footprint. Each of them occupies its own ecological niche with different trophic requirements, life histories and sediment preferences. This bewildering multitude of species, of which only about 10% may have been scientifically described, are systematically grouped into two large subtaxa. Ecological grouping is often related to substrate type (e.g., sand, mud) and feeding mode (e.g., microvores, predators). Nematodes are mostly sediment-bound and less dispersive than other meiofaunal groups. Most of them are bacterivores, but in tidal flats a large proportion of them graze on microalgae, and uptake of dissolved organic matter seems common. This high ecological differentiation and specialized physiological capacity of nematode species, which apparently corresponds to a hitherto unknown genetic diversity, makes any grouping of them into larger categories a problem. This becomes evident upon listing the life history or production details of nematodes. However, their relevance calls for generalizations in order to better assess their overall role. A P/B ratio of about 10–15 seems applicable to both the average marine and freshwater nematode. Their frequent and speciose occurrence in polluted environments and their often specific reactions to single pollutants make nematodes valuable tools for pollution studies, even when macrofauna have disappeared. Attempts have been made to define and possibly cultivate indicator species for experimental work. In order to overcome the problems of species identification (although competent literature does exist!), superficial parameters such as the nematode/copepod ratio have been developed. While simple to use, even by the nonspecialist, they are controversial. Considering the fact that in most biotopes >80% of all meiofauna are nematodes, findings for nematodes are often representative of the whole meiofaunal community.
156
5.6.2
5 Meiofauna Taxa: A Systematic Account
Kinorhyncha
The vermiform body covered with cuticular rings led Dujardin (1851), in the first description of a kinorhynch, to classify it as being intermediate “between worms and crustaceans.” The striking “in-and-out movement” of the eversible anterior end of their body is so characteristic that the scientific name “Kinorhyncha” (“animals with a motile proboscis”) was later derived from it. Today, their affiliation with Loricifera and Priapulida is considered, resulting in the “Scalidophora” or “Cephalorhyncha” (see Neuhaus 1994, Neuhaus and Higgins 2002). This morphologically based monophyly is supported by recent molecular studies (Petrov and Vladychenskaya 2005, Mallat and Giribet 2006, but see Sørensen et al. 2008). Some 170 kinorhynch species are grouped into 18 genera belonging to two orders distinguished by differences in the number of plates in the “neck” segment (the second of 13 segments) of the cuticular rings, sometimes referred to as “zonites” (Fig. 5.23). The head can be withdrawn in the trunk and the plates of the neck then serve as a closing apparatus. The zonites carry spines with different arrangements and shapes: an important taxonomic feature. Juveniles
100 µm
Echinoderes (lateral view)
a
Echinoderes (dorsal view)
b
Pycnophyes (dorsal view)
Fig. 5.23a–b Some characteristic Kinorhyncha. a Natural appearance; b schematic graph showing structure of body plates. (After Higgins 1981, 1986)
5.6 Nemathelminthes: A Valid Taxon?
157
hatch with a reduced number of trunk segments; two of them develop during the series of six molting stages. The hatching juveniles have a different spine arrangement and structure to that observed in adults, which has caused some confusion in their taxonomy. All cuticular structures are shed during the juvenile molting phases. Cyclorhagida, mainly represented by the genus Echinoderes with 60 species; second zonite is covered with 14–16 plates or “scalids” that are round-to-oval in cross-section. Semnoderes is another better-known genus from sandier sediments. Homalorhagida, mainly the genus Pycnophyes with about 35 species; second zonite with only 6–8 scalids that are triangular in cross-section. Kinorhynchus and Pycnophyes are well-known genera typical for very fine sediments. With a few exceptions (see below), kinorhynchs are infrequently recorded in meiofaunal samples, although their unique movements make them immediately noticeable. Their tiny size (120–1,100 µm) and the inadequacies of common extraction methods (decanting, elutriation, flotation, see Sect. 3.2.2 and below) contribute to their scarcity. The stiff body is made flexible through the articulation of the cuticle in 13 annular zonites, which additionally attain a certain degree of elasticity in lateral and medioventral furrows, subdividing them into separate plates. Rhythmical inversion of the first zonite into the neck and trunk region creates the typical kinorhynch movement in which the scalids serve as anchoring devices so that the body can be slowly dragged forward. When retracted, the anterior scalids can completely cover the front end; when fully extended, the nine long jointed oral styles of the first zonite point forward. They surround a small oral cone that often remains retracted in specimens from live samples. The subdivision of the body, externally evident in the zonites, is also internally documented in the epidermis, musculature and nerve system. The animals have separate sexes; spermatophores have been observed on the females of some homalorhagids. Ecological aspects. Kinorhynchs are purely marine animals that occur mostly in muddy-to-fine sandy sediments from the eulittoral (e.g., Echinoderes coulli) down to the deep-sea, where they seem to occur with the highest diversities, with many new species discovered. They are also found in the phytal, and occasionally in coarse clean sand (e.g., Cateria with a very slender body in intertidal high-energy beaches). Shallow water forms probably feed mainly on diatoms, which correlates with the marked population peak recorded in summer for some species. The diatom cells are taken up with the help of scalids and spines and protractor muscles that move a sucking pharynx. In deeper bottoms, bacteria and detritus are probably ingested. Kinorhynchs are mostly found in the oxygenated surface layers at abundances in the range of about 15 specimens per 10 cm2, with a decreasing tendency toward the deeper layers. Densities of Echinoderes coulli—one of only a few species that tolerate brackish salinities—reached as high as 72 per 10 cm2 (Higgins and Fleeger
158
5 Meiofauna Taxa: A Systematic Account
1980). Especially in polluted fine sediments, up to 200 kinorhynchs 10 cm−2 can represent a substantial part of the meiofauna. High densities (45 ind./10 cm2) have also been recorded in polar regions (Neuhaus and Higgins 2002). Pfannkuche and Thiel (1987) found that in Antarctic bottoms at 400 m depth kinorhynchs represent 5–6% of all meiofauna, exhibiting abundances of >250 specimens per 10 cm2. In some studies of the Pacific deep-sea and in Antarctic bottoms, kinorhynchs ranked third out of all meiofaunal groups. The kinorhynch body is covered by a strongly water-repelling cuticle, so the animals tend to adhere to the water surface once in contact with air. Hence, the best way to obtain kinorhynchs is to use the “bubble and blot method,” because, even when preserved, kinorhynchs adhere to the surface film, and they can be removed from this with blotting paper. Decantation with subsequent inspection of the water surface in the jar is only adequate in coarse sand. For quantitative purposes more sophisticated methods (Higgins 1988; Sørensen and Pardos 2008) must be used. More detailed reading: anatomy, Kristensen and Higgins (1991); Neuhaus (1994); monographs, Remane (1936a); Adrianov and Malakhov (1994); Neuhaus and Higgins (2002); biology and identification key, Sørensen and Pardos (2008).
5.6.3
Priapulida
Half of all known priapulids (some 20 species in total) are of meiobenthic size; however, these are rather heterogeneous in appearance and have wide distributions. Certainly the best known of these is Tubiluchus corallicola, which is relatively frequent in sublittoral coralline sands in the Caribbean. T. troglodytes from an Italian cave has been found at densities of up to 80 ind. 10 cm−2 (Todaro and Shirley 2003). Other meiobenthic forms are the interstitial Meiopriapulus fijiensis, found in the eulittoral of Pacific islands, and Maccabaeus (= Chaetostephanus), a tubicolous form from muds in the Mediterranean. The larva of the common macrobenthic Priapulus caudatus can be encountered in fine sediments as temporary meiobenthos. The body of a priapulid is covered with a chitinous cuticle, often bearing tegumental spines, setae and papillae. In its anterior part the body is structured as an eversible and retractable proboscis, the “introvert,” which is studded with various diagnostically relevant teeth and scales, the “scalids” (Fig. 5.24). A post-anal extension of the body can occur as a long tail, e.g., in the troglobitic (cave-dwelling) Tubiluchus troglodytes, which apparently serves as an anchoring organ. Besides the well-developed dermal-muscular layer, two strong retractor muscle strands traverse the body and insert into the proboscis. It has been argued that the body cavities of priapulids, for instance in Meiopriapulus, are coelomic with a mesodermal lining. In Tubiluchus spp., which are sexually dimorphic, the males have a stronger ventral setation. Except for Meiopriapulus, which may undergo direct development
5.6 Nemathelminthes: A Valid Taxon?
159
0.5 mm
0.5 mm
Meiopriapulus fijiensis
Tubiluchus corallicola
Fig. 5.24 Characteristic meiobenthic Priapulida. (Various authors)
(Higgins and Storch 1991), priapulids have characteristic larvae that differ from the adults in that they have a fortified cuticle, the “lorica” (latin: “coat, case”) consisting of solid plates (dorsal, ventral, lateral) equipped with scalids (Fig. 5.24). The larvae of Tubiluchus spp. lack the tails characteristic of the adults. The chitinous cuticle of priapulids is periodically molted. The meiobenthic priapulids probably feed on bacterial films and other small organisms, which they scrape off or sieve out of the sediment with their relatively wide scoop- or comb-shaped anterior scalids. Priapulida is a very old group; fossilized members dating from the Cambrian have been found, and present-day members are often similar to fossils from the Burgess Shales. Their monophyletic character is not always plausible (Park et al. 2006), but their relations to the Kinorhyncha and especially to the Loricifera (see below) seem to be confirmed by molecular data and by the similarity of the priapulid larva to loriciferan adults (Warwick 2000). This is accentuated by the establishment of the taxon Scalidophora, which has strong morphological and molecular support (Garey 2002; Mallat and Giribet 2006; bus see Sørensen et al. 2008). More detailed reading: Maccabaeus, Por and Bromley (1974); Tubiluchus corallicola, Kirsteuer (1976); Meiopriapulus fijiensis, Morse (1981); ultrastructure, Higgins and Storch (1989); monograph, van der Land (1970)
160
5.6.4
5 Meiofauna Taxa: A Systematic Account
Loricifera
This nemathelminth phylum was first described by Kristensen (1983) on the basis of Nanaloricus mysticus found on the coast of Roscoff (France). In the meantime, these bizarre, minute animals have increased in the number of species described to about 25, grouped into several families, and at present about 300 species still await formal description (Gad, pers. comm.). While the first specimen described lived in shell gravel in only 25 m of water, today most species are found in the deep-sea (Gad 2004). Reasons for the late discovery of this group might be their minute size (250–300 µm), their similarity (when fixed) to contracted rotifers, and their rare occurrence, which, in turn, might be a methodological problem, since their viscid surface strongly adheres to sand grains. Loriciferans were first noted in 1974 by Higgins and in 1975 by Kristensen, and probably also by other meiobenthologists. However, it was not until fresh material became available that their unique structures separating them from other nemathelminths were recognized (Fig. 5.25). Despite their minute size, loriciferans have a considerable number of small cells (>10,000), in contrast with similarly small nematodes and rotifers. Today, sufficient specimens have been found to suggest a worldwide distribution for the taxon in both sandy and muddy sediments, not just in deep-sea bottoms. Shell hash and biogenic sands seem to be favorable habitats for loriciferans. Herman and Dahms (1992) even found loriciferans in the polar regions, and their occurrence has been reported in “Atalante,” an anoxic and sulfidic brine basin in the eastern Mediterranean. As undescribed new species are often found, the number of projected species will increase rapidly. The dorsoventrally flattened bodies of the Nanaloricidae are covered with a solid armor, the lorica, which is divided by longitudinal furrows into six plates. These are studded with about 230 spiny scalids which are often of bizarre shape and arranged in 9 transverse rows. The head region has a non-eversible but telescoping mouth cone, which can be retracted (e.g., when fixed) along a flexible neck into the trunk; this is reminiscent of certain rotifers. The anterior proboscis contains an internal stylet apparatus with a complicated triradiate muscular pharyngeal bulb. The Pliciloricidae (Pliciloricus) have a round body covered with a set of plates corresponding to lorica structures in the priapulid larva. The separate sexes of Loricifera differ in the structure of the scalids (Fig. 5.25). The complicated larva (“Higgins larva”) has numerous long spines around its head end, and, at least in Nanaloricus, a pair of toes at its hind end which serve as adhesive organs. Three leaf-shaped, locomotory flosculi may be present, which probably propel and push the animal forward through the sand. The larva passes through several molts, shedding the lorica. Deep-sea samples also harbored huge paedogenetic larvae (almost 850 mm; Gad 2005), other deep-sea species (e.g., Rugiloricus) have aberrant developmental cycles that include paedogenesis and parthenogenesis as well as gonochoristic development and internal fertilization of females.
5.6 Nemathelminthes: A Valid Taxon?
161
20 µm
50 µm
c
b a
40 µm
Fig. 5.25a–c The first described representative of Loricifera, Nanaloricus mysticus. a Female, ventral view; b female, front end with mouth opening; c larva with floscula. (After Kristensen 1984)
A systematic relationship of the Loricifera with the Kinorhyncha and Priapulida (“Scalidophora”) is well supported by morphological characters (Kristensen 2002; Mallat and Giribet 2006), and their positions within the Ecdysozoa were confirmed by analyses of 18S rRNA (Park et al. 2006). However, recent corresponding analyses with several loriciferan species (Sørensen et al. 2008) do not consistently support the close relationships mentioned above. The structural relationship, especially with the larva of priapulids, is striking (Warwick 2000), and a progenetic origin for loriciferans (from priapulids?) has found some molecular support. Fossil records also indicate a close connection between the scalidoporan taxa and emphasize that they were archaic nemathelmiths that existed already in the Lower Cambrian (Maas et al. 2007). Similarities with singular rotifers (lorica) or with tardigrades (stylet apparatus) are believed to be convergences; perhaps both are based of plesiomorphism (Ecdysozoa). With their extremely high cell numbers and numerous scalids, loriciferans can be considered to be the most morphologically complicated meiobenthic animals, despite their minute size. More detailed reading: first description of taxon, Kristensen (1983); taxonomy, anatomy, biology, Kristensen (1991a); phylogeny, Kristensen (2002).
162
5.6.5
5 Meiofauna Taxa: A Systematic Account
Gastrotricha
Gastrotrichs are all of meiobenthic size; indeed, they are among the smallest metazoans (some only 60 µm). However, the characteristic gliding that they perform on their ciliated ventral side (their scientific name means “gastric hairs”), their general body shape (head, thorax, tail, often biramous furca), and finally their armature of body spines and haptic tubes make most of them easy to recognize microscopically (Fig. 5.26). The haptic tubes are extensions of the body cuticle into stiff tubes that contain rich adhesive glands. With their steady gliding they differ from the more writhing gliding of turbellarians and gnathostomulids; another distinctive feature is their well-developed terminal mouth and anus. About 720 species have been described so far, which are classified into two orders genetically quite different orders with 15 families. From Northern Europe alone approximately 200 species are presently known. Identification is mostly based on the shape of the caudal furca, the arrangements and shapes of the scales, spines and hairs on the cuticular surface, the positions of the haptic tubes and the structure of the radial pharynx musculature. Macrodasyida (also Macrodasyoidea): discovered in the 1930s, this group contains about 250 relatively primitive but radiating marine species that exhibit hermaphroditic sexual reproduction. The numerous haptic tubes extend in a symmetric arrangement anteriorly, laterally and caudally. Macrodasys, Urodasys (with a tail), Turbanella, Cephalodasys (Fig. 5.26). Chaetonotida (also Chaetonotoidea): This mostly limnic group of approximately 450 species is more derived and exhibits less form variation; it contains only a few marine and brackish-water taxa such as Xenotrichula. The cuticle is covered by conspicuous spiny or shingle-shaped sculptures, often with a circumoral whirl of setae. There is only one pair of haptic glands on the furcate toes. Most species reproduce parthenogenetically, but hermaphroditism is also widespread (Weiss 2001). Chaetonotus, Neodasys, Lepidochaetus. In Xenotrichula and Neodasys, males have been found that inject their sperm in a complicated way into their female partners (similar to Rotatoria, see Sect. 5.4.2). Biological and ecological aspects. The cuticular hard structures (free of chitin) and the occasional development of vacuolized epidermal cells are often interpreted as providing protection against sediment agitation and pressure. With their very effective haptic tubes, glandular toes, and in some species haptic “girdles”, the animals can momentarily cling to sand grains (see Fig. 4.6). This reduces the success of extraction by decantation considerably unless anesthetization or a freshwater shock is applied. Besides the typical continuous gliding motion, locomotion by crawling and alternating adhesion using the front end and the haptic toes (looping) occurs in gastrotrichs as well as sinuous swimming. Characteristic for meiobenthic organisms, just a few, large eggs are produced, from which juveniles hatch without passing through a larval stage. The lifespan of gastrotrichs is only a few weeks. They are microphagous, feeding on bacteria, protozoans, etc.
Thaumastoderrma (Macrodasyoidea)
Fig. 5.26 Some representative Gastrotricha. (Various authors)
Turbanella
50 µm
50 µm
Diplodasys
50 µm
Chaetonotus (Chaetonotoidea)
50 µm
5.6 Nemathelminthes: A Valid Taxon? 163
164
5 Meiofauna Taxa: A Systematic Account
Gastrotrichs are occasionally abundant in meiofaunal samples. The marine species live preferably in the interstitial of fine sand enriched with detritus. Especially in the lower tidal to subtidal shore they can attain maximal densities of almost 7,000 per 10 cm2. On rare occasions, aggregations of more than 200 specimens per cm3 have been found in a beach on the west coast of the USA (Hummon 1972) and up to 400 specimens per cm3 in a tidal sand flat of the North Sea (Potel and Reise 1987). However, a rich gastrotrich fauna has also been recorded in areas of coarse sublittoral calcareous sand (shell hash) (Todaro 1998; Kristensen et al. 2007). In some habitats, including polar fine sand, gastrotrichs outnumbered many other meiofaunal groups (Huys et al. 1992). In tropical beaches, gastrotrichs were found to be particularly common after the decline of most meiofauna due to monsoon rains (Alongi 1990b). The first records from deep-sea hydrothermal vents (Desmodasys abyssalis) have been reported recently (Kieneke and Zekely 2007). Exposed beaches harbor just a few adapted species (e.g., Xenotrichula). In contrast to abundance, gastrotrich diversity can be remarkably high, particularly in coarse, calcareous sands. Up to 18 species have been stated to occur syntopically in a few square centimeters. Sea caves (anchihaline caves) have also been considered “hot spots” in gastrotrich diversity and endemism (Todaro et al. 2006). Differing ecological demands, even between some congeneric species, resulted in a typical vertical and horizontal distribution pattern along a beach (Ruppert 1977). Muds are seldom populated, since most gastrotrichs prefer oxygenated sediments. However, some species seem to possess adaptive metabolic pathways for a specialized thiobiotic life (Boaden 1974; Todaro et al. 2000). Some 40 specimens were found in samples from North Sea methane seeps (Giere, unpublished). The brackish-water and freshwater gastrotrichs prefer zones of submerged or decaying vegetation (periphyton), ephemeral pools and organic debris. Macrodasyida and Chaetonotida are commonly considered sister groups, but molecular analyses indicate that they are probably not monophyletic (Wirz et al. 1999; Zrzavy 2003). Within these subgroups, sperm ultrastructure seems to provide useful character sets for cladistic studies attempting to classify the species (Marotta et al. 2005). Gastrotrichs, with their multilayered cuticular fine structure, seem to be a monophyletic sister group of the Ecdysozoa Cycloneuralia (Schmidt-Rhaesa 2002; Zrzavy 2003), although other affiliations (e.g., with Plathyhelminthes) are also discussed. The trend for reducing body structures, concomitant with decreasing size, confuses conclusions about their phylogenetic position. Even the coelomate nature of their body cavities (mesothelial coelom present?) and the phylogenetic condition of the triradiate pharynx are disputed (Teuchert and Lappe 1980; Schmidt-Rhaesa 2002). In any case, the taxon Gastrotricha plays a key role in hypotheses regarding the deeper links between phyla. More detailed reading. taxonomy, Hummon (1971); Clausen 2000; structure, phylogeny, Rieger (1976); Teuchert (1977); Ruppert (1982), Schmidt-Rhaesa (2002); Zrzavy (2003); ecology, Schmidt and Teuchert (1969); physiology, Hummon (1974); zoogeography, Ruppert (1977); freshwater, Kisielewski (1990); Ricci and Balsamo (2000); monographs, Remane (1936a); D’Hondt (1971); electronic database, Hummon (2004); identification key, Todaro and Hummon (2008).
5.7 Tardigrada
165
Box 5.6 Bizarre, Spiny Dwarfs: The “Lesser-Known Taxa” A film producer asked to design an alien life form from outer space could do worse than to depict a scalid-armored loriciferan, a long-spined gastrotrich or a “mud dragon” (kinorhynch) plodding its way through the mud. It seems hard to imagine the evolutionary pathways that have led to these exotic forms. Unconventional and tiny (only a few priapulids belong to the macrobenthos), these rather unearthly groups have been termed “the lesserknown protostome taxa” (Garey 2002). The Loricifera, which were discovered only a few decades ago, are particularly complex in appearance and anatomical detail. The Scalidophora (Kinorhyncha, Priapulida, Loricifera) are armored with a chitinous cuticle (rings, plates, spines) that is regularly molted during growth. Especially in the loriciferans, this indirect development entails several bizarre larval stages and complex life cycles, often with progenesis. The anterior end of Scalidophora functions as a retractable “introvert” whose numerous scalids (in kinorhynchs and priapulids) scrape small food items, e.g., diatoms, bacteria, detritus while the animal digs through the mostly muddy sediment. In contrast, loriciferans, which tend to prefer sandy sediments, are equipped with a complex stylet apparatus suggesting the piercing and sucking of small cells, but, as yet, no live specimen has ever been observed. Discovered in the shallow sublittoral, the Scalidophora seem to attain the greatest diversity in deep-sea bottoms, with numerous species still awaiting description. The Gastrotricha are distinguished from the Scalidophora by their nonchitinous, flexible cuticles that allow continuous development without molts or larval stages. Their characteristic appearance, with numerous stiff spines and sticky tubes, makes them easily recognizable provided one is prepared to look for these tiny creatures that glide swiftly on their ventral ciliary soles across the substratum. In contrast to the Scalidophora, gastrotrichs can become quite abundant (e.g., in detritus-rich sands) if the right extraction technique (freshwater shock) is applied. They also exceed the other groups of spiny, bizarre dwarfs in species number.
5.7
Tardigrada
Tardigrada or “water bears” are another meiobenthic taxon that has gained considerable phylogenetic relevance in discussions about the link between the supertaxa Ecdysozoa and Arthropoda (Garey 2001; Nelson 2002; see below). The approximately 1,000 described species of tardigrades inhabit marine and freshwater habitats, sands and muds, the supralittoral and the deep-sea, and this wide ecological
166
5 Meiofauna Taxa: A Systematic Account
and geographical range indicates a distinct physiological diversification over a long phylogenetic period. The high group diversity (most genera have just a few species) also points to marked radiation. The first description of an interstitial tardigrade (Batillipes mirus) was provided in 1909, and new species are continuously being found and described today, especially those from marine habitats. The tiny sizes of all tardigrades (50–1,200 µm) often necessitate a double check at high magnification when sorting a sample. The five-segmented bodies of tardigrades are sometimes bizarre (Fig. 5.27). The head segment has numerous sense organs on cirri, and their arrangement is of diagnostic relevance. The following three thoracic segments and the caudal segment have flexible legs without arthropod joints, mostly ending in claws that sometimes come from long toes (subgroup Arthrotardigrada; Fig. 5.27). The telescopic extremities are retractable via muscles and eversible by the turgor pressure exerted by the mixocoelom; the last pair of legs often bears spectacular appendages (Fig. 5.27). The pharynx is equipped with a complicated, diagnostically important stylet apparatus for piercing. Its specific muscles act as a suction pump, similar to that seen in many nematodes. As in nematodes, circular musculature and ciliated epithelia are absent. Respiratory, circulatory and excretory (nephridial) systems are not developed. The phylum consists of two classes separated on the basis of morphological characters (the elusive Thermozodium is not considered here), each with representatives in marine and freshwater biotopes. Heterotardigrada: This more ancestral group, often considered the sister group for all other tardigrade taxa, consists of more than 300 species (all have head cirri) that are divided in two subgroups. In most heterotardigrade species the body is well segmented and often armored with plates. Most live in marine sediments, among algae and barnacle thickets. Arthrotardigrada: Usually have a median head cirrus, and the telescopic legs have toes from which claws and/or sucking discs arise. Stygarctus, Halechiniscus and Batillipes are common species from marine eulittoral sands; altogether there are presently 35 genera with 110 species. Echiniscoidea: The claws are positioned directly on the legs without toes. Most echiniscoid tardigrades are terrestrial forms, but Echiniscoides is a typical marine genus and the rather vermiform Carphania lives in hyporheic freshwater. Echiniscoid tardigrades survive periods of desiccation without any problems; there are presently 14 genera and 190 species in total. Eutardigrada: This more advanced group consists of 34 genera with more than 700 mostly limnetic and terrestrial species. They never have head cirri, their body is not armored, and the legs terminate directly in claws without toes. The mouth is surrounded by a whorl of lamellae. Macrobiotus, Hypsibius and Milnesium are characteristic representatives. Many occur in mosses and lichens, but they also live in mesopsammal habitats. Some have secondarily readapted to marine biotopes (e.g., Halobiotus).
5.7 Tardigrada
167
A
B
30 µm
50 µm
Batillipes pennaki Stygarctus
50 µm
Echiniscus
50 µm
50 µm
Neostygarctus acanthophorus
Tanarctus bubulubus
Fig. 5.27a–b Some characteristic or spectacular Tardigrada. a Batillipes: piercing stylet apparatus. b Batillipes: viscid digital end plate. (Various authors)
Biological and ecological aspects. The aquatic heterotardigrades survive adverse periods via cyst formation, while terrestrial Eutardigrades have developed a remarkable capacity to survive as an almost metabolically inert “tun stage.” Extreme desiccation and freezing (30 K!) and high doses of radiation do not seem to affect the “tun,” and may even greatly extend the individual’s lifetime. This extreme cryptobiosis” or “anhydrobiosis” is not developed in the aquatic species,
168
5 Meiofauna Taxa: A Systematic Account
which only have less elaborate resting stages. The cryptobiotic stages are often in tune with propagatory phases, enabling tardigrades to colonize ephemeral or isolated biotopes (for an actual review of cryptobiosis, see Wright 2001; Crowe et al. 2002). The developmental biology of tardigrades is complicated by their diverse reproductive strategies, at least in eutardigrades (Bertolani 2001). The marine species usually have separate sexes that are sometimes dimorphous; the iteroparous females produce only one egg at a time; males of marine species, in contrast, are semelparous and have a shorter maturation period than females. Freshwater species are usually gonochoristic and iteroparous in both sexes. But hermaphrodities, with self-fertilization, parthenogenesis and polyploid cytotypes, occur. These mechanisms favor together with cryptobiosis dispersal to unstable and isolated environments. Bisexual reproduction in tardigrades is through copulation; eggs are often unusually shaped and ornate. Mitosis is limited to a few tissues and possible only during the juvenile phase; in the adult body, which consists of relatively few cells, the cell number remains constant (eutelic) in most organs. All hard structures (claws, stylet apparatus) consist of the chitinous material that covers the whole body as a cuticle, and so they have to be molted as the animal grows (which occurs continuously throughout its lifetime). Juvenile stages have fewer legs and claws than the adults. The developmental cycle of one of the most common interstitial tardigrades, Batillipes pennaki, is bivoltine with (in Italy) maxima in spring and fall, which are probably related to favorable food conditions; in boreal climates its population peaks in winter. During the life cycle of Halobiotus cyclomorphotic changes (i.e., regularly occurring alterations of certain body structures within a population) in the stylet apparatus have been noticed occasionally and in tune with the seasons (Kristensen 1982). The average lifespan is some months to a few years, but in the cryptobiotic stage they can survive for more than 100 years (Nelson and Higgins 1990). Tardigrades mostly feed on the contents of bacteria, fungi and algal cells (both macroalgae and diatoms), which they pierce with their stylet apparatus. A few species are carnivorous or detritivorous. Macrobiotus consumed in feeding experiments up to 100 nematodes in 4 h (Hohberg and Traunspurger 2005). However, feeding types related to specific mouth tube armatures, as in nematodes, could not be discerned. Many aquatic species are found in mosses, among algae, in seagrass meadows, and even in floating Sargassum. These phytal forms use their legs and claws to cling to the fine thalli. Another preferred habitat where tardigrades are regularly encountered is detritus-rich crevices among barnacle epigrowth. Sandy biotopes represent the other major habitat for (marine) tardigrades. Here, they characteristically and slowly crawl on sand grains (the group name is derived from the Latin for “slow walkers”). Batillipes, with its viscid digital sucker plates instead of claws, is fairly active, and “runs” busily up and down the grains. Halechiniscus, with its clumsy, barrel-shaped body, pushes its way through the finer sand. In the mud-inhabiting Coronarctus, found at greater depths and even in the deep-sea, the body is more worm-like and the legs are short.
5.7 Tardigrada
169
Similar to other meiobenthic groups, calcareous biogenic sands are populated by a highly diverse tardigrade fauna, often with particularly bizarre body appendages: 35 mostly new species were found in samples from the Faroe Bank (260 m water depth), i.e., 20% of all known marine species (Hansen et al. 2001). Recent data from the shallow calcareous Meloria Shoals in the Mediterranean underlined this trend. While in coarse sand most species live interstitially, the tardigrade community in fine sand with detritus tends to live epibenthically and it becomes less diverse. Apparently, sediment porosity and structure are relevant distributional factors. In sediments with a good oxygen supply (high-energy beaches), the vertical occurrence of tardigrades down to 150 cm and deeper is not unusual, but the food source at these depths remains unknown. The coastal zone of the sea is often considered the original habitat for primitive tardigrades, from which they entered both the freshwater and terrestrial biotopes as well as the deeper marine bottoms. Depressed forms with flat, extending body plates are considered particularly well adapted for passive transport by clinging to sand grains (Grimaldi de Zio et al. 1983). Despite the limited active distribution potential of tardigrades, many tardigrade genera and species are widely distributed, and some of them are cosmopolitans that occur in all climates. Their extreme survival capacities at various resting stages may have contributed to this ubiquity. A typical distributional profile along the sea bottom (Fig. 5.28) will have in the eulittoral the most eurytopic forms (Stygarctus, Batillipes) that are common in various types of sediment. Genera occurring in the sublittoral can be differentiated in mud- and sand dwellers, and rather specific tardigrade species of very limited distribution live in the deep-sea ooze. However, even genera that occur ubiquitously seem to have a well-developed potential to recognize subtle differences in habitat conditions, resulting in a remarkably heterogeneous colonization with distinct population centers in a beach (Pollock 1970). The overall abundance of tardigrades, even in favorable sites, is rarely very high. For marine forms, densities of >500 ind. per 100 cm3 of sand or 285 ind. per 10 cm2 must be considered unusually high. Patches with up to 3,500 B. pennaki per 100 cm3 (= 32,500 under 100 cm2) have been counted in the low water lines of a Portuguese beach (Thiermann, unpubl.), McGinty and Higgins (1968) decanted over 3,000 B. mirus from 100 cm3 of estuarine sand in Chesapeake Bay, USA. Especially in freshwater ecosystems, the presence of tardigrades in high quantities can reach ecological relevance. In the sandy shores of a freshwater lake in Brazil, Hypsibiidae were found to dominate the meiofauna with mean densities of 1,800 ind. 10 cm−2 (Flach et al. 2007)! Tardigrade abundance is often underestimated due to inadequate sorting procedures and methodological shortcomings. A good method for obtaining (interstitial) tardigrades fairly quantitatively is the “freshwater shock” decantation method, which causes them to release their grip on sand grains (see Sect. 3.2.2). Tardigrades are an old group; fossils attributed to their stem group date from the Middle Cambrian (Müller et al. 1995). The oldest true tardigrade was found embedded in amber that is 92 million years old. Earlier notions saw them affiliated
170
5 Meiofauna Taxa: A Systematic Account Halechiniscus Batillipes
Florarctus
Parastygarctus 0m sand
200 m
coarse detritus
Coronarctus
4000 m mud
Fig. 5.28 A horizontal distributional pattern of tardigrades along the sea floor. (After Grimaldi de Zio et al. 1984)
with the spiralians (Eibye-Jacobsen 1997); their segmented bodies, chitinous cuticula, metameric nerve systems and characteristic legs indicating some linkage to the “annelid–arthropod line of development.” Today, the annelid linkage seems more and more untenable, since the concept of the supertaxon “Ecdysozoa” means that Tardigrada is linked as “Proarthropoda” to both the Nematoida and the Arthropoda, an affiliation repeatedly confirmed both in morphological and molecular studies (Garey 2002; Petrov and Vladychenskaya 2005; Mallat and Giribet 2006). This could explain some parallels with nematoid structures (pharyngeal pump, eutelic fixation of cell number, lack of ciliated epithelia and circular musculature), features that are also considered convergences. More detailed reading: Marine species, taxonomy, identification and faunistics, Kristensen and Higgins (1984) Guidetti and Bertolani 2005; phylogeny, Grimaldi de Zio et al. (1987); biology, ecology, De Zio and Grimaldi (1966); Grimaldi de Zio and D’Addabbo Gallo (1975); Grimaldi de Zio et al. (1983); Renaud-Mornant (1982); freshwater species, Iharos (1975); Schuster et al. (1980); cryptobiosis, Wright (2001); short review, Nelson (2002).
5.8 Crustacea
171
Box 5.7 The Glamour of Clumsy Fellows: A Digression into the Psychology of Meiobenthos Any student course on meiofauna will prove that tardigrades are the stars in the wide world of meiobenthic animals. What is it that makes them so fascinating? Their tiny size? No—they are not any smaller than many rotifers or gastrotrichs! Their bizarre shapes, with floppy wings, decorative spines and baroque plates? No—kinorhynchs or priapulids have much more complex and ornamented scalid patterns! Their clumsy movements? Again, no—a kinorhynch’s method of pulling its body through the mud is more unusual! So, what makes “water bears” so different? Imagine: a spiny head, a flattened body, often armored with plates, strong claws on their four paired legs; all this would be repulsive to us if the creatures reacted quickly like a spider or an ant. However, these cuddly tardigrades set their stumped legs one after the other, as if pondering with each step where to go, they have to paddle hard to get a bit further. Many species have tiny eyes, useless for effective orientation. Altogether, they look touchingly helpless with their podgy bodies. Poor guys! In fact, tardigrades are real womanizers that primarily charm the female students. However, they have their male admirers too - they are record holders: They have resting stages to preserve their lives under most extreme conditions; their unusual modes of reproduction (progenesis, parthenogenesis) enable tardigrades to live in even the remotest of places. No wonder they have survived 500 million years—tough guys!
5.8
Crustacea
Nematodes aside, Crustacea is the meiobenthic group that dominates in abundance and species richness in most meiobenthic samples. There are, however, many taxonomic lines in crustaceans, with groups often consisting of just a few isolated species, and various species living in refuge habitats (e.g., groundwater, caves). Mainly because of their high phylogenetic and zoogeographic significance are these rarer forms included here too. Progenesis is of particular importance in the development of meiobenthic crustacean groups. They seem prone to this developmental abbreviation, which, on the evolutionary level, may have led to taxa such as Ostracoda and Anomopoda (Cladocera), which have a rapid generation turnover. In these groups as well as in the species-rich copepod suborder Harpacticoida, most are of meiobenthic size. However, there are other normally macrobenthic crustacean groups, e.g., isopods and amphipods, in which structural deviations and miniaturization have adapted a few aberrant taxa to the requirements of meiobenthic life. Many “microcrustaceans” are freshwater forms and are well reviewed in treatises by Dole-Olivier et al. (2000) and Galassi et al. (2002).
172
5 Meiofauna Taxa: A Systematic Account
Traditionally, the Crustacea as Arthropoda have been linked to the Annelida forming the supertaxon Articulata. With the advance of molecular methods, this unit has become controversial, while the concept of Ecdysozoa has gained increasing support (see Sect. 5.6; Giribet et al. 2005; Jenner and Scholz 2005). This concept would make Nematodes and Scalidophora related to Arthropoda, while the lophotrochozoan annelids are the sister group of Ecdysozoa and are thus only distant relatives to the crustaceans. The Crustacea seem to be a paraphyletic unit within the monophyletic Arthropoda (Garey 2001, Babbitt and Patel 2005; Mallett & Giribet 2006). Within the Crustracea the Malacostraca are apparently monophyletic while in other groups the relationships are inconsistent and debated.
5.8.1
Cephalocarida
This group, first described by Sanders (1955), presently consists of just five genera and a few species. Hutchinsoniella (Fig. 5.29), the first genus, was found in the
500 µm
Hutchinsoniella macracantha
Lightiella sp.
500 µm
Fig. 5.29 Some Cephalocarida. (Various authors)
5.8 Crustacea
173
sublittoral down to 300 m depth off Long Island (East coast of America), and then Lightiella was discovered in sands off the American Pacific coast, in other Pacific areas, in the Caribbean, and off Venezuela. Sandersiella occurs in mud bottoms off Japan, Chiltoniella in New Zealand, and Hampsonellus off Brazil. These small animals (about 2–3 mm in body length) have a slender body with numerous, homogeneous segments (nine thoracic and ten abdominal metamers). They lack a carapace and a caudal furca. Cephalocarida have articulate legs (phyllopodia) and a maxilla that is morphologically identical to the thoracic legs; compound eyes have apparently been reduced. Representatives of the rare Hutchinsoniella genus are found in the superficial detrital layers of sea bottoms that are flocculent enough to allow the animals to filter food from them when they are moving (the rows of legs form suction chambers). Lightiella, although oxybiotic, regularly occurs in the Caribbean in the deeper sand layers below the RPD layer (De Troch et al. 2000; Schiemer and Ott 2001). Few other biological details are known: these simultaneous hermaphrodites produce just two large eggs from which metanauplii hatch. A disjunct geographical distribution, a long nerve cord and heart with serially arranged ostia, and the primitive structure of the maxilla characterize this order as being an early offshoot from archaic crustaceans, i.e., a “living fossil,” but one without a fossil record. More detailed reading: Sanders (1959).
5.8.2
Anostraca: Anomopoda (“Cladocera”; “Branchiopoda”)
“Cladocera” is actually an artificial, polyphyletic unit that should be separated into several natural taxa. The approximately 270 species of typically meiobenthic “water fleas” are today grouped under one subgroup “Anomopoda” within the Anostraca, but as with the term “turbellarians,” the ecological literature often still refers to them as “cladocerans.” Also the frequently used term “Branchiopoda” denotes a colloquial, not phylogenetically based taxon. Cladocerans live in all kinds of freshwaters, from puddles to lakes, with most of them belonging to the Chydoridae (Fig. 5.30). The phytal species climb and swim among macrophytes (Eurycercus, Chydorus), the epibenthic forms dig through the surfaces of muddy bottoms (Ilyocryptus, Macrothrix, Pleuroxus), and some few species even represent typical subterranean stygobiota (Alona spp.). Although some adult cladocerans exceed the meiobenthic size range, the bulk of the benthic forms do not surpass 1 mm in length or represent small pre-reproductive instars. Diagnostic features include the general shape and sculpture of the head (often in the shape of a helmet) and the wide bivalved carapace (the “shells”) which completely encloses the unsegmented and short thorax, certain pores on the headshield, and the setation and spines of the terminal claws which are usually bent ventrally. All cladocerans molt five times before reaching maturity, thereby shedding their carapace. The populations consist almost exclusively of females, which
174
5 Meiofauna Taxa: A Systematic Account
IIyocryptus sordidus
Pleuroxus
Alona
Chydorus
Fig. 5.30 Some representative meiobenthic Anomopoda; 0.5–1 mm in length (Wesenberg-Lund 1939)
predominantly reproduce parthenogenetically by diploid eggs. These are brooded underneath the carapace. Each egg deposition is preceeded by a molt, and the development of the juveniles hatching from these “summer eggs” is very rapid. As the living conditions deteriorate (e.g., the temperature drops) males are produced, and, after normal meiotic division, haploid eggs that require fertilization by the males. The resulting thick-shelled resting eggs are stored in a chamber in the carapace, the ephippium. Here they can survive extremes of temperature and salinity for periods as long as years. During this resting phase of embryonic development, the eggs can also dry out completely. Moreover, the ephippium is an excellent raft for distribution, since it drifts in the water or adheres to plants and birds’ legs. Depending on the physiographic situation, the production of fertilized eggs occurs in winter (“winter eggs”). However, in species typical of ephemeral pools, which repeatedly dry up and then fill up again after new rain falls, several sexual cycles occur independent of the season. This “heterogonous” generation cycle, resembling that of rotifers, ensures genetic mixis and survival (of eggs) during adverse life conditions on the one hand, and, through its parthenogenetic phase, a huge potential to rapidly populate new areas and isolated ponds on the other. As with many meiofauna, progenesis plays a major role in the phylogeny of this taxon. The meiobenthic cladocerans dig, rake and climb with their large and muscular locomotory antennae and thoracic appendages rather than using them as swimming and filtering legs. The number of filter chambers is correspondingly reduced to just one or two. The terminal claws can also be used to scrape the substrate. Benthic cladocerans feed on small algal and detrital particles.
5.8 Crustacea
175
The highest diversity of meiobenthic cladocerans occurs in the phytal zone, where the habitat complexity is greatest. Due to their parthenogenetic development, they can develop dense populations in almost all types of freshwater (except streams). Abundances of 1 million per m2 substrate are not unusual (Whiteside et al. 1978). Considering their high consumption of bacteria, algae and detritus, Anomopoda represent members of the limnetic meiobenthos. They are also important as food for macrofauna. Anomopod-related fossils dating as far back as the mid-Cambrian have been found (Walossek 1996). One famous fossil related to the Anostraca is the Devonian Lepidocaris found in Scotland. More detailed reading (often still under the term “Cladocera”): taxonomy, Frey (1987); ecology, Goulden (1971); Whiteside et al. (1978); review, Rundle et al. (2002).
5.8.3
Ostracoda
Ostracoda are one of the most speciose crustacean groups. The bivalved carapace that characterizes them has been well documented in fossils that date from the Cambrian. Major taxonomic lineages seem to have diverged by then (Yamaguchi and Endo 2003). Today, > 60,000 extant and fossil species have been described, and the number of recent species is estimated to range between 5,000 and 15,000, with some authors even reporting 30,000, whereof about 2.000 species live in freshwater habitats. Important as stratigraphic and palaeoenvironmental indicators (De Deckker and Forester 1988), ostracods have been classified in a “geological system” based only on hard structures. In contrast, the zoological system of recent taxa is based on both hard structures and soft-body features; modern systematists try to integrate both systems (Hartmann and Puri 1974; Horne et al. 2005). For more detailed accounts on the limnetic ostracod groups, the reader is referred to the review on freshwater meiobenthos by Rundle et al. (2002) and references therein. Ostracods were first described in 1722 as “bivalved insects.” Their short bodies, consisting of only a few segments, are completely enclosed into the two valves of a carapace joined by a complicated hinge that is of taxonomic relevance (Fig. 5.31). Other diagnostic features used for identification are the shapes and sculptures of the shells, the structure of the single closing muscle and its scars on the shell, and the position of the spines on the short legs. The problems of basing the taxonomy on carapace structures become obvious when we consider that almost identical valve types have developed convergently in several independent lines, and that shell structures are subject to ontogenetic changes and sexual dimorphism. Moreover, shells can vary in their ornamentation and their patterns of lobes, depending on environmental conditions (e.g., salinity and chemical composition of the ambient water). The application of molecular studies has proven useful in this case, particularly because of the high degree of geographical and obligate parthenogenesis in many ostracods (Horne and Martens 1994).
176
5 Meiofauna Taxa: A Systematic Account
Candona candida
Cyprideis torosa
Cytherura sella Cypridopsis vidua
Iliocypris gibba Actinocythereis sp.
Polycope sp.
Psammocythere sp.
Fig. 5.31 Some representative meiobenthic Ostracoda; about 1 mm in length. (Various authors)
There are two relevant subgroups to mention here: Podocopa: Ord. Podocopida: Many marine (Cytheracea, Darwinuloidea), but also freshwater and brackish-water forms, prevailingly benthic. The ventral edge of the shell has a straight-to-concave contour; brackish water forms (Cypridacea) often have marked shell sculptures (Fig. 5.31) but no rostral incisure. The small species of meiobenthic size often show reduction phenomena like the lack of abdominal furca, heart and compound eyes. Progenesis is fairly common (e.g., Nannocandona). Representatives include Cyprideis, Paradoxostoma (interstitial; phytal), Bairdiidae (phytal), Microcythere (only 200 µm long),
5.8 Crustacea
177
Leptocythere, Limnocythere, Cythere, Xestoleberis (both interstitial and phytal species, also in brackish and freshwater water), Cytherissa (limnetic). Ord. Platycopida: Exclusively marine, with unequal, ovate valves, small body, most live in deeper waters. Myodocopa: Exclusively marine, valves are weakly calcified and thus are poorly documented in the fossil record. The ventral edge of the shell has a convex contour; there is often (but not in Cladocopida) a rostral incisure in the anterior edge of the shell. The small and roundish Cladocopida live in the marine interstitial. Cladocopa; Polycope, Polycopsis (Fig. 5.31). Biological and ecological aspects. Ostracods are most successful crustaceans ecologically and from an evolutionary perspective. This success is probably related to their wide variety of reproductive models for studies on the evolution of sexuality. One species can have bisexual populations in some areas while other lineages in other areas are parthenogenetic (e.g., within Cyprididae). Members of the Darwinuloidea, which have been exclusively parthenogenetic since Mesozoic times apparently, have an amazingly low genetic diversity. The genetically homogenizing mechanism in this long unisexual sequence is unclear. In contrast, the Cytherididae and Candonidae are holomictic. These mictic freshwater species often have amazingly large spermatozoa that are transferred by large copulatory organs. Eggs can become fixed to the sediment by adhesive fibers. The nauplius already has a two-shelled carapace, and after 5–8 molts the adult stage is reached. The eggs and the first naupliar stage are sometimes brooded between the shells. Ostracods are quite long-lived; many live for several years. In the sea, ostracods regularly populate lower tidal flats and sublittoral sands, where they live in fine or coarse sands; fine-grained sediments seem to support the largest populations. They can be particularly diverse and abundant in calcareous or coralline sands. They vigorously burrow and push through the sediment with jerky motions of their strong legs, which are often armed with claw-like setae. Occasionally is there a clear adaptive correlation between the structural or biological features of ostracods and their habitat. There are only 60–70 species that appear to be structurally adapted for life in the interstitials of sands. These have small bodies (only 0.2 × 0.3 mm) possessing a smooth carapace with an elongated, almost cigar-like or conical shape (Fig. 5.31). The shell may be laterally compressed (Paradoxostoma) or ventrally flattened (Xestoleberidae), and the segment number may be reduced and the limbs simplified. While crawling on the grain surface or swimming in the interstitial system voids, they can immediately clutch the grains with adhesive fibers produced by a spinneret gland complex and released from openings in the setae of the antenna. The loss of eyes and pigments in the more endobenthic or interstitial Podocopa or the reduction in the number of segments (Polycopidae) can be interpreted as adaptations to the interstitial of coarse sands (Hartmann 1973). The phytal zone (see Sect. 8.5) is another preferred habitat that is densely populated by sediment dwellers and typically adapted phytal species, e.g., the Bairdiidae with their specialized mouth parts and hairy carapaces. The deep-sea harbors ostracods in silty sediments (Dinet at al. 1988) as well as in hydrothermal vents (Sect. 8.3; Fricke et al. 1989; Van Harten 1992). Various substrates in the Antarctic sublittoral harbored a rich and diverse ostracod fauna (Hartmann 1990).
178
5 Meiofauna Taxa: A Systematic Account
In favorable shallow bottoms, ostracods can attain abundances of well over 200 specimens per 10 cm2 (when sorting, it is difficult to determine the live individuals from the empty shells). In the beaches of the Galápagos Islands, ostracods were found (Westheide 1991) to rank second after the nematodes and sometimes even to dominate (2,850 specimens 100 cm−3). In the Baltic Sea a few ostracod species represent most of the total meiobenthic biomass (Modig et al. 2000). Through their vigorous motions they are the main bioturbators among meiofauna, and have a considerable impact on the sediment’s fabric and geochemistry. While ostracods, like harpacticoids, are generally restricted to the well-oxidized surficial sediment layers, Cyprideis torosa survives highly sulfidic conditions (Jahn et al. 1996). Since the salinity is the dominant influence for most species, coastal areas with a brackish water gradient often have a marked pattern of species that are differently adapted to haline ranges. In typical brackish-water species (e.g., Elofsonia baltica), the shells are often rather thin. In freshwater, ostracods occur in almost every habitat, from high mountain lakes to ground and cave waters, and some species can also live in moist mosses and litter. Ostracods are typical dwellers of springs with elevated temperatures and ionic contents. Limnetic ostracods are often adapted to survive adverse conditions. Darwinuloidea can hibernate for months in a state of torpidity, and the Cytheroidea and Cypridoidea produce resting eggs that are resistant to desiccation and freezing. The water or air transport of these resting stages by other animals (e.g., birds) may have contributed to the wide distribution of some brackish-water and freshwater species. A rich endemic ostracod fauna with large species flocks evolved in ancient lakes (Cytherissa in Lake Baikal, Cyprideis in Lake Tanganyika). Among the ostracods there are scavengers that feed on debris and perhaps carrion, microphages that ingest silt containing bacteria and microalgae, and herbivores that devour diatoms; some of these have piercing mouth parts that suck plant (and animal?) tissues (Paradoxostoma spp). Candona neglecta is known to graze intensively on diatom phytodetritus in the Baltic, and it accounts for almost half of the total food uptake by meiofauna (Ólafsson et al. 1999). In many interstitial forms (e.g., Xestoleberidae), the second antennae are equipped with a complex of spinneret glands that release adhesive fibers from openings in long setae. In some species the detrital food material is fixed for ingestion by secreted fibers. Locally, Ostracoda feeding on settled planktonic diatoms seems to be an important benthopelagic link (Ólafsson et al. 1999). However, generally speaking, the role of ostracods in the marine food web is little understood. They seem to be the preferred prey mainly of small fish (Yozzo and Smith 1995), as well as halacarid mites and some polychaete worms. The lacustrine Cytherissa lacustris and Darwinula stevensoni are among the best-known ostracods and are often investigated in studies of the limnetic meiobenthos. The latter species is of particular interest for genetic studies on its evolutionary long-lasting parthenogenesis (Griffiths and Butlin 1994; Van Doninck et al. 2004). Ostracods are considered to have originated very early by progenesis from unknown crustacean ancestors. Loss of mictic reproduction and sexuality had already developed by the early Mesozoic (Martens et al. 2003). Interstitial ostracods
5.8 Crustacea
179
from the Cretaceous that closely resemble extant species indicate that morphological stasis has occurred in some lineages, in interesting contrast to the flexibility of other taxa. In spite of their rich fossil record, it is uncertain whether Ostracoda are monophyletic (Horne et al. 2005), and the relations between major subgroups are unresolved; the position of ostracods in relation to other crustacean classes are also disputed. More detailed reading: taxonomy, Elofson (1941); Hartmann and Puri (1974); Athersuch et al. 1989; monograph, Hartmann (1966–1975); freshwater, Meisch (2000); Henderson (1990); Maddocks (1992); Rundle et al. (2002); reproduction, Horne and Martens (1994); Horne et al. 2005; review, De Deckker et al. 1988; Ikeya et al. (2005)
Box 5.8 Bivalved Crustaceans: Sheltered, Adapted, Proliferous Cladocerans (Anomopoda) and ostracods are unrelated crustacean groups, but they do have some traits in common. In both the “water fleas” and the “water shrimps,” a short body of only a few segments is enclosed in two protective shells. The biological and (in ostracods) the palaeontological and even economic consequences of this are considerable. The protection from shells enables brooding, shelter for the offspring, survival in dry periods, and air/water transport over long distances. Parthenogenetic reproduction allows for the exponential growth of populations without any mating partners, even in remote and pristine habitats. Calcification of the shells (in many limnetic ostracods) has resulted in long-term fossilization and paleoenvironmental documentation of geological strata, a feature that is intensively used by oil drilling companies. Thus, the meiobenthic ostracods are probably the only meiobenthic taxon in which more fossilized than recent species are known, and for which an industry employs specialists. These small crustacean groups provide a variety of interesting scientific problems, such as the susceptibility of their reproductive modes to environmental cues. For example, at the onset of adverse conditions (e.g., winter temperatures), the series of parthenogenetic generations is interrupted and bisexual reproduction kicks in. In cladocerans this leads to a complex, seasonally tuned heterogonic generation cycle. Discussions about the origins and adaptive advantages of sexuality vs. asexual/unisexual reproduction in animals cannot exclude the darwinuloid ostracods, in which parthenogenesis without the existence of males has occurred since the Mesozoic without any apparent lack of adaptive potential. In other ostracod groups, (meta)populations of the same species can follow either a sexual or a parthenogenetic lineage, depending on unknown cues. Can meiofauna provide models for understanding one of the fundamentals of biology—sexuality?
180
5.8.4
5 Meiofauna Taxa: A Systematic Account
Mystacocarida
This entirely interstitial, marine group consists of some 12 species in only two genera (Derocheilocaris and Ctenocheilocaris). Although they occur locally in considerable numbers in well-accessible beaches, the group is scientifically young. They were originally found on the New England coast of America (D. typicus, Pennak and Zinn 1943), and on the French Mediterranean beaches north of Banyuls-sur-Mer (D. remanei, Delamare-Deboutteville and Chappuis 1951). Despite the close structural relations that bind the animals to the interstitial system and the lack of any distributional stages, populations of D. typicus live on conspecifically on both sides of the Atlantic. Other species have also been discovered along the African Atlantic coast. Ctenocheilocaris has been found along both Atlantic sides and also on Pacific beaches of South America and Australia. Hence, an assumed amphiatlantic distribution pattern can no longer be maintained. All mystacocarids are small (max. 0.5 mm) crustaceans with 11 free body segments that give the vermiform body high flexibility (Fig. 5.32). Equipped with long antennae, a primitively branched mandible and conspicuous claw-like caudal rami, a certain degree of convergence with some interstitial harpacticoid copepods is only superficial.
Fig. 5.32 The mystacocarid Derocheilocaris. (Left) body structure; (right) two animals in their natural environment. (After Lombardi and Ruppert 1982)
5.8 Crustacea
181
In a detailed study, Lombardi and Ruppert (1982) emphasized the high degree of specialization in the swift movements (fifty times their body length = 2.5 cm min−1) of these interstitial animals (Fig. 5.32). The long exopodites of the antenna and the mandible are held upwards, supporting and pushing the animal dorsally; their endopodite counterparts have the same function ventrally. All other appendages are non-locomotory; the antennules are merely tactile and the caudal claws are perhaps useful during copulation. Thus, locomotion is fully dependent upon an interstitial void system of a given width, while in other surroundings the animals move helplessly. Their occurrence is local and patchy, but in favorable beaches along the Portuguese Atlantic coast, they have been ranked third in overall meiofauna abundance (70 ind. 100 cm−3, Thiermann, unpubl.). They apparently prefer water-unsaturated median sand above the (high) water line that is poor in detritus. Here they are most common in the permanently moist subsurface layers above the groundwater horizon, often living under low-oxic conditions (see Villora-Moreno 1996a). Only rarely have they been found sublittorally. Their occurrence high up on the shore relates to their euryhaline and eurythermal nature (down to 10 PSU; 7–25 °C; Jansson 1966b; Kraus and Found 1975). The reduction of the vascular and respiratory organs, of egg numbers and eyes can be interpreted as consequences of the tiny body size. Secondary adaptations to life in the interstitial system seem to be (a) specialized locomotory organs, (b) development of antennules used as long tactile organs, and (c) abbreviated ontogenesis, omitting the naupliar stage. Conservative features reflecting a long and isolated phylogenesis are the biramous mandibles that are used for their original function, locomotion, as well as primitive traits in the nervous and cerebral system. These features justify the separation of Mystacocarida as a separate albeit small crustacean order, perhaps of progenetic origin. More detailed reading: bibliography, Zinn et al. 1982.
5.8.5
Copepoda: Harpacticoida
Aside from nematodes, harpacticoids are usually the most abundant meiobenthic animals in marine samples. In some tropical beaches they have been found to attain a higher relative share (35%) of the total meiofauna than nematodes (30%). Within the probably monophyletic suborder there are 55 families, of which approximately 17 are species-rich and relevant to this account (Boxshall and Halsey 2004; Wells 2007). 4,000–4,500 meiobenthic species have been described, but Huys et al. (1996) estimate that 30,000 more harpacticoid species exist. From the North Sea and around the British Isles, an area usually considered well investigated, Huys et al. (1996) listed 800 species, many of them only recently discovered. About 950 species belonging to 13 families have invaded freshwater biotopes, and the Parastenocaridae have evolved into typical groundwater forms. The slender, usually rather linear bodies of harpacticoids range in length from 0.2 to 2.5 mm; the thorax is set off only a little from the abdomen. Harpacticoid
182
5 Meiofauna Taxa: A Systematic Account
Calanoida Harpacticoida
Cyclopoida
Fig. 5.33 The general body structure of the three suborders of Copepoda (after Siewing 1985)
copepods are distinguished from calanoid and cyclopoid copepods by their short antennules (Fig. 5.33). They are distinguished from calanoids by the position of the articulation between the metasome and the urosome: in calanoids this lies between the last thoracic and the first abdominal segment, but in harpacticoids the last thoracic segment, i.e., the genital segment, is included in the urosome. Since the designation of body parts tends to vary in the literature, it is advisable to carefully review the pertinent definitions before performing identification.
5.8.5.1
Taxonomy and Systematics
The identification of copepod species mostly depends on examining individual appendages, which may be a difficult task for the beginner due to their minute sizes. The flattened pereiopod 5 is of particular taxonomic significance. The two-volume work of Boxshall and Halsey (2004) provides a basis for copepod identification. Dealing mainly with copepod taxonomy, anatomy and evolution, the book by Huys and Boxshall (1991) also has great importance. The marine harpacticoid species described through to 1996 are catalogued by Bodin, with the last update being published (also on diskette) in 1997. There are carefully prepared systematic keys for harpacticoids that provide a very good basis for taxonomic work (e.g., Lang 1988; Wells 2007, also on diskette). For description and identification of the ovoid naupliar stages (Fig. 5.34) of several species, see Dahms (1990, 1992, 1993). General features that are evident even to the nonspecialist and that are valid for entire families do exist in harpacticoids. Hence, the following short characterization of some more common harpacticoid families provides a rough orientation. The stated numbers of taxa are from Wells (2007).
Tachidius Parathalestris
Harpacticoides
typical nauplius
Ancorabolus
Fig. 5.34 Some harpacticoid copepods with different body shapes adapted to various biotopes; body lengths are between 0.2 and 2.0 mm. (Various authors)
Arenosetella
Leptastacus
Tisbe
Porcellidium
5.8 Crustacea 183
184
5 Meiofauna Taxa: A Systematic Account
Ameiridae: body shape similar to Miraciidae, but in contrast they have just one egg sac and a small or missing rostrum; primarily interstitial and burrowing species; about 325 species in 35 genera, some living in the phytal; Nitokra, Ameira, Ameiropsis. Ancorabolidae: body characterized by large dorsal spines and epimeral “wings” (chitinous extensions), apparently live covered with mud in deep-sea and polar bottoms, about 60 species in 22 genera; Laophontodes, Ancorabolus, Echinopsyllus. Canthocamptidae: largest harpacticoid family, with 50 genera and about 800 species; body elongate and cylindrical; segments are not distinctly set off from each other; cephalothorax without marked rostrum; one egg sac; primarily freshwater, benthic or phytal, and present in subterranean waters; a few brackish and marine species; Canthocamptus, Attheyelya, Mesochra. Cletodidae: body dorsoventrally depressed or a tube with a rostrum, segments very well separated by deep incisions. The roughly 115 benthic species in 24 genera prefer muddy sediments rich in detritus; especially common in tidal flats and salt marshes; also common in deep-sea muds; present worldwide; some species in freshwater; Cletodes, Enhydrosoma, Limnocletodes. Cylindropsyllidae (33 species) and Leptastacidae (74 species): mainly slender, interstitial forms of major importance in sandy habitats, about 100 species in total. Leptastacus excretes mucopolysaccharides from the caudal glands of its terminal segments, the mucus strands attract bacteria which are subsequently ingested (mucus trap feeding; see Sect. 2.2.3); Cylindropsyllus, Stenocaris. Ectinosomatidae: body spindle-shaped (e.g., torpedo-shaped or vermiform); no marked demarcation between thorax and abdomen; the fifth leg of each sex has a unique structure peculiar to the family; epi- and endobenthic ubiquists, typical opportunists; some 275 benthic and phytal species in 21 genera occurring worldwide from littoral to deep-sea bottoms; Ectinosoma, Arenosetella. Laophontidae: body has distinct segments and varies from being flattened to ovoid to slender and linear; flat forms particularly found in exposed sands; first leg unique and characteristic, with two segmented endopodite terminating in protruding claw which makes the family easily recognizable; one egg sac; common mostly in shallow (<100 m) marine habitats in sand, mud and as epibionts; 65 genera and about 330 species; Asellopsis, Laophonte, Paronychocamptus. Miraciidae (formerly Diosaccidae): body elongate, somewhat tapering towards the end, cephalothorax extending into a marked rostrum, females with two egg sacs; about 450 mostly benthic but also phytal and pelagic species in >50 genera occurring worldwide from littoral to deep-sea bottoms; Amphiascus, Schizopera, Stenhelia. Paramesochridae: body elongate–cylindrical to vermiform, small species, well adapted to the marine interstitial; about 125 species; may be the dominant harpacticoids in sandy shores, Apodopsyllus, Paramesochra, Scottopsyllus. Peltidiidae: body short and broad, dorsoventrally depressed, with large epimeral (lateral) plates; about 90 species in 11 genera live mostly among algae; Alteutha.
5.8 Crustacea
185
Tegastidae: body short, laterally compressed, like many amphipods (an unusual shape for harpacticoids); cephalothorax has conspicuous ventrolateral flanks; very small abdomen; some 75 species occur epiphytically on algae and in muddy sediments; Tegastes. Tisbidae: body shape cyclopoid (see below), more or less depressed; about 26 genera and about 220 mostly benthic and phytal species occurring worldwide from littoral to deep-sea bottoms; commonly found as epibionts; Tisbe, Scutellidium.
5.8.5.2
Biological and Ecological Aspects
The body size and shape of harpacticoids vary, often in accordance with the preferred biotope (Fig. 5.34). Typical interstitial species live in medium-to-fine sand and have thin, almost vermiform bodies with minute legs. Their uniformly metameric bodies with short, non-protruding appendages and setae give them high flexibility. They seem to swim rapidly through the interstices of the sand via a rapid writhing of the whole body, not just the legs (Arenosetella, Apodopsyllus, Hastigerella, Leptastacus, Parastenocaris). The term “interstitial fauna” was coined for the rich harpacticoid populations in a sample from a British sandy beach (Nicholls 1935). Species living in the phytal often have a stout, sometimes depressed, body with richly setose, often sturdy legs that are well adapted to clinging to plants as well as to swimming (e.g., Peltidiidae, Porcellidiidae, Tegastidae, with Thalestris, Porcellidium, Fig. 8.10). Other phytal forms squeeze their slender bodies through plant thickets (Laophontidae). Northern hemisphere estuaries, tidal flats and salt marshes are typically dominated by the genera Tachidius and Microarthridion, whereas in the southern hemisphere representatives of the Cletodidae and Cannuellidae commonly dominate shallow muddy substrates. A habitat correlation is less obvious in those harpacticoids that live in fine sands or muds. Their bodies are often spindle-shaped, sometimes almost cyclopoid-shaped (see below) or with distinctly set-off segments, and fairly large in size. Their stout legs help to dig in the mud; they prefer the surficial sediment and live mostly epibenthically (Ectinosoma, Cletodes, Tachidius, Paronychocamptus, Microarthridion). Some epibenthic deep-sea forms (e.g., Ancorabolidae) have developed bizarre dorsal spines to anchor mud balls stabilized by mucus as camouflage (Fig. 5.34; Thistle 1982). Species of Stenhelia and Pseudostenhelia are tube-builders living in tubes constructed from sediment (Chandler and Fleeger 1984). Harpacticoids have often been considered mainly “detritus feeders.” Other studies, however, revealed selective grazing on single food particles (bacteria, protozoans and particularly diatom cells) which the animals strip with their mouth parts from detritus, algae and sand grains (Marcotte 1983, 1984; Bouguenec and Giani 1989, Coull 1999; De Troch et al. 2005). Many harpacticoids also seem to devour fresh planktonic diatoms that sink to the sediment surface. Exudates of bacteria and plants (biofilms) are also included in the preferred trophic spectrum of these
186
5 Meiofauna Taxa: A Systematic Account
animals (Decho and Fleeger 1988b; Dahms et al. 2007). Bacteria are attracted and concentrated by the mucus that they produce. In some species a selective preference for chemo-autotrophic bacteria as food has been identified based on the very low isotopic signatures measured (Franco et al. 2008). On the other hand, phototrophic sulfur bacteria and cyanobacteria seem to be avoided, or are at least an inadequate food for many coastal species (Souza-Santos et al. 1996). Trophic specialization and a close relationship to the changing pattern of physicochemical conditions in the sediment favor a distinct distribution pattern for many species (Fig. 5.35). This zonation is especially well developed in the tidal soft bottoms (Coull et al. 1979; De Troch et al. 2002). In diatom-feeding species a close distributional correlation has been found with patches of microphytobenthos. Even within the “microalgal trophic niche,” harpacticoid species can exploit the various diatoms species-specifically with different preferences. This extreme partitioning can lead to a microdistribution that changes with food availability and seasonal growth phases (Pace and Carman 1996; Azovsky et al. 2005; De Troch et al. 2005).
Microarthridion littorale Halicyclops sp. Enhydrosoma propinquum Stenhelia bifida Nannopus palustris
Halectinosoma winonae
Schizopera knabeni
Paronychocamptus wilsoni Pseudobradya pulchella
Nitocra lacustris Diarthrodes aegideus
Pseudostenhelia wellsi Robertsonia propinqua
MHW High marsh Low marsh
MLW
Mud flat Subtidal
Creek bottom
Fig. 5.35 The horizontal distribution pattern of harpacticoids along the shore of a salt marsh in the southeastern United States. (After Coull et al. 1979)
5.8 Crustacea
187
On the other hand, flexibility in nutritional demands enables a shift to a broader food spectrum. Harpacticoids switch from one preferred food source to the other not only at different developmental stages but also sometimes with seasonal or tidal changes (Microarthridion). This flexibility explains why the (mass) culture of many harpacticoid species (e.g., Tisbe spp., Nitokra sp., Tigriopus; Cletocamptus, and others) is a successful and important method in experimental work, pollution monitoring, and fish aquaculture (Sun and Fleeger 1995; Chandler et al. 2004a,b; Brown et al. 2005). Beside food supply, temperature is a prime determinant of harpacticoid occurrence and development. Hatching and growth of the six naupliar and the five copepodite stages are mostly linked to increasing annual temperatures (Nodot 1978). Developmental time in situ is about 2–3 months in many harpacticoid species (Fleeger 1979), but can last a year in cold-water inhabitants. A prolongation also frequently occurs in groundwater species. Many limnetic and, so far, one marine species have been noted to develop cysts that are resistant to adverse conditions. Environmentally induced modulation of the developmental time can be accomplished by retarding naupliar development or a diapause phase during the copepodite phase. Parthenogenesis has also been reported for a few freshwater copepods. Species in crevices of Antarctic ice had copepodite resting stages (Dahms et al. 1990). On the other hand, Drescheriella glacialis from Antarctic sea-ice seems to compensate for the extremely cold temperatures by having a rapid life cycle, similar to the r-strategists of more temperate habitats (Bergmans et al. 1991). Most harpacticoids are sensitive to oxygen depletion, which restricts their occurrence in many sediments to the upper layers and favors epibenthic life. Nevertheless, in winter, migration into deeper sediment layers has been observed, provided there is enough oxygen available. In polar regions, populations of Tisbe furcata migrated during the winter months into the sea-ice layers, where they fed on the rich supply of algae (Grainger 1991). The often epibenthic harpacticoids are less sediment-bound than nematodes. Hence, they become easily disturbed by (tidal) currents, agitation during storms, or by the burrowing activities of macrobenthos (see Sect. 7.2.1). However, harpacticoids are also the classical “emergers” among the meiobenthos. Their morphology and setation of swimming legs discriminates them from persistent species (Thistle and Sedlacek 2004). Active emergence linking the benthic with the suprademersal biomes and the subsequent drift in the nearbottom water layers is influenced by favorable light and hydrodynamic conditions and often follows a diurnal rhythm (Bell and Sherman 1980; Palmer 1988). This behavior (Armonies 1989b; Buffan-Dubau and Castel 1996) leads to intensive dispersal, redistribution and colonization of new or disturbed habitats (Walters and Bell 1994). Despite this periodic “hyperbenthic” lifestyle, fairly consistent small-scale distribution patterns can be observed, although their formation is not yet fully understood. Some biological and physical factors are probably responsible for active and selective re-entry into the sediment, leading to non-random aggregations and patches in harpacticoids. In a comprehensive literature analysis of the distribution patterns of shallow-water harpacticoids,
188
5 Meiofauna Taxa: A Systematic Account
Chertoprud et al. (2007) found a set of six “life forms” that retained their species composition and suggested that the structure of the sediment and salinity gradients were controlling factors. High harpacticoid abundances are mostly encountered in shallow flats and lagoons with muddy, detritus-rich sand (from several hundred to several 1,000 specimens per 10 cm2), where they can even be the dominant taxon (e.g., in some areas of the southern North Sea). A density-dependent distribution was suggested by Sach and van Bernem (1996) for tidal flat harpacticoids, with more random patterns at low population densities and a highly patchy distribution at high densities. While the species composition can be rather monotonous in the eulittoral (about 20 species per sampling area), similar bottom types in the deeper sublittoral tend to show increased diversity (60–70 spp.), although with decreasing abundance. In deep-sea sediments there are often only 1–10 specimens per 10 cm2, but high diversity is maintained. Life history studies of meiofauna are often based on harpacticoid copepods (see Sect. 9.3.2) because some taxa (e.g., Tisbe or Schizopera) are easily cultured or because populations can be followed in the field due to their morphologically distinct ontogenetic life stages (see compilation by Ferrari and Dahms 2007). Generation times can be as short as 10 and 18 days in the field and can be accelerated in the laboratory. Compared to nematodes, the generally greater sensitivity of harpacticoids make them good indicators of pollution (Coull and Chandler 1992; Brown et al. 2005). In an extensive survey of the North Sea meiofauna (Huys et al. 1992), five well-defined harpacticoid groups could be discerned, based mainly on sediment structure and the impact of pollution. The frequent epibenthic occurrence of harpacticoids makes them a preferred prey for many small, often juvenile demersal fishes, carnivorous crustaceans (e.g., shrimps and their larvae) and polychaetes. In sandy tidal flats, but also in coral reefs, harpacticoids play a decisive nutritional role for small fish such as gobiids (Sect. 9.4.2; Gee 1987; Coull, 1990, 1999; Zander 1993; McCall and Fleeger 1995; Aarnio and Bonsdorff 1997; Fujiwara and Highsmith 1997). Derived from their preferred diatom food, harpacticoids have high fatty acid contents, which, in turn, seem to determine their nutritional value to fish (Coull 1999). The great diversity of copepods with their numerous parasitic families has been grouped in many different ways; the comprehensive cladistic phylogeny by Huys and Boxshall (1991) classifies Harpacticoida as a monophyletic group that is distant from the Cyclopoida. However, within the Harpacticoida the phylogeny remains debated. According to Noodt (1971) and Marcotte (1986b), harpacticoids with a more roundish and stout body shape (e.g., Tachidius, Tisbe) represent the more primitive type, and the vermiform sand dwellers are considered secondarily derived. This notion certainly needs molecular confirmation. More detailed reading: systematic monographs and identification keys, Lang (1988); Wells (2007), Bodin (1997); anatomy and phylogeny, Huys and Boxshall (1991); development, Ferrari and Dahms (2007); ecological reviews, Hicks and Coull (1983); Gee (1989); dormancy, William-Howze (1997).
5.8 Crustacea
189
Box 5.9 Harpacticoid Copepods: Bridging the Benthic Boundary In the multistory house of meiobenthos, harpacticoids usually live in the top stores and on the roof, so that they are always ready to make small excursions into the water layer above. In the near-bottom water layer, they prefer the phytal, the thickets of plants. Why do they favor this surfacelinked occurrence despite the associated enhanced risk of predation and physiographic extremes? At least in shallow reaches, it is probably the availability of the preferred food of many harpacticoids, diatoms, and freshly sedimented (phyto)detritus that compensates for these extra risks. As a result of this adaptive linkage, the developmental cycles of some species in tidal flats are concordant with the onset and distribution patterns of light-dependent diatom blooms. Life around the benthic boundary also has another advantage: it partitions the trophic niches of many harpacticoids from their most important, more sediment-bound competitors, the nematodes. However, the harpacticoid situation is not that easily understood. The populations of some species get regularly (and not only in shallow tidal waters) suspended in currents. Beside this suspension some harpacticoid species have been found to actively emerge into the water, especially at night and in calm weather. We cannot really perceive how the distributional advantages can compensate for the risk of drifting away in the water column, the risk of losing contact with adequate sediment and food, and the risk of being devoured by many plankton and nekton predators. Indeed, most harpacticoids are closely linked to the sediment conditions: despite their occasional excursions into the water body, we often find a clearly zoned distribution with defined local patterns and patches whereas resuspension would bring about rather homogeneous and accidental settlement. However, to gain a deeper understanding of their migrational and resettling phenomena we need to perform thorough autecological experiments: sequential drift and settlement studies; tests on the mobility behaviors and homing ranges of individual specimens; analyses of their sensing capacities. Ideally, we need to observe the functioning animals at eye level, not from above as a tiny mass of particles or portions of chemical energy. This holds true not only for harpacticoids. The drifters and emergers among them only serve to highlight our limited understanding of functioning processes in and among the meiobenthos.
5.8.6
Copepoda: Cyclopoida and Siphonostomatoida
This cyclopoid subgroup of copepods is widely believed to be restricted to planktonic life in freshwater. This is misleading since many cyclopoid copepods live on and in the sediment or the phytal. The Halicyclopinae and most Cyclopininae (about 30 genera in total) are marine taxa. They are quite common in the southern
190
5 Meiofauna Taxa: A Systematic Account
North Sea (Huys et al. 1992), but have also been found in eastern North America and in Brasilian beaches. While the stout bodies of most cyclopoids are adapted to living epibenthically or burrowing in muddy sediments, many of the smaller forms represent typical interstitial meiobenthos with vermiform bodies (e.g., Psammocyclopina) and reduced egg number (Halicyclops, Cyclopina, Metacyclopina). They are structurally well adapted to the mesopsammal and show some convergence with the interstitial harpacticoid family Paramesochridae. The genus Halicyclops can be abundant in brackish-water sediments, and in boreal estuaries it can be a dominant member of the meiobenthos. Most freshwater cyclopoids live epibenthically among macrophytes, with all transitions towards an endobenthic life. About 25% of all limnetic cyclopoids (i.e., about 160 species) can be considered endobenthic, subterranean or even troglobitic species. Aside from the reduction in body size, many of these species have also reduced their number of eggs (e.g., the troglobitic Eucyclops teras). The most specialized species have even lost their typical egg sacs, such that they carry their few eggs on long filaments (Speocyclops, Graeteriella). Some genera are restricted to the phreatic sediments of riverbeds and shores (Haplocyclops, Speocyclops racovitzai, Eucyclops subterraneus). In contrast with harpacticoid copepods, most Cyclopoida are predaceous carnivores that feed on meiofauna of an equal size or even larger. More detailed reading: Rundle et al. (2002). Copepoda Siphonostomatoida: Several species of one family, the Dirivultidae, are associated with hydrothermal vent biotopes (see Sect. 8.4.7). Here they live, probably as bacteria grazers, in the thickets of mussels, snails or polychaete tubes. Some even occur in the gill chambers of vent shrimps.
5.8.7
Malacostraca
This well-defined crustacean taxon emerges from morphological and molecular analyses as a monophyletic group, although the relationships within the malacostracan orders are not stable (Giribet et al. 2005). More commonly known by their macrobenthic representatives (Decapoda and Peracarida), there are also meiobenthic forms in almost all groups of malacostracan crustaceans. Some represent isolated, specialized miniatures of a speciose group of larger-sized animals (e.g., Isopoda, Amphipoda). Others represent the sole survivors of rare relict groups, often living in refuge biotopes (see Sects. 7.2, 8.7). The specialized miniatures mostly have typical convergent features that secondarily adapt them to a meiobenthic and particularly to an interstitial life. The relict forms, however, represent a composite of specializedderived and archaic-primitive features (e.g., Syncarida, Thermosbaenacea). Along with their often enigmatic zoogeography, this makes the few existing species a particularly rewarding study target for the phylogenetically interested zoologist. Among the malacostracan crustaceans mentioned in the following sections, many exceed the formal size limits that separate meio- from macrobenthos. However, they do live in habitats typical of meiobenthos, their ecology is typically
5.8 Crustacea
191
meiobenthic in many respects, and they have adaptations that are characteristic of the meiobenthos. Therefore, they are considered an “ecological meiobenthos” here, justifying their inclusion in this book. Their zoological, geographical and phylogenetical relevance would have made this book incomplete had they been omitted.
5.8.7.1 Syncarida This little-known crustacean superorder contains two orders, the Bathynellacea and the Anaspidacea, which lead a mostly subterranean life. While they are currently largely restricted to freshwater habitats, their fossilized ancestors are known to have dwelled in the marine strata of the Carboniferous. It is believed that they emigrated via the coastal, brackish groundwater. While almost all Bathynellacea are stygobionts, some species of the Anaspidacea that live superficially and in moist plant thickets should be considered stygophiles. The distributions of many syncarid crustaceans, especially the anaspidaceans, mirror the extent of the old supercontinent of Gondwana, corresponding to the recent Southern Hemisphere. Here, they are apparently still radiating. Bathynellid species are found globally; many of them appear to have a worldwide distribution. The increase in species descriptions is considerable and the group amounts now to 235 species. While the Bathynellacea have been investigated more thoroughly and some ecological details beyond their morphology and zoogeography have been described, little is known about the Anaspidacea. Syncarid morphology (Fig. 5.36a) consists of both primitive and derived features. This makes them, on the one hand, living representatives of primitive malacostracans (similar to Palaeozoic fossils), while, on the other hand, they are adapted convergently to the freshwater interstitial by developing a small, cylindrical and flexible body with uniform segments and reducing the carapace. The original eye peduncle is often abandoned, resulting in sessile eyes; sometimes the eyes become rather rudimentary. Bathynellacea: Some Bathynellacea with a body size of 0.5 mm represent the smallest malacostracans (Fig. 5.36a). Eyes and statocysts are reduced. They have retained a furca and styliform uropods. Bathynellaceans occur worldwide in groundwater systems, in wells and in river sands, where they appear to feed on bacteria and fungi colonizing the sand grains and detritus particles. Schminke (1981) showed that the group has clear homologies to the zoea/protozoea stages of the Eucarida (Penaeida) and thus suggested that it developed by progenesis from common ancestors of these other malacostracans. Like their penaeid relatives, bathynellids were originally larger in size (one species of length 50 mm still exists in Tasmania) and probably had free larval stages. Miniaturization and progenetic development enabled them to enter the mesopsammal of river mouths and from there the groundwater system, thus reducing competition from more “modern” crustaceans. Bathynellacea consist of two families, the Bathynellidae and Parabathynellidae, which have circum-mundane distributions with overlapping areas between the 10 and 20° longitude girdles (Schminke 1973).
192
5 Meiofauna Taxa: A Systematic Account
250 µm Syncarida: Bathynella
1 mm
a
Syncarida: Psammaspides
500 µm
b
Pancarida: Thermosbaena
Fig. 5.36a–b Representatives of the Syncarida (a) and Pancarida (Thermosbaenacea) (b) (a After Schminke 1986; b after Delamare-Deboutteville 1960)
Bathynella: 1–2 mm long, found in European and Californian groundwater; the first bathynellids, found in fountains. Allobathynella: mostly in Eastern Asia, includes many species formerly assigned to Parabathynella. Hexabathynella: cosmopolitan, mostly found in rivers; also found in brackish shore sand. Habrobathynella: India, Madagascar, in river sediments. Thermobathynella: Brasil, Central Africa, in thermal waters (55 °C) and river sand. Anaspidacea/Stygocaridacea: A relict group of about 20 known species grouped into five families restricted to the southern continents. Similar to some Palaeozoic fossils; several of them have archaic anatomical traits and give a picture of a primitive malacostracan. Psammaspides, Stygocarella: stygobiotic in New Zealand and Australia respectively. Micraspides (0.8 mm) and Koonunga (some species up to 8 mm): from crayfish burrows and moist mosses in Australia; furca reduced.
5.8 Crustacea
193
Stygocaris Noodt: 1.5 mm long, found in the mesopsammal of South America and Tasmania; caudal furca only retained rudimentary. New reports of stygocarid crustaceans from Australia and New Zealand indicate that this group, originally established as a separate syncarid order, represents a family (“Stygocaridae”) within the Anaspidacea. More detailed reading: taxonomy, zoogeography, Noodt (1965); Pennak and Ward (1985); ontogeny, phylogeny, Schminke (1973, 1981, 1986).
5.8.7.2 Thermosbaenacea, Pancarida These rare, meiobenthic malacostracans (largest species 4 mm, Fig. 5.36b) are known from about 35 species within three families. They are related to the Peracarida (see following section). Monodella: about ten species, they have a “north amphiatlantic” distribution from the Caribbean to Italy and the near East. The vermiform species are very euryoecious, occurring in sands and muds of both marine and freshwater biotopes. Halosbaena (1.8 mm): the only marine form with a “south amphiatlantic” distribution, mostly found in coral rubble. Tethysbaena: first discovered in a brackish karst outlet in Southern France; now about 20 spp. of vicariant distribution along the Atlantic coasts, mostly in the Northern Hemisphere; some in the coastal groundwater. Thermosbaena mirabilis: a monotypic species found in the sediment of thermal springs (42 °C) in Tunisia; five pairs of pereiopods only. Instead of typical peracararid oostegites, Pancarida have developed a dorsal brood pouch or “marsupium” beneath the posterior part of the wide carapace which covers the dorsum up to the fourth free thoracic segment; the animals are blind. The distribution center of Pancarida currently consists of refuge biotopes such as marine caves and sometimes the groundwater system. Probably several independent lines emigrated from the Oligocene shorelines of the Tethys Sea via brackish groundwater and colonized the crevices of subterranean freshwater aquifers (some Tethysbaena species). Ontogenetically, in parallel with the Bathynellacea (see above), Pancarida are related to the mysis stage of penaeid shrimps (Decapoda), which indicates a progenetic origin too (Coineau 2000). More detailed reading: Monod (1940); Stock (1976); Wagner (1994).
5.8.7.3
Peracarida
In several phylogenetic lines, peracarid crustaceans have independently developed meiobenthic forms. The whole taxon (Mictacea), a good part of it (Tanaidacea), or just some specialized forms (like in Isopoda and Amphipoda) may be meiobenthic.
194
5 Meiofauna Taxa: A Systematic Account
Mictacea. First discovered in 1985, there are now two monotypic genera: Mictocaris halope Bowman and Iliffe 1985, 3–3.5 mm long, from marine caves of Bermuda (Fig. 5.37a); and Hirsutia bathyalis Sanders et al. 1985, from the deep-sea benthos. These homonomously segmented crustaceans have, like all true peracarids, a ventral marsupium for egg brooding. They have a combination of body features that do not fit with any other existing peracarid order; they are distinguished by one unique character, their eyestalks, which lack functioning visual elements. The isolated occurrence of Mictacea in disjunct biotopes points to a long, independent evolution. Mictocaris has been observed to live epibenthically, swimming with the exopodites of its pereiopods. Spelaeogriphacea. An isolated monotypic taxon; Spelaeogriphus lepidops, 6–8 mm long, blind, from a cave near Cape Town, South Africa, where it lives in pools and a stream (Fig. 5.37b). This troglobitic group is phylogenetically close to Tanaidacea (see below). It is distinguished by the well-developed exopodites on all but the last thoracopods. Females have a ventral marsupium.
0,5 mm Mictacea: Mictocaris halope
1 mm Spelaeogriphacea: Spelaeogriphus lepidops
1 mm Tanaidacea: Gollumudes botosaneanui
Fig. 5.37 Representatives of meiobenthic Peracarida: Mictacea, Spelaeogriphacea and Tanaidacea. (Various authors)
5.8 Crustacea
195
Tanaidacea. A good part of this speciose and widespread marine group comprises numerous small species of meiobenthic size (a few mm in length, the smallest measuring only 0.8 mm) and with vermiform, narrow bodies. Distributed worldwide, they prefer muddy sediments (where they are tubicolous) and the phytal; only a few live in the interstitial of coarse sand. They occur frequently in deep-sea bottoms and can be among the most abundant crustaceans. Tanaissus lilljeborgi: 2.5 mm; Heterotanais oerstedti: 2 mm, occurs in muds of the North Sea; Anarthrura simplex: 1.5 mm, Atlantic Ocean; Psammokalliapseudes: from the mesopsammal off Brasil; Gollumudes botosaneanui: from beach sediment in Curacao (Fig. 5.37c). The tanaidacean carapace is highly reduced, allowing for high flexibility of the body, a trend enhanced by the numerous free and rather homonomous segments. This is important when digging U-shaped tubes in muddy sediments. In sandy bottoms and among algal mats the body is particularly vermiform; here they stabilize their tunnels by spinning silk from glands at the tips of peraeopods 1–3. All thoracopod exopodites are absent. The second thoracopod has a large distal chela. In some tanaidaceans, sex determination is phenotypic and complicated, sometimes with heteromorphic genders. In some families, different types of males and protogynous hermaphrodites have been described (a rare exception in malacostracans!). Most representatives are detritivores and scavengers, also feeding on diatoms; however, some predaceous species grasp nematodes and harpacticoids with their chelate legs. In shallow waters the tanaidacean abundance can reach densities of 100–1,000 individuals per 100 cm2, but their distribution is extremely patchy. In many areas tanaidaceans are common food for polychaetes, malacostracans and small fish. When numerous, they represent an important member of the marine benthic food chain. Isopoda. This large and diverse order has developed numerous meiobenthic forms through different adaptative lines known as the “micro-isopods” (Fig. 5.38). Structurally divergent, they belong to various suborders, superfamilies and families which have adapted to an epibenthic, interstitial–mesopsammic or endobenthic life. The groundwater species are often only 1 mm long (which is an unusually small size for the complex body organization of Malacostraca), but many species are 2–3 mm long, and yet they are well adapted to living in void systems, particularly those of river gravels (see Sect. 8.7.1), where they mostly feed on detritus. Large deep-sea isopods seem to selectively feed on Foraminifera. Many meiobenthic isopods belong to the superfamily Janiroidea; the genus Microcharon alone contains > 70 species. Microcerberoidea represent about 50–60 spp. of Microcerberus; all are meiobenthic in size and are probably of progenetic origin. Within the Cirolanidae (Flabellifera) there are about 45 stygobiotic freshwater species. The meiobenthic isopods from sand or gravel often have modified the typical dorso-ventral depression of the body into an almost round, vermiform shape with rather uniform segments. The resulting slender body is highly flexible. Reduction of eyes, pigments and long appendages occurs frequently in the numerous troglobitic and interstitial species, which all exhibit strongly thigmotactic behavior. In Stenasellus
196
5 Meiofauna Taxa: A Systematic Account
1 mm Cruregens frontanus
Angeliera phreaticola
300 µm
250 µm Microcerberus sp. Fig. 5.38 Characteristic representatives of meiobenthic Isopoda. (Various authors)
the lifespan and the intermolting periods appear to be unusually long. In Microcerberus, with reduced oostegites, eggs are deposited directly into the voids of the sand. Another developmental line is evident in the epibenthic forms. With their flattened bodies (large epimeres) and long legs they can easily move over soft surfaces. Microparasellidae: five genera with numerous species that are all only a few mm long. Their general distribution pattern corresponds to the region of the former Tethys Sea. Microparasellus lives in the shallow marine interstitial, as well as in freshwaters of European karstic regions; it is not present in America. Angeliera phreaticola occurs in the brackish coastal groundwater of Mediterranean beaches, as well as in surrounding freshwaters; it was also found off Madagascar. Microcharon is mostly known from the groundwater system, but some archaic forms are also marine and appear to be distributed worldwide: they have been found off Roscoff (coast of Brittany), along Mediterranean coasts, along the Gulf of Mexico coast off Belize, and on the Galapagos Islands. Microjaera lives sublittorally off Banyuls-sur-Mer (Mediterranean) and off Roscoff, France. Other taxa include the genera Protocharon and Paracharon.
5.8 Crustacea
197
Stenasellidae: with several genera that live in continental groundwater aquifers in all continents; Stenasellus spp. Microcerberoidea: the genus Microcerberus, containing species that are often just 1 mm long, occurs worldwide in continental groundwater systems, as well as in caves and aquifers of karstic bottoms. It also contains interstitial marine and coastal brackish species. This subterranean taxon is probably related to the epigean group Stenasellidae (Asellota). Phreaticoidea are restricted to subterranean freshwaters of the southern hemisphere; their bodies are strongly vermiform and attenuated. Neophreaticus, from New Zealand. Anthuridea: the center of this group seems to lie in the Caribbean and the IndoPacific. The species are mostly marine or live in freshwater in the vicinity of oceanic coastline. With their slender bodies they dig in the sediments, often occurring in tubes of polychaetes and in the culms of seagrasses, where they live as predators. The stygobiotic, often blind groundwater forms are sometimes only 1 mm long. Most-known genus: Cyathura. Flabellifera, Cirolanidae: these are mostly stygobiotic species with circumMediterranean and circum-Caribbean distributions. With their flattened bodies and reduced sizes (often only 2–3 mm long), they are typical of karstic habitats, where they live mostly as detritivores or scavengers. Cirolana, Speocirolana, Arubolana. Oniscoidea: Nannoniscus is a small (1–2 mm long), flattened deep-sea isopod living in the surface layer of the sea bottom. Most of the circum-Mediterranean meiobenthic isopods have been found in both marine and groundwater habitats, which led to discussions about whether they represent relict forms from the former Tethys Sea. Did they keep their old distributions, but, after the Tethys Sea had regressed, slowly adapt in several independent lines to brackish and finally to freshwater conditions? The alternative would be that the marine forms migrated independently and actively through crevices, often reaching deep down into the groundwater passing through the brackish mesopsammon of many beaches (see Chap. 7). More detailed reading: Faunistic and\ zoogeographic aspects, Delamare Deboutteville (1960); Botosaneanu (1986b). Amphipoda. Like isopods (see above), the dwarf forms of several amphipod orders appear to have developed independently. Even the morphological adaptations are fairly convergent with isopods, sometimes modifying the typical laterally compressed body shape of an amphipod into a roundish, slender body (Fig. 5.39). Some of the morphological species are extremely euryhaline, enabling them to live in marine as well as limnic habitats (different haline capacities in divergent physiological species or just different populations? See Kinne 1964). The data on the occurrence of meiobenthic Amphipoda do not yet allow us to draw a realistic picture of their geographical distribution. There seems to be a tendency for most marine meiobenthic amphipods to live in the southern hemisphere. Many freshwater forms are found in identical genera or even species in the Old and New Worlds, suggesting that these groups are very old, and originated on
198
5 Meiofauna Taxa: A Systematic Account
500 µm Niphargellus sp.
250 µm Ingolfiella sp.
500 µm Bogidiella sp.
Fig. 5.39 Characteristic representatives of meiobenthic Amphipoda. (Various authors)
Gondwana or on the archaic Laurasian continent before they broke up, which formed the Atlantic Ocean. Meiofaunal amphipods live both in the sea and in freshwater, either interstitially in sands (rarely in mud), and often in rubble from encrusting algae and corals. Many also occur in the culms and holdfasts of seagrass and algae and belong to the phyton. Globally, many small amphipods live in subterranean biotopes like groundwater, springs, caves and river beds.
5.8 Crustacea
199
Although the sizes of small Amphipoda may exceed the traditional limits that separate meio- from macrobenthos, they beautifully exemplify typical adapatations for meiobenthic, and in particular interstitial, life (Fig. 5.39): minute sizes (for the group), vermiform bodies with reduced epimeres, 1–2 eggs, reduced eyes, no pigmentation, and high flexibility of the uniform body segments. Ecological studies on meiobenthic amphipods (as well as isopods) are scarce, but it can be inferred that most of them feed on small detritus particles and bacterial films on and in the sediment. Niphargidae: Mostly found in subterranean waters of the holarctis. Niphargus and related species. Within this large group, widely known from European groundwaters, most species have kept the typical amphipod morphology. However, there are some that are small in size (only 2–3 mm) and that live in the interstices of sand. These typically possess a modified body organization: the slender, vermiform type with short legs (e.g., Niphargellus), or the stout, short type that can roll up its body. Numerous species are adapted to brackish-water conditions. Microniphargus: 2 mm, from wells in western Germany and Belgium. Psammoniphargus: from the groundwater of the former Yugoslavia and Brazil. Crangonyctidae: Most representatives have a holarctic, particularly American arctic, distribution, and are often larger than the generally accepted meiobenthic size range, although some typical stygobionts have attenuated bodies only 1–2 mm in length. All species are cold stenothermal, photonegative and mostly lack eyes and pigments. They occur in springs and caves in karstic areas of the southern and central United States, but have also been found in India, Eurasia and recently in Africa. The stygobiotic forms are clearly derived from surface-living ancestors. Crangonyx spp.: Occur mostly in the United States, only a few are known from Eurasia. Synurella spp.: Hypogean as well as epigean forms, mostly from eastern Europe, some found in Asia; smallest species are 1.5 mm long; found in springs, subterranean waters, and in the mud in creeks. Stygobromus spp.: mostly subterranean and from the United States, but also found in (western) Asia. Bogidiellidae: currently more than 100 species in 35 genera with global distribution. Bogidiella spp. are typically subterranean animals. Their small sizes (often only 2.5 mm length) probably result from progenesis; they are found in caves, springs, river beds and groundwater, but occasionally also in the marine interstitial (B. chappuisi), mainly in Europe and South America but also in India. Ingolfiellidea: Ingolfiella: About 25 species of this strongly modified and progenetic group of amphipods are described. Many are only 1–3 mm long and have vermiform bodies that lack oostegites in the females. They are found in refuge biotopes like caves, continental and coastal groundwaters, but also in deep-sea bottoms. Their vermiform bodies best demonstrate typical adaptations to the lebensform of the mesopsammon. There are also species that are meiobenthic in size in other amphipod families living in both marine and freshwater habitats, e.g., Uncinotarsus pellucidus (Aoridae), which is 1.5 mm long. Found in the shallow sublittoral off Roscoff, it is a typical interstitial form with characteristic adaptations to the mesopsammal. Salentinella (1.6 mm) is a characteristic element of the Pyrenean groundwater fauna.
200
5 Meiofauna Taxa: A Systematic Account
More detailed reading: faunistic and zoogeographic aspects, Delamare Deboutteville (1960); Botosaneanu (1986a); monographs, Barnard (1969, marine species), Barnard and Barnard (1983, freshwater species).
Box 5.10 Interstitial Malacostracans: Paradigms for Evolutionary Tales Upon hearing about Malacostraca we often envision lobsters, shrimps or perhaps gammarid amphipods. It is hard to believe that millimeter-small, vermiform, blind creatures that live deep in the groundwater, in karstic caves, in thermal springs or in sands of remote oceanic islands are also Malacostraca. Descendants of various classes with very different body structures and appearances, they all have to deal with the same constraints when adapting to the world of narrow voids in sand or gravel. The evolutionary outcome was a rather uniform, narrow and flexible body; not an easy task considering the normal sizes and body structures of the “higher” crustaceans with their heteronomous sections (tagmatization), bulky carapaces, long antennae and protruding walking legs. So how did they become small? They remained at an early and tiny ontogenetic stage but became mature: neoteny is fairly frequent in mesopsammic malacostracans. Eyes and colors were useless in the dark and so were reduced; eye stalks as well as protruding epimeres or oostegites were hindrances and were often reduced; the result was “smoothened,” wormlike, whitish and minute crustaceans with short legs. Although affiliated to groups as different as isopods, amphipods or tanaids, the student might easily lump them into one taxon—convergence par excellence! Another interesting feature of these micromalacostracans is their physiological evolution: within one genus, even within the same (morpho)species, both marine and freshwater forms exist (Ingolfiella, Microparasellus, Bogidiella chappuisi). This haline versatility, which is rare among other groups, may have opened up archaic pathways along which many of these originally marine interstitial malacostracans invaded the subterranean aquifers of the continents via (tropical?) coastal groundwaters. In many cases these evolutionary steps must have been taken in pre-Jurassic times when the uniform southern supercontinent Gondwana still existed, when Africa was still connected with Madagascar, when Australia and New Zealand were still parts of Laurasia (many bathynellids, anaspidaceans, e.g., Stygocaris, or amphipods, e.g., Bogidiella spp.). In the Tertiary the coastal groundwater system of the tropical Tethys Sea served as a bridge between shores that are today separated by a deep ocean (meiobenthic isopods). Thus, the present distributions of many subterranean freshwater malacostracans represent prehistoric patterns and mirror ancient zoogeographical connections. Migrations from the shallow marine interstitial into the continental groundwater, and often into caves, are interpreted as refuge routes for those taxa that are no longer competitive in the evolutionary “hot spots” of shallow coastal seas. Therefore, the study of stygobiotic and cavernicolous micromalacostracan groups in comparison with their radiating, usually larger, relatives from the surface often provides a way of investigating evolution (particularly regressive evolution) at work.
5.9 Chelicerata: Acari
5.9
201
Chelicerata: Acari
The small body size of mites has enabled them to contribute several independent subgroups to the meiobenthos (Fig. 5.40). The most successful is the superfamily Halacaroidea, with the family Halacaridae. In several independent lines, some 50 genera and more than 1,000 species from this family have invaded marine and brackish habitats since the Mesozoic (Bartsch 1996, 2004, 2006) and live preferably among plants and hard substrates. They even colonized with some 60 species freshwater biotopes (Limnohalacaridae). Other suborders of mites live mainly in the bottom substrates of lakes and rivers; most of them belong to the Hydrachnidia (for details see Di Sabatino et al. 2000). Representatives of some other mite groups can be encountered in marine sands and algal epigrowth. The Oribatei, with their uniform brownish color and heavily sclerotized dorsal shield represent a suborder that normally lives in terrestrial soils. The famous vermiform Nematalycus nematoides (Trombidiformes) was found in a beach near Algiers (Coineau et al. 1978, Fig. 5.40); its structural convergences with other insterstial groups are remarkable. Within the Rhodacaridae (Mesostigmata) there are two genera containing species that are regularly found in marine intertidal sand (e.g., Rhodacarellus). The slender bodies of some of these interstitial species seem well adapted to moving through the interstices of the sediment. Pontarachnidae, related to freshwater mites, are characteristic representatives in marine sediments and dense epigrowth in warm water areas.
5.9.1
Halacaroidea: Halacaridae
More than 1,000 species of halacaroids are known, most of which are meiobenthic in size and are inhabitants of marine sediments, the marine phytal and barnacle epigrowth. The family Halacaridae alone harbors >200 known species, and there is reason to assume that this represents only a small percentage of the real number. Some halacaraid species are specialized inhabitants of subterranean groundwater with wide distributions. Halacaroid mites are easily recognized by the division of their body into the “gnathosoma” carrying the chelicerae and pedipalps and the “idiosoma,” with the two first pairs of legs directed forwards and two pairs of posterior legs directed backwards. The distance between these laterally attached legs can become rather wide and the body can attain an elongated shape, thus adapting the species to a life in the interstices of sand (e.g., Anomalohalacarus, Fig. 5.40). The chitinous body cuticle can develop a pattern of sclerotized plates ornamented with numerous setae. The positions of the setae on the idiosoma and the legs and the shape of the plates are reliable diagnostic features, as is the articulate structure of the legs and the shape of the claws. Interstitial forms are either soft-bodied and slender with reduced body plates (Anomalohalacarus), enabling them to squeeze through the narrow voids, or stout, cylindrical and strongly armored with heavily scIerotized plates and legs (for interstitial forms), providing protection against sediment pressures (Acaromantis). In unfixed specimens, the remains of plants in the gut, visible
HALACARIDAE
100 µm
Halacarellus
Anomalohalacarus
100 µm
Ameronothrus ORIBATIDAE
200 µm
Fig. 5.40 Some typical representatives of different families of Acari. (Various authors)
Acaromantis
100 µm
300 µm
RHODACARIDA
Rhodacarus roseus
NEMATALYCIDAE
Nematalycus
100 µm
202 5 Meiofauna Taxa: A Systematic Account
5.9 Chelicerata: Acari
203
through the more or less transparent bodies of many halacaroids, make the idiosoma a characteristic color (dark green, reddish, brownish–black). Despite their relatively clear-cut diagnostic characters, halacaroids are rarely investigated, even though they represent common members of the meiofauna that regularly occur in samples from almost all biotopes, whether littoral or deep-sea (see below). The genus Copidognathus alone comprises about one third (> 300) of all halacaroid species. Other genera which are frequently encountered in boreal shores (tidal flats) are Halacarellus, Lohmannella, Acarochelopodia, Actacarus, and Rhombognathus (Fig. 5.40). Biological and ecological aspects. Most halacaroids reproduce only once (semelparous). The female takes up a spermatophore deposited by the male. They have 10–20 eggs (the mesopsammic species have only one to a few eggs); after a relatively long developmental time one egg reaches maturity at a time. There is one larval stage, which has only the first three pairs of legs, followed by 1–3 nymph stages in which the last pair of legs develops. In summer, samples often contain only juveniles; adult halacarids predominate in winter. This sometimes makes identification difficult. Average life span is 5–9 months. Their low reproductive potential and their very limited mobility make halacarids slow colonizers after events that destroy the population. Halacarids are hardy creatures, able to live in a wide range of biotopes without too many morphological variations of their general body organization. Although they seem to prefer defined conditions of moisture, pH or salinity (Bartsch 1974), they can withstand a range of salinities, from freshwater to 30% S, while maintaining full activity. This capacity enhances the overall abundance of halacarid mites in brackish biotopes. One requirement for this amazing ecological resistance, however, is good oxygen supply, since most mites are sensitive to hypoxic conditions and do not occur in hypoxic and sulfidic muds. Mites can also survive extremes of temperature, desiccation and (hyper)salinity via an inactive stage during which they reduce their respiration significantly. After the fixation of meiofauna samples halacarids tend to move their legs for a wearily long time! Halacarid mites occur in all kinds of interstices. Aside from the mesopsammal, they prefer the thickets of mussel beds and kelp holdfasts, and thrive among cirripede aufwuchs or colonies of hydrozoans or bryozoans. Halacaroid community composition seems to be mainly determined by micro-environmental factors and not so much by wave exposure or the nature of the aufwuchs substrate (Somerfield and Jeal 1995). Mites crawl characteristically slowly and somewhat awkwardly. The abundant phytal forms climb with particularly strong clinging appendages, while the often very small (200 µm) and slender mesopsammic forms have a rather flexible, concave body shape that attaches better to the sand grains (e.g., Anomalohalacarus). Those species living in exposed and agitated coarse sand protect their roundish bodies with armored, solid plates and keep their legs tightly pressed to the body in cuticle depressions. 90% of all halacarids live in shallow shelf biotopes: the phytal of boreal and temperate regions is preferred by some (e.g., Rhombognathinae). In algal mats and Enteromorpha canopies halacarids can make up 90% of all meiofauna, in the
204
5 Meiofauna Taxa: A Systematic Account
mesopsammal of medium sand they still comprise 15%, while in fine sand with a limited supply of oxygen this figure reduces to about 5% (Fig. 5.41). In this case, they never enter the deeper horizons and tend to live epibenthically. Silty muds are devoid of halacarids. In the epigrowth of larger animals (crustaceans, gastropods), mites are regular phoretic guests. The set of preference reactions (see above) allows certain species of marine mites to be distributed differentially along the shoreline. However, halacarids are well known to stay in a site even when it provides harsh conditions. A few stress-resistant taxa are associated with algae above the midwater line (e.g., Isobactrus), while most mites live in the lower tidal to subtidal range (Somerfield and Jeal 1995). Distribution with depth seems less defined; deep-sea samples sometimes harbor the same genera as from the shore. Their euryecious nature along with their long evolutionary lines and an often parthenogenetic mode of reproduction contribute to the wide distribution areas and limited endemism of many halacarids. No halacarid genus is restricted to just one zoogeographical province. This explains why the halacaroid fauna of the Baltic Sea and the North Sea are more or less identical and the inter-province fidelity of amphi-North Atlantic species is still 45% (Bartsch 2004). Like many other chelicerates, halacarids have piercing and sucking mouth parts and an extra-oral digestion. Most of them are carnivorous, feeding for instance on crustaceans and oligochaetes; the phytal forms (see below) feed on the soft parts of hydrozoan and bryozoan colonies. Rhombognathus is phytophagous, piercing algal
Halacaroidea other Acari
Upper slope, sand
Nematoda Oligochaeta Harpacticoida
Middle slope
Ostracoda Others
Enteromorpha Tidal flat
Fig. 5.41 Proportion of halacarids in the meiofaunal spectrum along a beach profile. (After Bartsch 1982)
5.9 Chelicerata: Acari
205
cells. In some biotopes a certain degree of competition with oribatid mites can be expected. In turn, marine mites have been observed to serve as prey for some small fish, as well as for hydrozoan polyps, but there is generally not much predation pressure on them. Halacaroid mites can best be retrieved by decantation, perhaps along with a hot freshwater (60 °C) shock. Fixation with formalin should be avoided. For further preparation, clarification of the body plates in lactic acid is required. An identification key for the marine genera of halacarids (Bartsch 2006) can be downloaded from the net in the PDF format; a key to the species around the British Isles has been developed by Green and MacQuitty (1987).
5.9.2
Freshwater Mites: “Hydrachnidia,” Stygothrombiidae, and Others
Aquatic mites from many orders and families are combined in the group “Hydrachnidia” or formerly “Hydrachnellae” (more than 6,000 spp). They live among the phytal, in the sands of river beds (hyporheos) and lotic streams, as well as on mud of stagnant freshwater biotopes. One of the dominant hyporheic genera is Atractides; the larvae of this genus are parasites on chironomid midges. In many hydrachnellid groups, these biological ties to insects seem to have prevented successful colonization of the stygobial. The inhabitants of the hyporheic interstitial often show analogous specializations that are typical of the habitat: reduction of eyes, elongation of the body (Wandesiidae, Stygothrombiidae). The interstitial forms are clearly smaller than the epigean ones, exhibit strongly positive thigmotactic responses and have more or less reduced eyes. They never swim but prefer to crawl on sand grains. For further ecological details on freshwater mites see Sect. 8.2. More detailed reading (Acari): taxonomy, Viets (1927); Bartsch (1979); faunistics and zoogeography, Bartsch (1989, 1996, 2004); ecology, Bartsch (1974, 1989); Pugh and King (1985a,b); freshwater groups, Schwoerbel (1961b); Di Sabatino et al. (2000, 2002).
5.9.3
Palpigradi (Arachnida)
It is with hesitation that representatives of this primitive and rare group of Arachnida are included here. While most Palpigradi live in moist terrestrial soil and caves, a few species from three genera have been found in the eulittorals of tropical beaches and shallow coral sand. Considering their small sizes (less than 2 mm), slender shapes, long flagella and their fairly thin and flexible appendages, they seem well matching to the requirements of an interstitial life. The most interesting features that provide convincing arguments for life in the marine meiobenthos are of an ecological nature: specimens of Leptokoenenia scurra, the best-studied species, normally
206
5 Meiofauna Taxa: A Systematic Account
crawl on and clutch sand grains. When extracted, they can successfully move back into seawater. While all terrestrial palpigrades have a hydrophobic cuticle, which would trap them at the water surface, these marine forms can easily pass through the surface film. The occurrence of marine palpigrades has been considered supporting a new suggestion of a possible marine origin for Arachnida and the colonization of terrestrial habitats from the seashore. More detailed reading: Condé (1965); Monniot (1966).
Box 5.11 Halacarid Mites: A Story of Invasion and Re-Invasion It is not true that there are no arachnids in the sea. More than 1,000 species of halacarid mites refute this popular but erroneous opinion. Their amazing tolerance to harsh environmental conditions, especially to salinity fluctuations, underline the invasive potential of the ancestral mites that made their way into the sea and radiated there. Steadily climbing in algal aufwuchs on rocks, hiding in the crevices of barnacle, bryozoan or hydroid colonies, and even squeezing through the voids of beach sand, they conquered the shallow sites and, from there, even reached the deep-sea. Some 60 species have reinvaded freshwater regions as descendents of marine genera and populate, together with other meiobenthic mites, limnetic biotopes. Their euryoecious nature might be the reason for another characteristic: although they are widely distributed and live under different habitat conditions, halacarids did not much alter their typical body structure, having a bipartite body and four pairs of walking legs: two forwards and two backwards pairs. When they invaded the mesopsammal they did so with little more than elongation and body flexibility. Therefore, halacarid mites, especially when alive, are easily recognized between algal fronds or on sand grains. What a different evolutionary strategy compared to the meiobenthic malacostracans or polychaetes, which often changed their body structure beyond recognition! Since they play a minor ecological or phylogenetic role, the Halacaroida remain a niche taxon, although they can dominate the meiofauna of suitable phytal habitats.
5.9.4
Pycnogonida, Pantopoda
Among the meiobenthos extracted from sublittoral marine sand samples, there are occasionally also some minute representatives of the usually macrobenthic Pycnogonida or Pantopoda. These are about ten species of Anoplodactylus, Nymphonella and Rhynchothorax. Except for their small sizes (the smallest is about 1 mm in body length) and, in some species, reduced ocular tubercles and eyes, there are no characters that strongly differ from the general body organization of this strange animal group, which is associated with the chelicerates. More detailed reading: Child (1988).
5.11 Annelida
5.10
207
Terrigenous Arthropoda (Thalassobionts)
There is a heterogeneous assembly of normally terrestrial arthropods that are meiobenthic in size and occur so regularly in marine biotopes that they should be briefly mentioned here. These animals are characterized not so much by morphological idioadaptations (perhaps their normally well-developed clinging legs are important), but by highly interesting and genuine physiological and ecological features. The strong affinity of these arthropods to the marine realm seems to be based on an ecological niche realized in marine (temporarily) moist sand and in the algal cover of supra- and eulittoral hard bottoms. The typical and highly specialized forms, although mostly widespread, are rarely investigated. Most of them belong to chelicerate groups, followed by Tracheata: 1. 2. 3. 4. 5.
Mites: Hydracarida, Gamasida (Hydrogamasus ), Oribatei, Uropodidae Pseudoscorpiones Aranea Chilopoda (centipedes) Insects: Collembola (Anurida, Archisotoma), Diptera (larvae of Ephydridae), Coleoptera (Staphylinidae)
Compared to the normal meiobenthos, the problem for thalassobiotic terrigenous arthropods is not a lack of moisture in the environment, but rather inundation by the sea, which may last too long. On the other hand, since they are adapted to living in genuinely marine habitats, like regularly flooded cliffs and islets, their ability to survive flooding by the sea is amazing, and they can sometimes survive such situations for up to several months. During this time they of course remain air-breathing, often adopting the principles of “physical lungs” (plastron respiration). Moreover, many of the thalassobiotic arthropods have a resting stage with only minimal oxygen requirements. Osmoregulation remains effective through their coxal glands. The adaptations are so specific that survival after inundation is only possible in sea water, not freshwater. As an indication of their terrestrial descent, the foods of this ecologically defined assembly of animals are not of marine origin. Terrestrial detritus, fungi, lichens and carrion washed ashore on even the most isolated islands is sufficient to sustain a small outpost of strange terrestrial life in the marine biome. More detailed reading: Schuster (1965, 1979).
5.11
Annelida
The traditional Articulata concept, which unites Annelida and Arthropoda, is largely based on the monophyly of segmentation. It contrasts with a growing set of morphological and molecular (including Hox genes) studies (Peterson and Eernisse 2001; DeRosa et al. 1999) in which molluscs, annelids and sipunculids are combined into a taxon termed “Lophotrochozoa,” thus separating Annelida
208
5 Meiofauna Taxa: A Systematic Account
from Arthropoda. The implication of this is that serial segmentation is more plastic than originally thought during evolution, and so it cannot serve as the pivotal character that bonds large groups of phyla together. Acccordingly, we arrange the Annelida in context with Sipunculida and Mollusca here. Although there is now a consensus among specialists that Annelida is a rather diverse group, with the “polychaetes” and “oligochaetes” being paraphyletic (McHugh 2000, 2005; Rouse and Pleijel 2006; Erséus et al. 2008), we retain these colloquial terms in this ecologically oriented book in order to enable a broader understanding. The actual morphological and phylogenetic status of the Annelida and especially the Polychaeta is summarized in Bartolomaeus and Purschke (2005) and Rousset et al. (2007).
5.11.1
Polychaeta
In any meiofaunal sample, polychaetes are among the most striking and beautiful animals due to their multitude of structures, sizes and movements. Although there are only about 250 polychaete species of meiobenthic size, belonging to approximately 25 families, their abundance is fairly high, usually ranking fourth in meiofaunal samples. Besides the typical polychaetes that belong to the meiobenthic size spectrum as adults (permanent meiofauna, see below), many polychaetes pass through a juvenile phase in the meiobenthic size range (temporary meiofauna). In particular, juvenile Hesionidae, Syllidae and Capitellidae are often found in meiobenthic samples. Many of the meiobenthic forms were previously classified as a separate group, the “Archiannelida.” This group was characterized by rather aberrant features (e.g., irregular or absent segmentation, reduced parapodia, ciliary rings), which were held to be archaic. The structure that was often regarded as the most valid one for unifying the archiannelids was the ventral pharyngeal bulb, which is everted by a complicated musculature used to dab up bacterial and diatom epigrowth from sand grains. It has been shown, however, that the structures that form the ventral pharyngeal bulb are not homologous, that the similarities are convergences, and so they defy a synapomorphy for all the archiannelids (Purschke 1988). Today, it is well established that no single synapomorphic character can unify the “Archiannelida,” a fact documenting that this taxon is artificial. Rather, archiannelids are a convergent assembly of about 60–100 meiobenthic species of aberrant polychaetes belonging to approximately 12 families with numerous reductive, neotenic or highly modified features (Worsaae and Kristensen 2005; see below). It is, however, problematic to group them within the polychaetes, since polychaetes themselves cannot to be defined by clear synapomorphies (McHugh 2000). There are segmented worms that are so aberrant that classifying them as “Annelida incertae sedis ” (e.g., Aeolosomatida, Lobatocerebrum, see Sect. 5.11.3) would be more appropriate than forcing their classification within the traditional annelid subgroups, polychaetes and oligochaetes. With the absence of any segmentation in the Diurodrilidae, even their assignment to annelids is
5.11 Annelida
209
questionable. Some of these strange forms play a central role in phylogenetic discussions. Many characteristic convergent adaptations typical of meiobenthic and particularly interstitial life are beautifully realized among polychaetes (Fig. 5.42): – mostly very small (mature Nerillidium are only 250 µm long!), with only a few segments, – parapodia often reduced and not protruding (Protodrilus); sometimes even chaetae are reduced (Polygordius), – often ciliated, with ciliary rings or a ciliated “creeping sole” (with gliding locomotion; e.g., Dinophilus, Trilobodrilus, Ophryotrocha), – no circular musculature, no peristaltic movements. Some of the features mentioned above clearly relate to the progenetic nature of many interstitial annelids (Westheide 1987a; Worsaae and Kristensen 2005; Struck 2006), which is supported by other structures: the epithelial nerve system, the simplification of the dermal ultrastructure, no circular musculature of the vascular system, no coelomic cavities. Other morphological features seem to be secondary adaptations to the void system of the sediments: – Threadlike form with many segments (Polygordius is up to 10 cm long with up to 185 segments!) – Flattened body with a ventral “creeping sole” (Protodrilus) – Numerous adhesive glands on parapodia and caudal appendages (Hesionides, Sinohesione) – Eyes and pigments reduced – No planktonic trochophore larva – Complicated genital organs, often hermaphroditic (Ophryotrocha); with copulatory structures (Hesionides, Microphthalmus, Questa, Sinohesione)
5.11.1.1 Taxonomy and Classification The wealth of structural details useful as diagnostic features makes it possible to identify meiobenthic polychaetes to the generic level in most cases; often even the species can be identified without further dissection (Westheide 1990). Below, some of the more frequent and interesting genera are introduced and commented upon; their classification into families follows Westheide (1988, 1990a). Polygordiidae: Polygordius (with 15 species) occurs worldwide in the lowest tidal and subtidal of sandy shores (Nordheim1984; Villora-Moreno 1997). Together with Nerilla (see below) the genus was one of the first “archiannelids,” discovered in 1848 in sublittoral sand near the Island of Helgoland, Germany. The long, threadlike body with its smooth surface that is devoid of any appendages or setae and lacks circular musculature seems rather nematoid and makes the
Saccocirrus
Fig. 5.42 Some typical meiobenthic Polychaeta. (Various authors)
Polygordius appendiculatus
Protodrilus sp.
500 µm
1 mm
2 mm
Trilobodrilus axi
Nerillidium troglochaetoides
100 µm
Hesionides arenaria
100 µm
Stygocapitella subterranea
100 µm
200 µm
210 5 Meiofauna Taxa: A Systematic Account
5.11 Annelida
211
genus easily recognizable. The numerous segments are not visible externally. The two prostomial tentacles are short and stiff; the pygidium with adhesive glands is set off. Protodrilidae: A family that may not be monophyletic, with two genera among which Protodrilus has about 30 species. They are distributed worldwide (Nordheim 1989), occurring mainly in the sublittoral but also in sandy tidal flats. The slender body consists of many discernible segments without parapodia but with setae visible. A continuous ventral band of cilia serves for gliding locomotion. Circular musculature is lacking. In contrast to Polygordius, the two head tentacles (palps), which easily break off, are flexible and longer. The pygidium has 2–3 lobes, which are of diagnostic relevance. The animals are hermaphrodites; some species have larval dwarf males which attach to the partner. They produce spermatophores and fertilization of the eggs is internal. The larva is known to metamorphose only in “natural” sand from its habitat, probably due to a certain bacterial composition (Gray 1966a). The two species of Protodriloides, characterized by numerous refringent epidermal glands, lay their eggs in cocoons; fertilization is external, possibly within the cocoons. Parenterodrilus is a gutless protodrilid from coral sand in the Pacific Ocean (Jouin 1992). Saccocirridae: Saccocirrus has 18 species distributed worldwide in eulittoral and sublittoral coarse sand. They are thin and very active worms with short parapodia and setae, one pair of eyes and two sticky pygidial appendages. The long and flexible tentacles are used as tactile probes. In some species, the males use eversible papillae in their genital segments as copulatory organs; the females have corresponding spermathecae. In other species of Saccocirrus, spermatophores are attached to the female partner. Saccocirrus is probably predaceous. Nerillidae: Most of the >50 species (classified into 17 genera of which Nerilla and Nerillidium are the best known) are small polychaetes (<1 mm long) with only 7–9 segments. They are ventrally ciliated, have characteristic, slightly lobate anterio-lateral palps, and relatively long parapodia and setae. The pygidium has anal cirri; all appendages are very flexible. Development is mostly direct; eggs are sometimes brooded in an epidermal “hood.” Ecologically, Nerillidae are very diverse: some (Mesonerilla) are limnic or troglobitic (Troglochaetus), while some live in symbiosis with bacteria (Trochonerilla). Often Nerillidae are brought into experimental seawater systems accidentally (e.g., Tzetlin and Saphonov 1995). Some species occur worldwide, which is a much-discussed phenomenon considering their limited means of distribution. Although many species possess jaws, they are microphagous bacteria and diatom feeders. Dorvilleidae: Besides numerous macrobenthic forms there are characteristic interstitial genera such as Ophryotrocha, Parapodrilus and Apodotrocha. Previously considered to belong to separate (archiannelid) families, Westheide (1984) clarified their relationship to dorvilleid polychaetes. The approximately 70 meiobenthic species, which each glide on a ventral ciliated “creeping sole,” are mostly progenetic, with only a few segments of larval character. As with nerillids, dorvilleid species (particularly Ophryotrocha) of uncertain origin suddenly “occur” in seawater aquaria. Parthenogenesis is considered possible in this group. Ophryotrocha
212
5 Meiofauna Taxa: A Systematic Account
species are protandric hermaphrodites, protecting their eggs in a cocoon. They have conspicuous jaws. Parapodrilus adults are only 0.6 mm long. Dinophilidae are now considered to belong to the Dorvilleidae (EibyeJacobsen and Kristensen 1994). Dinophilus, discovered in 1848, comprising about ten sometimes pigmented species, represents the most reduced form in a progenetic regressive evolutionary series. The mostly clumsy body of Dinophilus has ciliary rings but lacks tentacles and appendages. Males and females have different numbers and sizes of chromosomes (see Simonini et al. 2003). After phenotypical sex determination, the dwarf males inject their sperm into the females. This copulation is performed directly after hatching from the eggs, and sometimes while still in the egg cocoon. Trilobodrilus is one of the meiobenthic polychaetes that frequently occurs in North Sea tidal sands. Diurodrilidae: Six species of Diurodrilus have been defined on the basis of their primitive, although not larval, organization as “primary small interstitial annelids” (Worsaae and Kristensen 2005). They mostly live high up on tidal shores (Villora-Moreno 1996a). Recently, the existence of any segmentation has been challenged, which would make their annelid nature questionable (Worsaae, pers. comm.). Their similarity to dorvilleids is probably superficial. Hesionidae: Hesionides and Microphthalmus are common meiobenthic genera within this polychaete family. Both have long, flexible parapodia and setae. Hesionides moves quickly over the sand grains, resembling chilopod centipedes, with the parapodia orientated ventrally. The genera and species differ in the arrangement of prostomial tentacles and parastomial cirri and the formation of anal lobes equipped with adhesive glands. They are hermaphroditic species with complicated copulatory organs, spermatophores and encapsulation of eggs in cocoons. Sinohesione has unique external genital organs (Westheide et al. 1994). Syllidae: This large family of Polychaeta contains numerous small, meiobenthic forms belonging to genera such as Sphaerosyllis, Streptosyllis, Petitia, Exogone and Brania. Most of them have a conspicuous barrel-shaped proventricular musculature. The parapodial tentacles often have a characteristic articulated structure. Some species have eyes. Representatives of meiobenthic syllids occur both in sand and mud, but most often between coral rubble and encrusting epiphytic algae. Some live in the interstitium of sponges, such as Amblyosyllis sp. depicted in the cover photograph. Paraonidae: Several genera; chaetae are bifid crotchets; species often have a dorsal anus; complex genital organs; frequent genera are Aricidea and Levinsenia. Pisionidae: The mostly small, psammobiotic species (the best-know genus is Pisione) of this family occur worldwide, even in continental groundwater. They have many adaptations to the interstitial system: adhesive glands, complicated genital organs and reproductive patterns with numerous copulatory appendages (males with penes, females with spermathecae, see Fig. 5.43). Psammodrilidae: An isolated group of aberrant, neotenic polychaetes (Psammodrilus, Psammodriloides) related to the Maldanidae and Arenicolidae (Meyer and Bartolomaeus 1997), with complete body ciliation; no appendages, but with three pairs of long cirri. They live interstitially, some of them in a housing that
5.11 Annelida
213
+
+
Fig. 5.43 Copulatory structures in the meiobenthic polychaete Pisione remota. (Ax 1969)
is glued to sand grains. The validity of the genus Psammodriloides is doubtful (Worsaae and Sterrer 2006). Parergodrilidae: Stygocapitella and Parergodrilus are aberrant progenetic annelids whose morphological affiliation to polychaetes was recently recognized and is now also supported by molecular analyses (Rota et al. 2001; Struck et al. 2002; McHugh 2005; Rousset et al. 2007). Earlier, they were often assigned to oligochaetes. Stygocapitella, long considered an enchytraeid oligochaete, has no head appendages or parapodia, but it does have a characteristic setation (long hair setae in segment V) and a conspicuous S-shaped intestinal curvature in the posterior body (see Fig. 5.42). It occurs regularly in the supralittoral rather dry zones of beach sands. The monotypic species, S. subterranea, may, in fact, consist of a complex of cryptic species. Its slow movements and the features given above make it easily recognizable. Parergodrilus is a rare terrestrial monotypic genus that sometimes co-occurs with another soil-dwelling annelid, Hrabeiella, whose phylogenetic position is unclear.
5.11.1.2
Biological and Ecological Aspects
Many meiobenthic polychaetes are euryoecious and well adapted to living even under the fluctuating conditions of eulittoral beaches. Along some sandy coastlines, e.g., in the Mediterranean, they can become so abundant locally that they provide 30–40% and sometimes even 50% of the total meiofauna—more than nematodes (Villora-Moreno 1997). Many interstitial polychaetes often have distinctly zoned distribution patterns on beaches (Fig. 5.44), showing clear preferences for ecological gradients. These preferences probably relate to the sediment structure and
214
5 Meiofauna Taxa: A Systematic Account dry sand d
an
ts ois
slo
V
III
m e:
p
IV
VI
II I
−10
0
2
4
6
8
10 m
Fig. 5.44 Differentiated distributional pattern of some interstitial polychaetes along a beach profile. I, Hesionides gohari; II, Diurodrilus sp.; III, Hesionides arenaria; IV, Diurodrilus benazzi; V, Protodriloides sp.; VI, Stygocapitella subterranea. (After Westheide 1972)
organic content (Westheide 1972; Villora-Moreno 1996a, 1997). They occur and breed most frequently in the warmer season. Some interstitial polychaetes have entered the continental groundwater system from the often brackish sea shores (Meganerilla, Thalassochaetus), Troglochaetus became troglobitic and Parergodrilus and Hrabeiella even live in (moist) soil far from the sea. In many tropical regions of the Pacific Ocean, annelids play a major role among the meiobenthic groups, often second only to nematodes. However, polar subtidal sediments can also harbor about 500 ind. 10 cm−2 (De Skowronski and Corbisier 2002). Their (local) abundance and hardy nature have made meiobenthic polychaetes favorite animals for experimental work: Protodrilus was shown to “recognize” its natural sand due to a preference for particular bacteria (Boaden 1962; Gray 1966a). Ophryotrocha and Dinophilus have long served as convenient culture animals for bioassays and studies on speciation and sex determination (Åkesson 1975; Pfannenstiel 1981). Respiration experiments have been performed with Protodrilus and Ctenodrilus (Boaden 1989a). Ctenodrilus has been studied in detail for ontogenetic determination processes at the level of gene sequences and expression (Dick and Buss 1994), and for its sexual reproduction pattern. Nordheim (1984) estimated the lifetimes of many interstitial polychaetes around the island of Helgoland at about 10–15 months. Despite the frequent lack of planktonic larvae in meiobenthic polychaetes, the distribution of many genera and species is fairly wide. Identical genera, sometimes even species, can be found on both sides of the Atlantic, and occasionally have a cosmopolitan occurrence. Considering the often highly apomorphic body organization, it seems doubtful that this wide distribution derives from the old age of a formerly ubiquitous animal group. Molecular scrutiny will be needed to verify the uniformity of disjunct populations (see Chap. 7)
5.11 Annelida
215
More detailed reading: taxonomy and reproductive biology, phylogeny, Westheide (1984, 1985, 1987a); Rouse and Fauchald (1997); Rouse and Pleijel (2001, 2006); McHugh (2000, 2005); Worsaae and Kristensen (2005); Bartolomaeus and Purschke (2005); monograph on archiannelids, Remane (1932).
5.11.2
Oligochaeta
Although modern taxonomy tells us that the names Clitellata and Oligochaeta are synonymous, the vernacular term Oligochaeta is retained here for a broader understanding. The same applies to one of the dominant subtaxa. the tubificids, which taxonomically correct are now termed Naididae with several subgroups, among others the new taxa Tubificinae and Naididinae (Erséus et al. 2008), Oligochaeta, considered to be either primarily freshwater inhabitants (Erséus 2005) or to be of terrestrial origin (Purschke 2002), contains about 750 meiobenthic species, of which about 450 have radiated intensively into marine habitats, with some of them even occurring in the deep-sea. Most of the numerous interstitial species belong to the (former) Tubificidae, followed by the Enchytraeidae. The families are largely distinguished by the position of the genital segments in combination with features of setation (Fig. 5.45). Genera and species are mostly defined by setation and genital organ details. Potamodrilida and Aeolosomatida do not have genuine oligochaete features; they are covered here as “Annelida incertae sedis.” When sufficiently small, the body organization of an oligochaete does not require specific adaptations for a meiobenthic life: oligochaetes never have any head tentacles or protruding parapodial appendages, their chaetae are retractable or flexible. In contrast to polychaetes, they have only fairly simple (never “compound”) chaetae (Fig. 5.45). All Oligochaetes are hermaphrodites with complicated genital structures and copulatory modes. Only the aberrant tubificid Mitinokuidrilus is a functional gonochorist (Takashima and Mawatari 1998). Reduction phenomena are frequent: planktonic larvae are not developed; eyes and a marked coloration are mostly lacking. 5.11.2.1 Taxonomy and Classification (note remarks on new classification above) Tubificidae: About 300 marine tubificid species from numerous genera belong to the meiobenthos occurring preferably in sandy sediments, while the frequent limnetic forms are mostly of macrobenthic size. The characteristic bifurcate chaetae can be modified into pectinate and hair chaetae. Many have salient “penial chaetae” in segment XI and some are equipped with penis-like structures for sperm transfer. Identification is based on a combination of metric, chaetal and genital characteristics (usually mature specimens are required). Among the numerous sand-living species, there are many characteristic adaptations to the interstitial: thread-like,
216
5 Meiofauna Taxa: A Systematic Account
Aktedrilus (Tubificidae)
Enchytraeus (Enchytraeidae)
Fig. 5.45 Some typical meiobenthic Oligochaeta and their chaetae (below). Length of the animals about 2 mm
elongated (Olavius longissimus) as well as flattened bodies, with tails (O. planus), adhesive glands, and turgescent cells for stabilization (Aktedrilus monospermathecus). The genera Inanidrilus and Olavius, comprising numerous species, live in symbiosis with sulfur bacteria, which adapts them to a thiobiotic life and allows for the complete reduction of gut and nephridia (see Sect. 8.4). Other frequent marine genera of meiobenthic tubificids are Heterodrilus, Limnodriloides, Thalassiodrilus and Phallodrilus spp. Among the phallodriline tubificids, a few species of the normally marine genera have entered the stygobios (e.g., Abyssidrilus, Gianius) and are encountered in caves and groundwater. Enchytraeidae: While most members of the family occur in terrestrial biotopes, there are an estimated 200 meiobenthic species from aquatic biotopes, some of which are adaptated to the interstitial habitat: the usually simple-pointed chaetae (see
5.11 Annelida
217
Fig. 5.45) occasionally become more or less reduced (e.g., Achaeta, Marionina subterranea, M. achaeta), and the body is thread-like and elongated (Grania sp.). In this genus of about 70 species, the solid longitudinal musculature reduces peristaltic movements and leads to a more nematoid wriggling motion. Grania also has a complex statocyst (Locke 2000), an unusual feature for oligochaetes. Some Marionina—the other speciose meiobenthic genus (80 spp.?)—can attach to sand grains by secreting coelomic fluid. In Lumbricillus and Marionina the species are mostly marine, occurring predominantly in the upper eulittoral. Grania is exclusively marine and contains species that occur even in deep sublittoral and deep-sea sediments; some have been reported from Antarctic benthos (Rota and Erséus 1996). The blackish Mesenchytraeus solifugus populates the fissures in Alaskan glacier ice. Living on microalgae on and in the ice, this species is optimally adapted to temperatures just below 0 °C (Shain et al. 2001). The genera Enchytraeus and Henlea, frequent in the upper shore and in salt marshes, combine terrestrial, limnic and marine species. Most enchytraeids are extremely tolerant to fluctuations in abiotic factors (except oxygen supply) and are abundant in the wrack zone of the seashore, where they have an important role in degrading the debris washed ashore (Giere 1975; McLachlan 1985). An exotic microhabitat, the plant parenchyma of Spartina stems, is the preferred habitat of a few Marionina species (Healy 1994; Healy and Walters 1994). The identification of enchytraeids requires a combination of metric, chaetal and genital features. Since the shape of the spermatheca is of particular importance, species determination is only possible in mature specimens. Naidinae: This group comprising about 200 mostly meiobenthic species, has to be ranked as a subgroup of the tubificid oligochaetes as a result of cladistic analyses (Erséus and Gustavsson 2002). Most naidines live as epibenthos in the limnetic phytal (Nais, Dero); only a few have entered brackish habitats (e.g., Nais elinguis, Paranais litoralis, Amphichaeta sannio). The chaetae of most naidines are conspicuously bifurcated and have a nodulus. Long hair setae occur frequently in dorsal bundles. Sexual maturation with the development of genital organs is rare, and often linked to decreasing water temperatures. Hence, identification must be based on the form, arrangement and metric relations of chaetae. The predominantly asexual reproduction by budding, fission or fragmentation leads to sudden outbursts of naidid populations, often in tune with blooms of their favorite food, benthic diatoms. Dissolved free amino acids are also of nutritive relevance for naidines (Petersen et al. 1998). As an exception, Chaetogaster is predominantly predaceous. Lumbriculidae: The subterranean freshwater genera Dorydrilus and Trichodrilus belong to the continental groundwater fauna, the stygon, and contain a few truly meiobenthic albeit rare species with body lengths of < 10 mm (see Sect. 8.7.2).
5.11.2.2 Biological and Ecological Aspects Most oligochaetes do not reproduce before their second year, while the naidids have an annual lifespan. Reproduction in the limnetic and boreal eulittoral forms is mostly seasonal (often in summer/autumn), in contrast with the numerous marine species
218
5 Meiofauna Taxa: A Systematic Account
from subtropical to tropical habitats, which apparently reproduce continuously. Oligochaetes in tropical beaches were found to restore their populations particularly rapidly after a general massive decline in meiofauna due to monsoon rains (Alongi 1990b). In New Zealand beaches oligochaetes seem to be scarce, which might also explain the low numbers of otoplanid turbellarians, their main predators (Riser 1984). In polar shore sediments, which harbor oligochaetes in relatively high abundance, mainly enchytraeids contributed to the degradation of experimentally induced detritus among the meiobenthos (Urban-Malinga and Moens 2006). In mangroves, experiments showed that the numerous oligochaetes are linked into the detrital food chain. More specifically, food competition for detritus between meiobenthic and macrobenthic species was postulated (Schrijvers and Vincx 1997). Despite the existence of some deep-water forms, marine oligochaetes are essentially eulittoral and sublittoral meiofauna. Especially in the upper, often waterunsaturated shoreline, enchytraeids can dominate in abundance all other meiofauna while in the lower tidal flats tubificids are the more frequent oligochaetes. They represent, particularly in the fine sediments of deeper horizons, the typical interstitial marine oligochaete (Fig. 5.46). In limnetic habitats, especially among plant aufwuchs, the predominant meiobenthic oligochaetes are naidids. Eutrophic lake bottoms are often densely populated by tubificids. Freshwater oligochaetes are usually an important member of the benthos in any limnetic habitat (see Sect. 8.7). Macrobenthic species whose juveniles belong to the temporary meiofauna play an enormous role in limnetic food webs and in pollution ecology. Accordingly, their life histories have been thoroughly studied. Permanently meiobenthic relatives are rarer. In the continental groundwater and hyporheic system about 60 small species have been recorded, belonging mainly to lumbriculids, haplotaxids and rhyacodrilids. Some marine tubificid genera (Aktedrilus) have species that occur in groundwater (Sambugar et al. 1999). Vertical distribution is mostly limited by oxygen supply to the uppermost layers, except in the case of gutless thiobiotic tubificids, which accumulate in slightly sulfidic sediments of warm regions around the oxic/sulfidic interface (Giere 1975; Giere et al. 1991). Where oligochaetes are abundant, they can become a significant food source for other meiobenthos (e.g., halacarids, turbellarians) and for small fish. More detailed reading: taxonomy, identification, Brinkhurst (1982a); Erséus (1980, 1984, 1990b); phylogeny, Brinkhurst (1982b, 1984, 1991); Erséus (1987, 1984b, 1990a; 2005); Erséus and Källersjö (2004); reviews, Giere and Pfannkuche (1982); Giere (2006).
5.11.3
Annelida “Incertae sedis”
Progenetic trends may have played a major role in the development of these rare but possibly phylogenetically relevant meiobenthic worms. Potamodrilida and Aeolosomatida: These “microannelids” (Fig. 5.47) do not match the usual definitions of the larger groups above and have been grouped as “Aphanoneura” (Brinkhurst 1982b) into a separate annelid taxon. While their
naidids, tubificids (meiobenthic and macrobenthic); a few enchytraeids (in sand)
eulittoral and shallow bights tidal : fine sand, mud atidal: mostly mud
terrestrial enchytraeids, e.g. Marionina; a few tubificids, e.g. Aktedrilus, Phallodrilus
low shore wrack moist, often zone medium sand
terrestrial enchytraeids
high shore, beach dry sand
supralittoral
Fig. 5.46 General distribution pattern of marine meiobenthic Oligochaeta. (After Giere and Pfannkuche 1982)
tubificids (meio- and macrobenthic); a few enchytraeids, e.g. Grania (in sand)
sublittoral mud or sand
5.11 Annelida 219
220
5 Meiofauna Taxa: A Systematic Account
Fig. 5.47a–b Representatives of the annelid groups Aeolosomatida (a) and Potamodrilida (b). (Various authors)
sister group relationship has been confirmed by molecular analyses (Struck and Purschke 2005), their systematic links to other annelids remain unclear, particularly since miniaturization (body lengths are only 200–1,000 µm) and progenesis have led to various reductive phenomena. The few segments are not always clearly delimited, the nerve cord is epithelial, a ventral ciliation (particularly of the prostomium) is responsible for gliding movements, the setation consists of hair setae only, etc. Potamodrilida live stygobiotically in the gravel and sand on river banks. Potamodrilus fluviatilis, a hermaphroditic species that only reproduces sexually, has just seven segments and a typical “archiannelid-like” protrusible ventral pharyngeal bulb (see Fig. 5.47). Aeolosomatida consists of about 25 species characterized by numerous epithelial “oil glands” of different colors. Aeolosomatids occur preferably in the phytal and epibenthic detritus layers of more or less stagnant, oxygen-rich freshwater habitats (Aolosoma hemprichi). A. litorale is common in brackish coasts; A. maritimum was reported from the interstitial of a Tunesian beach. Rheomorpha neizvestnovae is a rare species without
5.12 Sipuncula
221
setae from the hyporheos of boreal rivers and lakes Fig. 5.47). Aeolosomatids normally exhibit asexual reproduction by paratomy, leading to sudden mass multiplication and chains of 2–8 zoids. In the few mature specimens obtained, the spermatheca consists of just one cell and a clitellum is not developed. Lobatocerebrida: In 1980, Rieger described the genus Lobatocerebrum from coarse sands off the North Carolina coast (USA). The vermiform, fully ciliated animal lacks any setation and segmentation. Its body cavity is filled with mesenchymal cells; a mesodermal lining is not discernible. The mouth and pharynx are ventrally located, and the genital system is hermaphroditic. Thus, superficially, it can be characterized as a turbellariomorph. However, a well-developed hindgut and anus, remnants of a blood vessel system and a strange inversion of the muscular layers contradict any body plan of turbellarians. Additionally, many autapomorphic histological details do not allow an alignment into the Plathelminthes. They instead suggest that the peculiar Lobatocerebrida, with its three species, should be defined as “secondary acoelomates.” Rieger (1991) argued that the Lobatocerebrida might be examples of an evolutionary line from coelomate ancestors to acoelomate flatworms. Also, on the basis of a combined morphological and molecular study, Zrzavy (2003) attributes a basal position (“basal Trochozoa”) to the group, but denies any relation to a simplified interstitial annelid. Jennaria pulchra: This peculiar slender worm has been found in fine sand at the highwater line along the US Atlantic coast (Rieger 1991; Rieger and Rieger 1991). Its ciliated, unsegmented body, which is only about 2 mm in length, is divided by grooves into a rostral and a caudal part set off from the trunk. The body musculature consists almost exclusively of unicellular, obliquely striated longitudinal muscle cells. The mouth is ventral and the anus terminal. The body cavity is filled with large enchymatous cells and contains some protonephridia. The genital system is possibly gonochoristic, so they probably have separate sexes. Like Lobatocerebrum, this worm had also been considered a “secondary acoelomate” that originated as an isolated evolutionary line from the annelid stock or a closely related unknown root. The cladistic studies of Zrzavy (2003) vaguely suggest a “basal spiralian.” More detailed reading: Aeolosomatida and Potamodrilida, Bunke (1967); Parergodrilida, Hrabeiella, Rota et al. (2001); Lobatocerebrum, Jennaria – phylogenetic considerations, Rieger (1991); Zrzavy (2003).
5.12
Sipuncula
Only two species of this phylum of unsegmented marine coelomates with affinities to the annelids (Halanych et al. 2002) can be considered permanent meiofauna. Other sipunculans are temporary meiofauna during their juvenile stages. The body of Sipuncula is divided into a posterior trunk and a narrower anterior introvert that can be retracted by ventral muscles. At the anterior end, the terminal mouth opening is surrounded by a tentacular crown of varying structure. In close vicinity to the
222
5 Meiofauna Taxa: A Systematic Account
Box 5.12 Annelid Evolution Towards the Meiobenthos: Opposite Pathways A sample decanted from the sandy marine littoral usually harbors meiobenthic annelids. Despite their minute size, a zoologist’s eye would quickly recognize the well-segmented oligochaetes with their uniform bodies that lack head appendages and their characteristic chaetae. As different as the habitats may be, this body structure is mostly maintained across all meiobenthic clitellates, with variations predominantly found in the segments with complex genital organs. However, then we also find various “worms” in the sample, which are for the untrained eye hardly to recognize as annelids: these are the meiobenthic (interstitial) polychaetes such as Polygordius, Diurodrilus and Psammodrilus. Often no segmentation, parapodia or chaetae are visible, but strange rings of cilia and furcate tail appendages are developed—not very typical of a polychaete! During their evolution, all of the characteristic features of annelids appear to have been disposed of and any unifying structural frame rescinded. Nevertheless, the various habitats colonized by both groups are not so different. The clitellate annelids radiated from a primary limnetic origin into groundwater, cave ponds, migrated along with numerous species into the marine littoral, and even reached the bottom of the deep sea. The non-clitellate annelids, the so-called “polychaetes,” which are genuine inhabitants of the sea, conquered all of the marine habitats from exposed beaches to abyssal depths with meiobenthically small species. They also developed species that are able to live in fresh and groundwaters, and even in the soil. Both annelid groups propagated along different pathways into similarly diverse habitats, the clitellates without any planktonic stages and the ancestral polychaetes either with planktonic larvae or perhaps progenetically without this propagative phase. On the route towards meiobenthic life, we find in clitellates conservatism in external morphological features, while a lot of variation evolved in the genital organs. In the non-clitellates, high morphological plasticity even nullified the unifying features of the group called “polychaetes.” Two contrasting evolutionary principles with opposite radiative pathways—one can only speculate about the reasons for this divergence.
anterior tip of the body is the anal opening. Thus, what seems to be the posterior body end is morphologically a median “sac.” Internally, the intestine is coiled with descending and ascending portions. While macrobenthic Sipuncula live mostly in crevices and holes of porous rocks, in animal tubes and in gastropod shells, the tiny and almost transparent meiobenthic species have been found in shelly coarse sand in the Atlantic sublittoral off Florida (Phascolion sp.) and in the brackish interstitial of intertidal sands from some Caribbean islands (Aspidosiphon exiguus). These species are about 4 mm in length, but their flexible bodies can squeeze through narrow voids in the sand. Aspidosiphon has two characteristic “shields,” hardened areas of the skin. More detailed reading: Edmonds (1982); Gibbs (1985).
5.13 Mollusca
5.13
223
Mollusca
Morphological and molecular data place Mollusca in a lineage with Annelida and Sipuncula, which together form the Lophotrochozoa (De Rosa et al. 1999; Peterson and Eernisse 2001; Garey 2001). The calcareous shells of molluscs seem to have developed independently several times for different molluscan subgroups (Scheltema 1985). After metamorphosis most molluscs pass through an initial benthic phase during which they belong to the “temporary meiofauna” and can become ecologically important (Elmgren 1978). Later, they grow to macrobenthic size. However, there are numerous “microgastropods” and “microbivalves” which remain about 1–2 mm throughout their lives and are therefore permanent meiofauna. Members of the heterogeneous group of microgastropods that largely belong to the Opisthobranchia occur globally and tend to show internal fertilization and brooding. They preferably live in the interstices of well-oxygenated medium to coarse sand (Fig. 5.48). The microbivalves, mostly protobranchs, live as detritivores in the uppermost sediment layer. Occurring preferably in deep-water sediments, they grow extremely slowly and can reach ages of 50 years or more (Haszprunar, pers. comm.). Because of their remarkable deviations from the typical mollusc body plan, a mixture of “idioadaptations” and regressive features, it is sometimes not easy to recognize their small and vermiform bodies as being those of molluscs, especially when fixed. They can be confused with turbellarians (Rhodope) and nemerteans (Philinoglossa, Platyhedyle). While most of them retain a (simple) radula as the main synapomorphic structure of molluscs, they have reduced their gills, ctenidia and in many cases their shells, suppressed the planktonic veliger larvae, simplified the gonads and digestive gland, and became whitish-opaque without pigments. On the other hand, most meiobenthic gastropods have developed rhinophores and tentacles in a specific arrangement at their anterior ends, and some have subepidermal spicules to stabilize their bodies (Hedylopsis, Rhodope, see Figs. 4.9, 5.48). Most interstitial gastropods have an enhanced number of adhesive glands in the epidermis. The aplacophoran molluscs are externally characterized by their coats of cuticular spines and needles which are thought to provide protective and reinforcing functions and to be of locomotory relevance. The small number of meiobenthic genera facilitates an introduction to some of the more common or characteristic forms.
5.13.1
Monoplacophora and Aplacophora
The archaic group of monoplacophoran molluscs is represented by almost 30 species, many of them of meiobenthic size (Micropilina, partly Rokopella). They are mostly found in deep-sea sands (Warén and Gofas 1996, Hazprunar and Schäfer 1997; Marshall 2006). Details on their ecology and biology are largely unknown. The Aplacophora (Fig. 5.48–1,2) are peculiar and probably ancient marine molluscs (Scheltema and Ivanov 2002). Discovered in 1844 off the west coast of Sweden (Scheltema 1998), almost 200 species are now known. They have vermiform,
3
4
6
5
7
50 µm
8
9 500 µm 500 µm
1 mm
500 µm
500 µm
100 µm
Fig. 5.48 Some meiobenthic Mollusca: 1, Meioherpia; 2, Prochaetoderma; 3, Hedylopsis spiculifera; 4, Microhedyle; 5, Philinoglossa; 6, Embletonia; 7, Rhodope; 8, Pseudovermis; 9, Caecum. (Various authors)
1
2 500 µm 500 µm
224 5 Meiofauna Taxa: A Systematic Account
5.13 Mollusca
225
bilateral-symmetrical bodies without a shell (the name of the group derives from the Greek for “not bearing a shell”) but with many spicules arranged in a characteristic manner. Neomeniomorpha (= Solenogastres s.str.): These are characterized by a longitudinal pedal groove (raphe) with mucus glands, and by many refractive calcareous spines and plates. These hard structures can be hollow (Biserramenia) or solid (Lepidomenia, Meiomenia, Meioherpia); their shapes and arrangements are taxonomically relevant features, as are the positions and forms of the radula teeth. The Neomeniomorpha lack tentacles and eyes and exibit many reductions; they are anatomically well adapted to living mostly in the interstitial of coarse sand and shell hash at shallow sublittoral depths. They also occur in colonies of hydrozoans, where they feed on the polyps. The deep-water forms (below 100 m depth down to the deep-sea) burrow slowly in muddy bottoms. Helicoradomenia spp. occurring at hydrothermal vents harbor a variety of bacterial symbionts (Katz et al. 2006). Chaetodermomorpha (= Caudofoveata). These have a frontal shield, a set of long circum-cloacal spines and a pair of posterior ctenidia; they lack a pedal groove. While most Aplacophora are larger than meiobenthic sizes, about 20 species are only 1– 2 mm long and are considered meiobenthic (Prochaetodermatidae, genera Chaetoderma, Prochaetoderma, Chevroderma). They have a rasping radula and feed on foraminiferans and detritus. Most of those reported were from the continental shelves and the upper slopes of the deep-sea, where they are common but never numerous. Lepidomenia hystrix: from the Mediterranean; Meiomenia, Meioherpia, and Psammomenia: in detritus-rich coarse sand (“Amphioxus sand”) off the coast of Roscoff (Britanny, France) and from the North Sea.
5.13.2
Gastropoda
Opisthobranchia: Acochlidioidea: about 30 spp without shells and ctenidia, with two pairs of anterior tentacles (see Fig. 5.48). Hedylopsis: Six species with long and broad posterior tentacles; with spicules in the dermis (e.g., H. spiculifera); hermaphroditic. Members of the genus occur only in sublittoral sand. Strubellia, related to Hedylopsis. Microhedyle, Unela: With short, slender posterior tentacles (rhinophores), with spermatophores and integumental fertilization. They live eu- and sublittorally in sand. Cephalaspidea: some species with an internal rudimental shell and ciliated creeping sole (Philinidae), as well as a small visceral hump. The ctenidia are sometimes reduced (Philinoglossidae). Philine, Philinoglossa, Pluscula: The vermiform body is square in cross-section. Nudibranchia: With clusters of dorsal processes (cerata), mostly containing “cleptocnides” (functional cnidocysts adopted from their cnidarian prey). Aeolidiacea: Seven meiobenthic species, some of them (as in their macrobenthic relatives) with cleptocnides in midgut gland processes. They prey on interstitial hydroids.
226
5 Meiofauna Taxa: A Systematic Account
Pseudovermidae: Several species of Pseudovermis, head end swollen, set off from body, devoid of appendages; they continuously probe the sand when creeping. Embletoniidae: Several species of tiny Embletonia that are similar to their macrobenthic relatives. Rhodopidae: This group has a doubtful position (between opisthobranchs and pulmonates). They are hermaphrodites without shells. Rhodope has numerous internal verrucose spicules (Fig. 5.48), similar to a turbellarian since foot is not delimited and radula is reduced, but with gastropod (euthyneuran) nerve structure. Discovered in 1847 (the first interstitial opisthobranch to be discovered). Prosobranchia: Omalogyra: With a coiled, operculated shell, about 1 mm in height. Caecum spp: Have a characteristic curved tubular shell with an open apical end, but juveniles have a coiled shell; adults are about 1.2 mm long; common in calcareous sediments, preferably in warm water regions, may be confused with juvenile tusk shells (Scaphopoda). Meiobenthic gastropods are characteristically found in water-saturated intertidal and subtidal sands with a well-developed interstitial system of high permeability and with a rich oxygen supply. In the deep-sea, they occur preferably on sea mounts (see Sect. 8.3). The shapes of the grain particles seem to play a major role determining their occurrence, since densely packed sediments are rarely inhabited. Coarse sands, shell hash and coralline rubble with a loose structure are preferred. The typical gastropod adaptations prevent them from being displaced or damaged by sand agitation. Only a few species are found in sands with higher organic contents or in polluted areas. In the Fiji Islands, some intersititial species have invaded freshwater habitats and can be found in mountain streams (Morse 1987). More detailed reading: taxonomy and biology of meiobenthic gastropods, Swedmark (1968); Poizat (1985); Arnaud et al. (1986); anatomy, Neusser et al. (2006); Aplacophora, Salvini-Plawen (1985); Scheltema (1985); review, Scheltema (1998).
5.14
Tentaculata
Molecular phylogenetic studies (Hausdorf et al. 2007; Helmkampf et al. 2008) indicate that Tentaculata are related to the Lophotrochozoa (Mollusca and Annelida), with Brachiopoda and Bryozoa (Ectoprocta) clustering as sister groups while the Kamptozoa (Entoprocta) are somewhat more distant.
5.14.1
Brachiopoda
Among this group of macrobenthic, sessile animals, there is one meiobenthic species (only 1 mm diameter) from the British and Brittanic (North Sea Channel) coasts, where it occurred in sublittoral shell–gravel sedimen temporarily attaches to
5.14 Tentaculata
227
the particles of shell hash. Gwynia capsula Jeffreys was described in 1859, but only much later was it recognized as an aberrant brachiopod (Swedmark 1967; Fig. 5.49A). Gwynia lives together with psammobiotic foraminiferans. Its simple lophophore and shell articulation may indicate a progenetic nature. The few eggs, which are brooded for a long period, develop into benthic larvae.
5.14.2
Bryozoa, Ectoprocta
The usually sessile and colonial bryozoan Tentaculata contain some meiobenthic, hapto-sessile species that belong to different families and live in sublittoral sediments. The rather conspicuous Monobryozoon and its relatives and the recently described lunulitiform bryozoans will be discussed here. Monobryozoon ambulans, found by Remane (1936b) in sublittoral coarse sand (“amphioxus sand”) off the island of Helgoland (German Bight, North Sea) was the first bryozoan described to live as a non-sessile, solitary organism (2 mm in length), anchored by viscid stolons (rhizoids) in the sand (Fig. 5.49B1). Although a rare animal, it has also been recorded on other European coasts. M. bulbosum and M. sandersi are congeners from the American east coast. Another, more slender, form was originally described as a new European Monobryozoon species (Franzén 1960) but is now classified as Nolella limicola (Fig. 5.49B2). It is found in deep-water muddy sediments of some West Scandinavian fjords (Berge et al. 1985). Belonging to the Gymnolaemata, the above bryozoans have a non-calcified but still fairly compact cystid which is separated from the soft-bodied retractable polypid by a deep annular furrow. With their semi-transparent yellowish color, the animals are well hidden among the sand grains. The mouth opening (“orifice”) is surrounded by a circle of ciliated tentacles. The numerous characteristic stolons are hollow ambulatory processes supplied with musculature for (slow) contractions that pull the animal through the sediment. Terminally they are equipped with adhesive glands and sensory hairs. In Monobryozoon, stout, non-contractable buds that are structurally different from the rhizoids are attached directly to the cystid. This is in contrast with Nolella, where buds arise from slender processes of the cystid, probably representing “kenozoids,” i.e., thread-like, modified bryozoan individuals. Consequently, Nolella limicola is a small mobile colony rather than a solitary bryozoan like Monobryozoon. Lunulitiform bryozoans: Today, the numerous small lunulitiform colonies (Cook 1963, 1966, 1988) of variable but mostly conical to discoid shape are grouped in various genera belonging to several families of vagile bryozoans (Fig. 5.49B3). They occur circum-mundane in deep-water sands and muds (Atlantic Ocean, Red Sea, Indian Ocean). At 1.5– 9 mm, they are, at least during their initial stages, considered meiobenthos. Their long and flexible setae, the vibracularia, help to transport food, remove foreign particles and serve as the locomotory organs of the colony. Although food
228
5 Meiofauna Taxa: A Systematic Account
0.5 mm
0.5 mm
A. Gwynia capsula
B1. Monobryozoon ambulans
0.5 mm B2. Nolella limicola
0.5 mm B3. Pachyzoon
Fig. 5.49 a–b Meiobenthic Tentaculata: a Brachiopoda; b Bryozoa; b1, b2 Gymnolaemata; b3 Lunulitiformes. (a Swedmark 1967; b1 Remane 1936b; b2 Franzén 1960; b3 Cook 1988)
can be taken up in each position, when turned upside down, the colonies follow a distinct vertical polarity, with the vibracularia pulling them back into an upright position. If buried too deeply, they can also be brought back by the vibracularia into more superficial layers. As usual, reproduction in these bryozoans is both asexual (by marginal buds) and sexual. In the conical species the apex consists of a foreign particle, e.g., a sand grain, to which the first zooid produced attaches. It subsequently forms a new colony by budding further zooids. If they are broken away, parts of the colony will regenerate. Despite their wide distribution, lunulitiform bryozoans were not discovered until recently. This is partly because of their restriction to deep-sea sites, but, as is the case for many benthic foraminiferans, they have also certainly been overlooked because of their variable and often irregular shapes.
5.15
Kamptozoa, Entoprocta
Among this isolated group of sessile animals that usually form colonies of individuals connected by stolons, there are some strange solitary representatives of meiobenthic size. Loxosoma isolata, described by Salvini-Plawen (1968) as occurring in coarse sand from the northern Adriatic Sea, is less than 1 mm in size (Fig. 5.50a)
5.16 Echinodermata
229
and has only 16 ciliated tentacles. At the base of its stalk there is a glandular disk that is used for temporary fixation onto shell fragments in coarse, sublittoral sands. Like their colonial relatives, solitary Kamptozoa seem to use their ciliated tentacles to feed on microparticles. Other currently undescribed meiobenthic species of kamptozoans have been found off North Carolina and Florida (USA) as well off Roscoff (Britanny) (Rieger, Monniot, pers. comm.), and additional forms will probably be discovered if unfixed samples of coarse shell hash are scrutinized for them. Asexual reproduction by budding has been observed. More detailed reading: see citations in the main text of this section.
5.16
Echinodermata
Most echinoderms pass through a temporary meiobenthic phase directly after metamorphosis and settlement. Judging from their abundance, particularly the numerous species of juvenile Ophiuroida can be an ecologically important faunal element in the crevices of coarse sand and shelly sediments. However, almost nothing is known about their ecological roles during this stage of their lives. Permanent meiobenthic species with characteristic adaptations to a mesopsammic life have only evolved within the holothurians.
5.16.1
Holothuroidea
Within this group of (usually) large animals, there are some species within several apodal families where the adults are only one to a few mm in length (Fig. 5.50b). Most of them belong to the family Synaptidae (Labidoplax, Leptosynapta, Rhabdomolgus, Psammothuria), but there are also some small Chiridotidae (Chiridota; Trochodota) and Myriotrochidae which can be considered meiobenthos. Apart from their size, other characters also adapt these tiny sea cucumbers to meiobenthic life in sediments. The often transparent body is rather tough and in some species calcareous platelets, the ossicles, mechanically protect the animals from abrasion. Other species (e.g., Rhabdomolgus) lack these ossicles. Conspicuous statocysts occur regularly. The mouth tentacles are strongly adhesive and their contractions pull the body slowly forward through the sediment. These large and sticky branched podial appendages are also used for food uptake (sediment particles) and for temporary anchoring. Only a few eggs are developed and brooded after fertilization through spermatophores. Leptosynapta minuta is a hermaphroditic species. The “Pentactula larva” of the meiobenthic species is not planktonic, as would be typical of the macrobenthic forms. Among the few meiobenthic species described so far, the best known are the synaptids living in shallow sublittoral coarse sand, but Rhabdomolgus ruber occurs
230
5 Meiofauna Taxa: A Systematic Account
100 µm
250 µm
a
Loxosoma isolata
b
Labidoplax buskii
Fig. 5.50 a–b Interstitial representatives of Kamptozoa (Entoprocta) (a) and Holothuroidea (Echinodermata) (b). (a Salvini-Plawen 1968; b Swedmark 1971)
in more silty bottoms and the Myriotrochidae have been found in the deep-sea. The regular occurrence of meiobenthic holothurians along the European, American and Indian coasts suggests a wider distribution than documented today. More detailed reading: Swedmark (1971); Salvini-Plawen (1972).
5.17
Chaetognatha
Besides the macrofaunal “arrow worms,” known to be marine planktonic predators, there are a few species of the genus Spadella that became secondarily epibenthic. A dwarfed, meiobenthic Spadella boucheri (1.3 mm) from coralline sand in a shallow Japanese Bay (Casanova and Perez 2000) has some structural features that indicate a life in the interstitial void system (thin hooks, little protruding, stout teeth). Recently, in coarse silicious sand samples off the island of Elba (Italy, water depth of about 12 m), live chaetognath specimens were found to quickly wriggle among the grains and were described as Spadella interstitialis (Kapp and Giere 2005). With
5.18 Tunicata (Chordata)
231
their small sizes (adult specimens 1.6 mm long, 0.2 mm wide), swift movements of the narrow, smooth bodies with confluent lateral fins (in contrast with the separate fins in their planktonic relatives), and their stiffened alveolar epidermis, these new forms represent a structurally well-adapted, genuinely interstitial meiobenthic species. This is underlined by their main occurrence well below the sediment surface at 5–10 cm depth (sometimes down to 15 cm).
5.18
Tunicata (Chordata)
5.18.1
Ascidiacea
While ascidians are usually sessile and macrobenthic, there are some 20 species with individuals of adult size 3–4 mm that are found in sand and have the typical adaptations of mesopsammic animals. These small tunicates originated apparently convergently from different orders and families, often through progenesis (Fig. 5.51a). Described by Weinstein (1961) from “Amphioxus sand” off Banyuls-sur-Mer (French Mediterranean coast), Psammostyela delamarei was the first known tunicate of meiobenthic size. Other genera are: HeterostigmaPsammascidia, Polycarpa, Dextrogaster, Heterostyela, and Molgula sp.
500 µm
250 µm
a
Heterostigma fagei
b Hexacrobylus indicus
Fig. 5.51 a–b Meiobenthic Tunicata. a Ascidiacea (Monniot and Monniot 1988). b Sorberacea (Monniot and Monniot 1984)
232
5 Meiofauna Taxa: A Systematic Account
In contrast with the sessile, macrobenthic ascidians, the body wall of the meiobenthic species is mostly transparent and lacks pigments. The sticky tunic is usually covered with sand grains or foraminiferan shells. Often the meiobenthic ascidians have the openings for ingestion and egestion currents at opposite ends. Corresponding to their minute sizes, the characteristic branchial filter system is often simplified. The planktonic ascidian tadpole larva is usually suppressed. These features indicate that meiobenthic ascidians have originated by progenetic processes. Typical adaptations to a hemi-sessile life are the numerous, mostly ventral rhizoids that anchor the body temporarily in the soft substrate. The animals move by slow contractions of the rather tough muscular body wall (Monniot 1965). In contrast to the other species representing solitary individuals, Arenadiplosoma migrans Menker and Ax (1970) is a colonial aggregate of 1–3 mm length found in the North Sea. It moves by contractions of its rhizoid-like “tunic vessels,” which have a viscid terminal end. Meiobenthic ascidians tend to avoid sediments in which their delicate filtering apparatus might become clogged and where oxygen is limited. Thus, they are found sublittorally in the well-exposed surficial layers of coarse, silicate sands (“amphioxus sand”) and seldom in muds and fine calcareous sediments. However, meiobenthic ascidians have also been encountered in fine deep-sea sediments. They have either ventral anchoring rhizoids or bristle-like rhizoids covering the tunic. The meiobenthic ascidians are oviparous and their simplified internal organs indicate progenesis (Monniot and Monniot 1988). Details on their ecology are lacking so far. Meiobenthic ascidians are mostly found in temperate and boreal latitudes, in deep bottoms of the Atlantic Ocean from northern to southern latitudes, and in the Mediterranean Sea. A few littoral species have been reported worldwide.
5.18.2
Sorberacea
This exotic group of tunicates (Monniot et al. 1975) with unique morphological and ecological characters consists of about 20 species. Many of the slowly moving individuals measure only 1–5 mm (Fig. 5.51b); Hexacrobylus; Hexabranchus; Sorbera, Hexadactylus. Encased in a tunic, the body is densely covered with adhering sediment particles. It is anchored in the sediment by rhizoids. Divergent from all other tunicates, the oesophagus does not filter. Instead, these tunicates are carnivorous; they have a wide mouth opening surrounded by six prehensile lobes or “fingers” that are well suited to grasping prey and that fold over the oral cavity. These muscular organs are well innervated and efficiently coordinated by the welldeveloped dorsal neural system. All animals have one large excretory organ (kidney) and are hermaphroditic and oviparous. Their ontogeny is thus far unknown. Vegetative regeneration and colonial forms have not been found. The intestines of these predatory tunicates have been observed to be filled with nematodes, harpacticoids, acarids and polychaetes, but also ophiuroids and gastropods. Although predominantly found worldwide in deep-sea samples, Sorberacea
5.19 Meiofaunal Taxa: Concluding Remarks
233
are known also from the continental shelf. In fact, it now seems that the first sorberacean tunicate was found at only 92 m depth off the English coast, but was described as being an ascidian (Bourne 1903). More detailed reading: Monniot (1965); Monniot and Monniot (1990).
Box 5.13 Meiobenthic Molluscs, Bryozoans, Chaetognaths, Echinoderms, and Tunicates: Exotics of Value When teaching about meiofauna, discoveries of tiny, transparent creatures that do not fit into the scheme of the traditional phyla are highly welcomed. Their aberrant body structures can wonderfully exemplify the adaptive trends that are characteristic of meiobenthic life. Adult snails creeping on a sand grain; tiny sea cucumbers showing rings of statocysts, dragging themselves with their mouth feet through the narrow passages between grains; a bryozoan on a piece of shell, with the cilia rotating in its intestine; all of these features in an animal of size 1–2 mm will impress most zoologists and their students, who still can envisage the beauty of the whole animal and its functions. But there is more: progenesis keeps the body at its larval size, and sometimes (in ascidians) larval organization is retained too. Suppression of a planktonic larval stage, even in taxa where dispersal by larvae is the rule (holothurians, ascidians), prevents unfavorable drift. Structures also change their function: feeding arms or stolons become ambulatory processes (holothurians, bryozoans), cirri of the mouth orifice become grasping arms (sorberacean tunicates); organs become simplified or reduced (filter systems in tunicates, shells and radula in gastropods). And all of these remarkable features are “wrapped up” in a transparent body wall, ready for microscopic observation - meiofauna as valuable tools for demonstrating the principles of adaptation to ecological constraints, to which evolution responds convergently.
5.19
Meiofaunal Taxa: Concluding Remarks
This account of meiofauna is far from being complete and will surely have to be supplemented in the future, particularly since recent discoveries of high-ranking taxa have brought meiofauna into the mainstream of invertebrate phylogeny. What had been only postulated now seems a realistic option: small-sized benthic metazoans were probably the first to utilize the enriched nutrients on the ocean bottom once traces of oxygen allowed complex body constructions (see Sect. 7.1; Fenchel and Finlay 1995; Chen et al. 2004). The examples of meiobenthic Sorberacea, lunulitiform Bryozoa, and archaic monoplacophoran molluscs suggest that the deep-sea is one region where further discoveries of new and exciting meiobenthos can be expected. Additionally, sea mounts, due to their unique hydrodynamics and their evolutionary interesting isolation are other sites where new and important
234
5 Meiofauna Taxa: A Systematic Account
meiobenthos can be expected (see Sect. 8.3). The recent findings of Mictacea, Cephalocarida, and, of course, Loricifera underline that even shallow bottoms, especially anchihaline caves, harbor spectacular new animal groups. With closer examination of calcareous sands (see Box 8.2) from tropical shores, better accessibility to polar sediments and thorough investigations of caves, the discovery of a large number of new animal forms can be expected, many of which will be of meiobenthic size. Also freshwater habitats (e.g., springs) have added to the set of meiobenthic exotics: the new phylum Micrognathozoa (Kristensen and Funch 2000) turned out to be of general importance for phylogenetic considerations. One need not to be a prophet to predict that the classification of higher animal taxa, phyla and classes, will greatly depend on further investigations of the meiofauna.
Chapter 6
Evolutionary and Phylogenetic Effects in Meiobenthology
Discussions of phylogenetic links between animal groups often center on small animals, particularly when remote habitats such as deep-sea bottoms, aquatic caves, and groundwater systems are envisaged. Their body organization is often simple; does this simplification and the small size result from derived or progenetic trends, or does it reflect an archaic and evolutionary primitive trait? The analysis of genetic similarities through PCR methods, new and powerful tools in our attempts to disentangle the evolution of animal taxa, is increasingly being used in studies of meiofauna, especially since new methods are being developed that can process the DNA from even very small animals. The systematic interrelationships resulting from these analyses have been considered in the previous taxonomic account (Chap. 5). Here, we will examine the role of structural specializations, paleontological findings, and distributional processes related to the evolution of meiofauna.
6.1
Body Structures of Evolutionary Relevance
The relatively stable physiographic conditions encountered by endobenthic meiobenthos living in the subsurface layers favor long-lasting evolutionary trends. Such trends often appear to follow an “orthogenetic” pattern in that the structural adaptations of the meiobenthic inhabitants appear to be directed towards an end-point. The uniformity of the physical milieu favors convergent evolution where numerous morphological and physiological features are subject to identical selective constraints. The result of this is similar and analogous “solutions” within diverse and unrelated meiobenthic groups (e.g., adhesive organs, caudal appendages, ciliation; see Sect. 4.1.2). This is particularly evident in the numerous cases of regressive evolution within the stygofauna (Sect. 8.7.2). Size reduction does not necessarily lead to simplification of a complex functional design; the elaborate structural complexity of Loricifera renders this argument questionable. The fundamental prerequisite for any phylogenetic analysis is to carefully examine the nature of each structure considered. It must be determined whether we are dealing with true homology or convergence, whether a structure is plesiomorphic
O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
235
236
6 Evolutionary and Phylogenetic Effects in Meiobenthology
and archaic or apomorphic and derived. The interstitial fauna offer good examples of a complex structural mosaic of features that differ in their evolutionary significance. In cases where the whole taxon is restricted to the interstitial of sediments and has entirely evolved in this ecological refuge (“plesiotope”), we frequently find a combination of archaic and highly specialized and derived features. The fairly homonomous body segmentation and the biramous mandible of Mystacocarida are plesiomorphic features, while the terminal claws are clearly derived. Similar examples can be found among Cephalocarida, Gnathostomulida, Gastrotricha, Bathynellacea, etc. Within nematodes, the Epsilonematidae are characteristically shaped and sculptured; within phallodriline oligochaetes, some are gutless and in symbiosis with bacteria; within the turbellarians, the acoels have a reduced intestine and a digestive parenchyma instead. These specialized features, which the whole subgroup has in common, are well-founded synapomorphies of the respective taxon, particularly when supported by identity in structural or functional details. They are based on common ancestry and express the natural relationship of the subgroup. The situation is particularly straightforward in those cases where a few small, specialized members live in restricted interstitial niches while their “normal” relatives are of macrofaunal size and belong to large and diverse animal groups. Combined with their reduced size and vagility, these lineages often result in a simplified but derived body organization. These cases represent an apomorphic and phylogenetically fairly recent development: Halammohydra among cnidarians; Psammostyela, Heterostigma among ascidians; Monobryozoon among bryozoans; Labidoplax, Leptosynapta among holothurians; Sphaerosyllis among polychaetes; Microhedyle among molluscs. Ultrastructural investigations have often been helpful when examining animal relationships in meiobenthic taxa. Remane (1952b) summarized some general guidelines to ascertaining homology in his “criteria of homology.” These have been specified for ultrastructural features of meiobenthic animals by Rieger and Tyler (1979) and Ruppert (1982). Below are some examples of (ultra)structural analyses on meiobenthos that contribute to a better assessment of natural relationships between phyla as well as between subgroups of higher taxa, e.g., between the fairly wellstudied polychaetes. On the basis of number and arrangement of cilia, Rieger (1976) concluded that the cells of the more primitive phyla (at least of their larvae) such as Placozoa, Cnidaria, Gnathostomulida, and Gastrotricha (partim), are “monociliated,” i.e., flagellated, while the Spiralia (with their typically multiciliated arrangements) were considered more derived. On the basis of the cuticular ultrastructure, Rieger and Rieger (1976) further derived that the annelids were descended from platyhelminth ancestors. Ruppert (1982) considered a myoepithelial pharynx to be a symplesiomorphic character that is important enough to interlink gastrotrichs, nematodes, bryozoans and tardigrades. The ultrastructure of adhesive glands was the basis for phylogenetic considerations of those groups conventionally regarded as acoelomates, such as plathyhelminths and nemerteans, and their links to typical coelomates (Tyler 1977; Rieger and Tyler 1979; Rieger 1985). Ultrastructural characters, applied in a strictly cladistic analysis, refuted the classical platyhelminth
6.1 Body Structures of Evolutionary Relevance
237
grouping and suggested that the “Turbellaria” is a polyphyletic taxon of several convergent lines which are better combined under the term “free-living platyhelminths” Ehlers (1985). In annelids, ultrastructural details of the nephridia and the pharyngeal bulb have revealed that these structures, traditionally used as criteria for natural relationships, do not represent a valid basis for polychaete classification. According to Bartolomaeus and Ax (1992), metanephridia independently arose several times from protonephridia. Protonephridia, in turn, are homologous structures that connect various primitive animal groups. Similarly, the analogous nature of the pharyngeal bulb that emerged from ultrastructural studies indicated that archiannelids represent an artificial group of miniaturized and often neotenic worms belonging to various annelid families (Westheide 1985, 1987a; Purschke 1988). As early as 1974, Fauchald postulated that neither the archiannelids nor the typical errant polychaetes are close to the hypothetical primitive annelid. Instead, he envisaged the probable annelid ancestor to be a worm of meiobenthic size with simple setae and head appendages. These examples underline that studies of the structural details of meiobenthic groups provide a fruitful field for drawing phylogenetic conclusions. When considering the phylogenetic relationships between taxa, the role of larval structures has always been important. In meiobenthic groups, where retention of larval structures, i.e. progenesis, is characteristic, these are of particular relevance. Nielsen (1994, 2001) states that larval ciliation (downstream or upstream ciliary bands) is of fundamental relevance for the separation of metazoans into protostomes and deuterostomes. Larval ultrastructure has documented the relationship of some “archiannelids” to various polychaete families. The study of nauplius larvae clarified the position of syncarid crustaceans as primitive Malacostraca. According to Westheide (1987a), the loss of planktonic stages and adaptation to a permanent presence in sediment voids led to the evolution of an interstitial meiobenthos (Fig. 6.1). Worsaae and Kristensen (2005), taking nerillid polychaetes as an example, contend that this step towards a permanently small size was not restricted to intersitital habitats. They point out that many meiofauna live in the flocculent surface layers covering the bottom, where they have a richer food and oxygen supply than in the sediment. If this fluffy interface is evolutionarily preferred, a regressive trend towards small size would be advantageous because of reduced exposure to predation. So, even in muds, the progenetic scenario with an intermittent planktonic larval lifestyle suggested by Westheide (1987a) for sands would be advantageous. The frequent occurrence of larval trends in meiobenthos would support the contention that an archaic benthic fauna had a dispersive, larval phase. Secondarily then, this phase became abbreviated and eventually omitted (Hadzi 1956; Rieber 1994; see Sect. 4.1.4). Focusing on the basic metazoan groups, Haszprunar et al. (1995) contend that the planktotrophic larva is probably a derived feature. Direct development in the epibenthic realm would then be the plesiomorphic condition. The most powerful tool for elucidating kinships between groups, the study of genetic similarities by molecular biological methods, has challenged many of the established morphological classifications. Initially this new method had severe
238
6 Evolutionary and Phylogenetic Effects in Meiobenthology
eggs pelagic adult stage larvae Macrofauna pelagic juvenile stage
benthic adult stage progenetic development Meiofauna eggs
adult
Fig. 6.1 Hypothetical progenetic origin of the interstitial meiobenthos, which evolved from the macrobenthos with a meroplanktonic larval phase. (After Westheide 1987a)
flaws, since the first gene sequences were often based on too few and not significant genes. The resulting computations created often improbable neighborhood relationships that strongly contradicted current morphological understanding. Today, large data matrices consisting of numerous base pairs, combinations of 18 S and 28 S r RNA genes, and preferably considerations of Hox gene sequences (De Rosa et al. 1999) and biochemical characteristics (cytochromes, haemoglobins) give promising results which are continuously refined as the number of sequenced genes increases and additional species are included. New PCR techniques even allow DNA to be extracted from minute meiofauna (Schizas et al. 1997). In the future, automated laboratory power will allow more genomic and metagenomic analyses of meiofauna (see Parkinson et al. 2004; Woyke et al. 2006) and provide evidence of new relationships, especially those at critical branching points. However, considering the difficulties involved in replicating DNA material from nematodes. A great deal of this molecular work should be concentrated on the central meiobenthic group, the nematodes, particularly when considering the difficulties involved in replicating their DNA material. The comparison or even combination of molecular outputs with traditional computer-assisted cladistic approaches based on morphological characters will yield reliable and repeatable, but often revolutionary
6.2 Meiofauna in the Fossil Record
239
new affinities between taxa. With a more comprehensive data pool and a greater “openness” of researchers, both approaches will achieve increasing concordance, as already indicated by some examples described in the taxonomic part of this work (Chap. 5).
6.2
Meiofauna in the Fossil Record
The new molecular approaches enable us to extrapolate genealogies and relationships of selected morphological characters and gene sequences, but they do not reveal the morphology of the stem lines, and often fail to resolve the deep branching nodes which are of Precambrian origin (Runnegar 1991; Wray et al. 1996). Considerable inaccuracy also remains in the chronological assignment of radiative events, since calculations of timescales for evolutionary steps based on molecular clocks remain debatable due to the lack of fossils available for calibration (Kerr 1998; Rokas et al. 2005). Convincing evidence can be drawn from fossils linking present-day animals to prehistoric stem groups. However, discoveries of meiofaunal fossils are rare and fortuitous because of the small size and delicate nature of the animals. Hence, the Duoshantuo fossils of meiobenthic size that date some 50 million years before the early Cambrian (Chen et al. 2004) are biological highlights and confirm the archaic age of multicellular meiobenthos (see also Martin et al. 2000, with even earlier metazoan fossils from the White Sea). Despite their minute size, the remains of hard-shelled meiobenthos that date from the early Cambrian could even be assigned to existing taxa (e.g., Ostracoda). Later, in the early Cambrian, the remains of hard-shelled meiobenthos could be assigned to existing taxa such as ostracods, despite their minute sizes. However, information on other microfossils of meiobenthic size remains fragmentary (the first nematodes and kinorhynchs from the Cambrian; an early anostracarelated microfossil, Lepidocaris, from the Devonian; most primitive syncarid crustaceans from the Carboniferous, etc.). While many meiobenthic fossils remain to be discovered in the well-known paleozoic lagerstätten (e.g., Burgess shale), the discovery of tiny phosphatized arthropods in the rich “Orsten fauna” marks the beginning of a new era of studies on the evolution of meiobenthos (Müller and Walossek 1985, 1991; Walossek and Müller 1990). Despite their old age (Upper Cambrian), their extremely good preservation together with a meticulous fine-preparation revealed unprecedented details such as teeth structures, limb ramifications and chaetation patterns. While some of these fossils are doubtlessly related to extant higher taxa, for instance Rehbachiella to the branchiopods (Walossek 1996), there are other findings which cannot be clearly allocated to any known animal group, not even to a higher taxon (Fig. 6.2). Older findings (Middle Cambrian) of Orsten-type fossils from Siberia open up further exciting discoveries of extinct meiofauna, one of which was the earliest tardigrade (Müller et al. 1995). New lagerstätten from the early Cambrian, especially in China (see overview in Waloszek et al. 2005), and new, non-destructive
240
6 Evolutionary and Phylogenetic Effects in Meiobenthology
Skara anulata
Fig. 6.2 A Cambrian meiobenthic crustacean from the Orsten formation of unknown taxonomic allocation (total length 1.2 mm). (Müller and Walossek 1985)
paleontological techniques provide us with new insights into the origin of the “Cambrian explosion” and will fill many gaps in the sparse records documented so far (Meysman et al. 2006a). From these early fossil findings it is evident that all phyla that exist today were already present at the Precambrian/Cambrian boundary, and that many bilaterian animals of meiobenthic size arose well before the “Cambrian explosion” (Chen et al. 2004). Was there enough dissolved oxygen for respiration in these early periods of animal life? Conway Morris (1998) confirmed earlier hypotheses (Jenkins 1991; Runnegar 1991) that even in the early Ediacaran, a primitive, probably small and simply structured benthos existed under the low-oxic conditions that prevailed during these periods. Low oxygen pressure (less than 5 % of the current level) and the concomitant minimal oxygen diffusion distances enabled an oxic metabolism only for meiobenthos-sized metazoans (Thomas 1997; Canfield 1998; see Sect. 8.4). On the other hand, as fauna with biomineralized skeletons evolved and became bioturbative at the shift to the Cambrian, they disrupted, for the first time, the prevailing two-dimensional bacterial mats that covered the bottoms of the Proterozoic oceans. By burrowing and digging into the substrates, they caused a flux of surface water, so that deeper horizons were supplied with oxygen and food (Dornbos et al. 2005; Meysman et al. 2006a). This “Cambrian substrate revolution” (Bottjer et al. 2000) is believed to be one of the most important factors that instigated the rapid radiative evolution of metazoa in the early Cambrian. The extent to which the new paleontological findings shift the origin of early metazoans further backward in the proterozoic periods, thus supporting the “thiobios hypothesis,” remains unclear (Boaden and Platt 1971; Boaden 1975, 1977). This hypothesis contends that the early metazoans belonged to a “thiozoic” meiofauna that, without a planktonic phase, was frequently exposed to hydrogen sulfide without regular access to oxygen. It was contended that some of the extant primitive meiobenthic taxa are directly derived from these thiozoic roots (Haszprunar et al. 1995). Even today, these taxa live in anoxic and sulfidic habitats corresponding
6.2 Meiofauna in the Fossil Record
241
to the living conditions of the primeval (benthic) biotopes. However, the thiobios hypothesis has been placed in serious doubt by the fact that many of the recent thiobiotic animals represent highly derived specialists and are not ancient survivors (for details see Sect. 8.4). More detailed reading: Ax (1963); Rieger (1976); Rieger and Tyler (1979); Conway Morris (1998); Conway Morris et al. (1985); Bryant (1991); Waloszek et al. (2005); Meysman et al. (2006a)
Box 6.1 The Origin of Meiobenthos: Hypotheses and Fossil Documents Many meiobenthic groups have archaic roots, as documented either by their primitive body organizations or by early fossils. Maintenance of serial homonomy in some meiobenthic crustaceans (Cephalocarida) or molluscs (Micropilina) is interpreted as evolutionary basal. On the other hand, the small sizes of meiofauna and specific requirements of the meiofaunal biotope have often secondarily led to the simplification of organs, with convergent trends concealing phylogenetic relationships. Thorough studies of homology, often achieved by analyzing ultrastructural details, are needed to reveal genuine relationships. In this field, larval structures have always played a prominent role, especially since the maintenance of a larval organization in adults (progenesis, neoteny) is typical of many meiofauna. The discussion on the origin of the meiobenthos often centers on the presence of larvae as planktonic, dispersive stages. Did some early macrobenthic animals give up their planktonic larvae and become permanently (endo)benthic, retaining larval size and (sometimes) structure? Miniaturization would have provided access to the protected and nutrient-rich surface layers. While this issue will remain debatable, the rapid advances in molecular methods also for meiobenthic research have led to new insights into obscure phylogenetic alliances between taxa that separated in archaic times. They indicate, for instance, a position of the meiobenthic Acoela basal to all bilaterian metazoans. Other most convincing and informative documents remain the microfossils that are reconstructed in wonderful completeness by sophisticated methods. The new findings from the lower Cambrian or even Precambrian lagerstätten, which often have unclear relationships to present-day phyla, challenge the idea of a Cambrian “explosion” of new phyla. However, they also indicate that many metazoan taxa were originally meiofauna adapted through their small size to living under hypoxic or perhaps temporarily anoxic conditions. This sheds new light on the “thiobios theory.” As a result of mutual fauna–habitat interactions, the bioturbative activities of the early Cambrian fauna are believed to have instigated the “Cambrian substrate revolution” which, in turn, allowed the rapid evolution of new animal groups in a physicochemically novel benthic environment.
Chapter 7
Patterns of Meiofauna Distribution
In a world of global environmental deterioration with an ever-increasing number of endangered species, the study of species distribution and dispersal has grown in importance and scientific attention among meiobenthologists. The potential to recolonize impoverished habitats and the risk of invasions by “neozoa” are fields of increasing awareness. This supports assessments, also on the level of meiofauna, of species distribution and richness, particularly in less studied biotopes, for instance in polar, tropical and deep-sea regions. Considerations relevant to this issue involve the distributional pathways that underlie the present patterns of distribution and the genetic diversification of populations. These general aspects will serve as a basic orientation before the meiofauna from selected biotopes is characterized (Chap. 8). Meiofauna occur in all aquatic biotopes and climatic zones, from polar ice to alpine lakes, from hadal troughs to mangrove swamps. Of course, distribution patterns due to historical and/or biological characteristics underlie this ubiquitous occurrence. One of these patterns is the vicariant amphi-oceanic distribution of coastal species, well explored from the North American and European Atlantic coastlines (Ax and Armonies 1987; Stock 1994). The mechanisms of dispersal underlying these distributional patterns are not conclusive and are still remain debated. Potential pathways will be addressed, as will the opposite effects—mechanisms of restriction to isolated localities with a high rate of endemism. Especially for meiofauna, restricted distribution patterns are problematic for two reasons: (1) these patterns are often not based on comprehensive studies and only reflect local findings, while investigations from many potentially connecting areas are incomplete, and (2) the taxonomic affiliation often does not stand up to detailed examination with modern methods by experienced specialists.
7.1
Evolutionary Aspects
It is widely accepted that the marine littoral represents one of the earliest and most favored zones for animal life due to its rich supply of nutrients and food (bacteria, microalgae). On the other hand, the astatic, often rigid conditions typical of this zone O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
243
244
7 Patterns of Meiofauna Distribution
require evolutionary versatility and force the fauna to continuously respond and adapt to physical and biological constraints. Remane (1952a) termed the littoral mesopsammal “a zone of radiation into adjacent habitats,” because in this evolutionary scenario those species that were not able to cope with the demands of the marine littoral as a competitive and radiative center became displaced. According to this conception, the small, less protected benthic animals in particular moved from the surface into the endobenthal and, if they were small enough, into interstitial habitats where they experienced a lower number of competitors and a more balanced physiography. In this subsurface refuge area of enhanced stability, evolution allowed many meiobenthic forms to maintain their original biology and sometimes also their morphology. It is this refuge character of many meiobenthic habitats which makes the study of meiofaunal distribution so valuable for zoogeographic approaches. Since it was protected, this habitat not only enabled the survival of archaic forms, but it also allowed for reduced reproductive rates, limited distributional means via propagative stages, and adaptative specializations; it favored the development of K strategists. These biological prerequisites, which are often also consequences of miniaturization, are thought to promote allopatric speciation and to make meiofauna good zoogeographical indicators for possible evolutionary connections. This is particularly evident in the freshwater stygofauna. In the ecologically stable, often isolated groundwater habitats, a fauna with relict characters survived or long-lasting adaptive lines evolved that were often of a regressive nature. In order to avoid the rigid conditions of the marine epibenthic littoral or to survive periods of ocean regressions, one important pathway was immigration into the “anchihaline” continental shores and further into the subterranean freshwater system, the stygobial (see Sect. 8.7.2). Delamare-Deboutteville (1960) and Stock (1994) underlined the role of the brackish coastal groundwater as a predominant route of immigration into the continental groundwater system. This brackish coastal colonization pathway would explain why many of the groundwater Amphipoda (Bogidiella; Ingolfiella), Thermosbaenacea (Tethysbaena) and Isopoda (Microcerberus, Microcharon), which now live far from the coast, have close marine relatives. This relationship is particularly striking in isolated groundwater lineages of marine groups, e.g., the polychaetes Troglochaetus beranecki and Hesionides riegerorum (Westheide 1979), nematodes of the genus Desmoscolex, some macrodasyoid gastrotrichs (Kisielewski 1987), and some turbellarians (see Sect. 8.7.2 for a detailed account). A recent example of this colonizing pathway is the distribution of the water mite Halacarellus subterraneus. Without any morphological variations, it lives in habitats ranging from the marine littoral deep to continental groundwater aquifers. Another pathway of refuge apparently used by many meiobenthos led away from the shallow seashore into deeper and protected regions of the sea. Both the deep-sea bottom and marine caves are stable and conservative environments and often considered to represent refuge biotopes (Iliffe et al. 1984; Sket 1996). Indeed, there are striking taxonomic relations between the meiofauna of anchihaline caves and the deep-sea. Good examples of this linkage include the tardigrades and ostracods (Villora-Moreno 1996b; Boesgaard and Kristensen 2001, Danielopol 1990b).
7.1 Evolutionary Aspects
245
These interrelations between troglobitic and abyssal fauna have been explained by a contiguous pathway from the depths passing through the sheltered and deepreaching interstitial and crevice system which purportedly connects deep-sea bottoms with shallow marine caves (Wilkens et al. 1986;Iliffe 1990, 2005). From there, immigration routes would have led further to the freshwater stygobios (Ward and Palmer 1994). This would imply that many deep-sea fauna were plesiomorphic, a feature that cannot be demonstrated and that neglects the long-lasting anoxic periods in the oceans (Stock 1986). Moreover, good evidence for crevice connections between these disjunct habitats has never been provided. The more probable option seems a gradual immigration of a formerly widespread littoral meiobenthos into the refuges of anchihaline habitats, caves or bottoms of the deep-sea (after the last Cretaceous/Tertiary oxygen crisis). After repeated phases of marine transgression and regression, some meiofauna succeeded in colonizing the more stable continental stygobial. The amphi-Tethyan distribution of the amphipod Pseudoniphargus speaks in favor of this contention (Stock 1986, 1994; Danielopol 1990b). The ancestral forms characteristically have had wide and coherent distribution areas, while their descendents “stranded” on geographically or ecologically isolated refugia, which gave rise to an independent genetic/phenetic evolution resulting today in endemic forms in brackish and (later) subterranean freshwater (Stock 1994). Mayr (1982) speaks of “vicariant events” that caused a “peripatric speciation,” as opposed to allo- and sympatric speciation. Hence, the vicariance between deep-sea and anchihaline taxa probably originates from a common shallow-water ancestral pool and does not indicate elusive migratory connections. One example is the recent finding of the harpacticoid Kliopsyllus, a genus that is typical of interstitial beaches, in deep-sea muds (Veit-Köhler 2004). Taxonomic connections between the deep-sea and shallow water have also been documented for several other meiobenthic groups. In a shallow, previously ice-covered Arctic cove, Gooday et al. (1994) found a “deep-sea assemblage” of Foraminifera and macrofauna. Can we document with meiobenthic examples the differentiating steps that are required for a species to diverge and form two new taxa with meiobenthic examples? On the route towards the separation of an originally uniform gene pool, slight differences in physiological reactions, behavioral inventories and ecological preferences can cause cryptic speciation despite an unaltered morphology. A close morphological similarity does not necessarily mean an identical physiology and ecology, or the same genetic background. A population exposed to an unusual physiographic milieu may initially cope with this altered environment through its wide adaptive capacity (non-genetic adaptation). However, its genetic range may gradually shift in order to provide an energetically less costly solution. Eventually this will result in a metapopulation with an adaptive variant that better responds to the new environmental demands (Battaglia and Beardmore 1978). Heterozygosity offers adaptive advantages, but bottleneck effects often also cause chances for selection, particularly in stress areas of environmental gradients (e.g., salinity, pollution). Before any differentiation becomes apparent in structural details, it may be expressed and genetically fixed in divergences such as community and life history parameters, physiological reactions and ecological preferences. When experimen-
246
7 Patterns of Meiofauna Distribution
tally tested, this may result in differences of growth curves, reproductive effects, and metabolic rates. Eventually, even in an identical morphospecies, the tolerance ranges of disjunct populations may display clear differences (e.g., Ophryotrocha; Levinton 1983; Levinton and Monahan 1983; Aktedrilus; Giere 1980). Consequently, flocks of closely related species should be numerous among meiofauna, and the percentage of morphologically similar or even identical, yet genetically divergent species should be high. This has been documented by Westheide (1991) for polychaetes and ostracods from the Galápagos archipelago. The flock of cryptic species in the harpacticoid genus Tisbe or Tigriopus, the species complex of Canuella perplexa or Paraleptastacus macronyx (Harpacticoida), Paracanthonchus caecus (Nematoda) and Pseudomonocelis spp. (Turbellaria, Proseriata), and Microphthalmus listensis (Polychaeta) all differ in their ecological ranges but are very similar in terms of structural details (Battaglia et al 1978; Gabrich et al. 1991; Casu and Curini-Galletti 2006). The well-studied model nematode Caenorhabditis elegans differs despite close morphological congruence with its sister species, C. briggsae, for almost 10% of its genes (Blumenthal and Davis 2004)! The remarkable changes in allele composition (cytochrome oxidase C1 gene) among closely adjacent populations of the cosmopolitan nematode Pellioditis marina indicate intensive metapopulation dynamics (Derycke et al. 2008). Subtle differences in substrate preference and biometry also characterize the forms of Cobanocythere labiata (Ostracoda) (Westheide 1991). Another powerful mechanism favoring segregation and ultimately speciation is the trophic niche-partitioning of many meiofauna, e.g., the narrow selection of bacterial strains by nematodes (Moens et al. 1999b), or the attractiveness of discrete patches of diatoms to harpacticoids (Santos et al. 1995). This “internalized” basis for sympatric mode of speciation contrasts with the classical allopatric speciation based mainly on geographical isolation (see below). One characteristic feature of meiofauna that supports sympatric speciation is the need for energy conservation. This leads to food specialization, organ simplification/reduction, efficient reproductive structures and modes (hermaphroditism, internal fertilization, brood protection), short generation and life times, often with abbreviated progenetic trends, e.g., in cladocerans, ostracods, syncarids. In phylogenetic terms, all of these typically meiobenthic strategies that favor specialized, often K-selective, animals have a high potential for sympatric speciation, which can open up new niches and evolutionary pathways without any geographic separation (e.g., in harpacticoids, see Ivester and Coull 1977; Marcotte 1984). Diverging reproductive strategies are another root of sympatric speciation in meiobenthic animals. Cryptic species may be brought about by sexual isolation through polyploidy. In the marine oligochaete Lumbricillus lineatus, a local 3 n population has developed which must be pseudofertilized by the normal 2 n stock (Christensen and O’Connor 1958) to produce offspring. Reproductive isolation in the uniform and ubiquitous harpacticoid Phyllognathopus viguieri has been documented in crossbreeding experiments (Glatzel and Königshoff 2005). Sibling species with different gonochoristic and hermaphroditic lines occur in the polychaete
7.1 Evolutionary Aspects
247
Ophryotrocha. This genus contains simultaneous or consecutive hermaphrodites, some of which are oviparous, while some are viviparous (Åkesson 1973). Hence, compared to macrofauna, trends towards allopatric and sympatric speciation are well recognized in meiofauna, which would lead in turn to a high diversity that manifests itself as an enormous and yet unexplored species richness (see Sect. 9.1). How do speciation processes in meiobenthos compare to the situation in micro-organisms? Their scenario follows the postulate “everything is everywhere” (Fenchel and Finlay 2004). Micro-organisms, with their huge populations, small body sizes and high rates of dispersal, tend to have a cosmopolitan distribution with a relatively low diversity. However, many meiobenthic species also have a wide distribution along continental shorelines, despite of the limited reproductive and dispersive potential that characterizes meiofauna. Many have even an amphi-oceanic or cosmopolitan occurrence. So, we have a “meiofauna paradox” which comes down to two questions. (1) Why are so many meiofaunal taxa from distant areas so similar despite their limited means of dispersal? (2) How can meiofauna have bridged oceans and occupied disjunct shores in the absence of large populations and competitive propagative stages (Fig. 79; Gerlach 1977b; Westheide 1987b; Danielopol et al. 1994)? In many examples of wide or even cosmopolitan distributions of meiofaunal species, differences that were not evident from morphological characters have been revealed. Originally this was accomplished by physiological and immunological methods, i.e., enzyme electrophoresis and antiserum tests. The latter had already been successfully applied for predator–prey relationships (Feller et al. 1979, Feller 1982). In some rare cases natural relationships were even be demonstrated by successful positive or negative infestations with parasites (the “rule of Fahrenholz”). Åkesson (1977) succeeded in transmitting host-specific sporozoan parasites from Ophryotrocha spp. to Dinophilidae, implying a close relationship between the groups. Nowadays, molecular methods are mostly applied in order to scrutinize geographically disjunct yet morphologically barely differentiated meiofaunal species, and the number of pertinent studies is rapidly increasing. However, the picture resulting from gene sequencing and molecular comparisons (facilitated by efficient computer programs) is inhomogeneous. The global distributions of many former “species” will have to be differentiated and split into various restricted areas inhabited by genetically diverging species: Monocelis lineata (Plathelminthes), Xenotrichula intermedia (Gastrotricha), nerillid or syllid species (Polychaeta), epsilonematid nematodes, and various harpacticoids (Microarthridion littorale, Nannopus palustris, Cletocamptus deitersi (Todaro et al. 1996; Schmidt and Westheide 1997/98; Soosten et al. 1998; Gourbault and Decraemer 1996; Schizas et al. 1999; Rocha-Olivares et al. 2001; Staton et al. 2005). In Hesionides arenaria, a polychaete often considered an example of a cosmopolitan distribution (Fig. 7.1a), further molecular studies will probably reveal genetic differentiation among disjunct populations. The various so-called “panabyssal” ostracods with seemingly wide distributional ranges are probably also genetically distinct species, each with a restricted occurrence (Brandao, pers. commun.). Other examples of cryptic clades are provided by nematodes, one of the
248
7 Patterns of Meiofauna Distribution
a
b Fig. 7.1a–b Cosmopolitan distribution of meiofauna species. a An example of a morphospecies (Hesionides arenaria, Polychaeta) whose genetic uniformity remains to be elucidated. b Gyratrix hermaphroditus (Plathelminthes). (a Westheide 1977b; b Sterrer 1971)
quintessential animal taxa. At the species level, some ubiquitous forms (e.g., Pellioditis marina) have remarkably different genetic backgrounds. Terschellingia longicaudata, a common shallow-water nematode of ubiquitous occurrence, demonstrated its genetic diversity only after thorough analysis of several nuclear and mitochondrial molecular sequences (Bhadury et al. 2008). Many genera and families encountered in the deep-sea are believed to correspond to shallow-water taxa. Is there really a gene flow between these separated taxa? This will be revealed by molecular analyses. For the turbellarian Monocelis lineata, allozyme and genetic analyses revealed the existence of distinct Atlantic and Mediterranean clusters as well as discrete sibling species in fully marine and brackish habitats of the Mediterranean (Casu and Curini-Galletti 2006). On the other hand, in some cases
7.2 Zoogeographic Aspects
249
of cosmopolitan occurrence (the polychaete Ctenodrilus serratus), comprehensive molecular methods have reinforced the notion of a common amphiatlantic gene pool, which implies ongoing transoceanic genetic exchange (Westheide et al. 2003; see Sect. 7.2). A high ongoing gene flow was also revealed between some species of Arctic and Antarctic rotaliid Foraminifera. Together with biochemical methods, genetic analyses, especially those that resolve multiple comprehensive sequences, have a considerable potential to solve questions about cryptic species and the geographical differentiation of species flocks. They will stimulate further morphological scrutiny, which remains an indispensable approach since most cases of “identity” are defined by morphological standards, based on the “archives of diversity”, the museum collections. Inspection by experienced taxonomists, including the application of novel microscopical methods (immunohistochemistry, environmental electron microscopy, confocal laser microscopy), remains important (see Chap. 10; Gourbault and De Craemer 1996). The combined efforts of traditional and molecular methods will probably reveal that there is a much lower uniformity at the species or genus level in meiofauna than assumed so far, and will thus partly resolve the meiofauna paradox of the wide-ranging distributions of many “species.” More detailed reading: Sterrer and Ax (1977); Botosaneanu and Holsinger (1991); Fenchel (1993); Danielopol et al. (1994b); Fenchel and Finlay (2004).
7.2
Zoogeographic Aspects
The traditional zoogeographic regions have been established on the basis of macrofaunal distributional patterns. It remains to be seen whether they will also be applicable to the distribution of meiofauna, with their different ways of propagating. The huge dispersive potential of microorganisms, which results in a random, often ubiquitous distribution pattern (see ciliates, e.g., Remanella rugosa; Fenchel 1993; Fenchel and Finlay 2004; Finlay et al. 1996; Danielopol et al. 1994), was mentioned above. A cosmopolitan distribution has developed in various meiobenthic groups too, despite their limited vagility (see Fig. 7.1), e.g., in turbellarians, gnathostomulids, gastrotrichs (Urodasys viviparus), interstitial polychaetes (e.g., Hesionides spp., Microphthalmus complex), mystacocarid crustaceans (Derocheilocaris spp.), harpacticoid crustaceans (Phyllognathopus viguieri) and tardigrades (Halechiniscus remanei). In the limnic meiobenthos and in microorganisms, the dispersive relevance of resistant eggs and cysts requires attention. Resting stages that facilitate long-distance dispersal by wind, migrating shore birds or water currents are also known to occur in some marine metazoans (ostracods, tardigrades, some harpacticoids and turbellarians). But what are the mechanisms that disperse the numerous meiofaunal taxa without these resting stages? Water drift, rafting on seaweed, flotsam, marine snow, human transport activities, or even geological processes such as continental drift?
250
7.2.1
7 Patterns of Meiofauna Distribution
Mechanisms of Dispersal
Although mechanisms preventing them from being washed out of the sediment are characteristically developed in meiobenthos (see Sect. 4.1), recruitment and colonization via the water column are common explanations of meiofaunal distribution patterns. The “hyperbenthic” occurrence of meiofauna that are regularly caught above the sediment (Palmer 1988; Armonies 1994; Walters and Bell 1994; Commito and Tita 2002; Ivanenko 1998) has also been experimentally confirmed (Hagerman and Rieger 1981). This emphasizes the need to consider water transport an important means of meiofauna dispersal, and to differentiate this into several natural, often connected processes that are relevant for distribution: passive, physically controlled erosion and active, behaviorally controlled emergence. The subsequent drift phase is further influenced by the process of rafting. Erosion. Passive erosion and subsequent transport above the sea bed are particularly relevant in hydrodynamically active areas exposed to tidal currents and breaking waves, which scour the sediment and may suspend the meiofauna of the upper sediment layers (Fig. 7.2a). Regular tidal currents have been found to alter small-scale
a
b
low ...
... medium ...
night
high current velocity
day
Fig. 7.2a–b Current-induced (a) and diurnal (b) emergence, erosion and re-entry of meiofauna. (After Armonies 1994)
7.2 Zoogeographic Aspects
251
distributional patterns via suspension and resettlement, especially in ostracods and harpacticoids; less so in nematodes (Bell and Sherman 1980; Decho and Fleeger 1988a; Fegley 1988; Armonies 1990). Within the harpacticoids, the more exposed epibenthic species were more easily suspended than the interstitial ones, and copepodites easier than adults (Hicks 1992). Trap experiments recorded that dispersal rates of harpacticoids were 65-fold higher than those of nematodes (Commito and Tita 2002), and within nematodes, epigrowth feeders had much higher dispersal rates than deposit feeders. Even weak near-bottom currents, which become enhanced as they pass over bottom ripples, can suspend and transport particles (Huettel and Rusch 2000). Other above-ground structures, such as boulders and plant culms as well as biogenic tubes or mounds, can also modify hydrodynamic currents and cause turbulence, aiding the suspension of meiofauna. Palmer and Gust (1985) compared suspended meiofaunal organisms with quartz grains of 40 µm diameter, inferring passive transport of the animals once suspended in the water column. Breakers produced during gales have been shown to erode the bottom down to a water depth of 25 m and to carry particles over distances of 50 km. Erosion by tidal currents and erratic hydrodynamic forces with subsequent passive drift along continental coastlines and with large ocean current systems may be a relevant means of dispersing meiofauna, even at continental shelf depths (Sedlacek and Thistle (2006). Meiofauna dispersal at velocities of 10 km per day is possible through erosion and passive drift (Hagerman and Rieger 1981), resulting in rapid local fluctuations. In lotic freshwater ecosystems (streams and creeks), erosion and passive watercolumn transport are important means of meiofauna dispersal (Palmer and Gust 1985; Palmer 1990a). Recruitment of meiobenthic populations from the water column and the streambed surface is even more relevant than immigration from adjacent areas, and it can explain the rapid recolonization within some weeks after severe disturbances (e.g., spring floods). Under these conditions, up to 2,500 meiofaunal organisms per 1 m3 (mainly rotifers, chironomid larvae, oligochaetes and copepods) have been found drifting in the waters of a stream (Palmer 1992). They preferentially settle in more quiescent sediment patches (Palmer et al. 1996). In streams, drift phases were terminated by the re-entry of meiofauna into deeper, hyporheic sediments of the streambed. This has been shown to be of significance for the recruitment of rotifers and copepods (Palmer et al. 1992). However, an active biological component even persists in highly dynamic creeks: the rates of drifting organisms increased significantly at night. Emergence. Analysis of suspended meiofauna in near-bottom waters revealed that some groups such as harpacticoids, ostracods, and turbellarians are much more often subject to aquatic drift than nematodes and other groups. This repeated observation, which has even been reported from sublittoral and deep-sea bottoms (Thistle et al. 2007b), does not support the contention that animal drift results from erosive forces only and that meiofaunal individuals are comparable to passive particles or sand grains. Instead it points to a biological component of some meiofaunal suspensions and dispersals. Active emergence would behaviorally influence the rate of dispersal (Armonies 1994; Buffan-Dubau and Castel 1996; Commito and
252
7 Patterns of Meiofauna Distribution
Tita 2002). Particularly for the phytal and epibenthic meiofauna, emergence and subsequent dispersal seem to be important factors. A periodic emergence of harpacticoid copepods into the “hyperbenthal” is pronounced in seagrass meadows, where more than 50% of all specimens ascend nightly into the water column, especially in calm weather (Fig. 7.2b; Arlt 1988; Hicks 1986, 1988a; Walters and Bell 1986). However, the rate of emergence varies between 1 and 80% of the whole population, apparently depending on body morphology and the microtopography of the bottom surface, such as ripples or depressions (Sedlacek and Thistle 2006). The biological role of this periodic suspension is underlined by the observation that some harpacticoid species even feed preferentially during their stay in the water column. Apparently it is mostly emergence rather than passive drift that disperses meiofauna (Armonies 1994). Emergence behavior may vary in different species: those turbellarians that feed on harpacticoids synchronized their emergence behavior with that of their prey, while those preferring the less emergent nematodes as prey stayed in the sediment (Armonies 1994). Epiphytic meiofauna in the Black Sea changed position during the night: while the nematodes descended from the plants to the bottom, the harpacticoids entered the water column (Kolesnikova et al. 1995). Among the harpacticoids of a sublittoral community, the emerging species could be discriminated from the non-emergers by some adaptive details of the swimming legs (Thistle and Sedlacek 2004). However, active emergence does not seem to be restricted to the phytal. Armonies (1988a,b) showed that in the sandy flats of the Island of Sylt (North Sea), particularly on calm nights, 87% of all harpacticoids, 67% of ostracods, and 42% of turbellarians left the sediment. Catching the suprabenthic biota with a “plankton sledge” (see Fig. 1 in Armonies 1994) should also reveal in other areas a considerable amount of the drifting meiofauna. The preferred nightly emergence of meiofauna (Fig. 7.2b) was explained by the avoidance of strong predation pressure in the benthal layer. Ascending into the water column at night would diminish predation by (mostly visually oriented) predators (Armonies 1989b, 1994). Even nematodes that are often considered sediment-bound showed emergence and active re-entry with selecting algae-enriched sediment patches (Ullberg and Ólafsson 2003). An active emergence behavior is also thought to occur in the tardigrade Florarctus salvant, which floats out and moves by lateral cuticular “wings,” and perhaps also in Batillipes bullacaudatus with its inflatable bubble tail. While calm seas seem to promote active emergence, strong currents keep meiofauna suspended by erosive forces rather than active ascent. Emergence sometimes seems to be associated with reproductive, feeding or fugitive periods. The degree of emergence may also vary depending on sex distribution, developmental stage (Bell et al. 1989), and—mainly in temperate regions—on seasonal cycles. Environmental deterioration also seems to elicit active mass evasions of meiofauna with subsequent drift phases (Armonies 1994). Re-entry. The counterpart to active emergence is the process of re-entry. The original contention of a merely passive and more or less random resettlement from the suspended stock (Eckman 1983) has been redefined. Rather it is a mixture of physically controlled resettlement and, in the direct vicinity of the bottom, fairly
7.2 Zoogeographic Aspects
253
selective re-entry (Butman 1989). Here physical details like structural complexity, roughness of the bottom and microturbulences (Palmer 1988; Eckman 1990; Ullberg and Ólafsson 2003a) are important modulating factors. Turbulences reduce the suspension time after erosion by increasing the “hitting time” of an eroded particle onto the substratum (McNair et al. 1997). These mathematically proven processes can give turbulence a new, biologically relevant aspect. However, experiments and mathematical considerations have shown that resettlement is certainly also an active biological process that depends on chemical and microbiological signals perceived at small distances and in calm water (Fegley 1988; Butman 1986, 1989; Fleeger et al. 1995). In experiments with harpacticoids, an active behavioral component was only recorded at low current velocities, while at higher flume flow the animals were preferably deposited in sediment depressions in the manner of passive particles. Active re-entry would be an explanation for the puzzling fact that meiofauna, despite tidal currents and suspension, can be found redistributed under favorable conditions in an almost identical pattern: micro-scale distributions with certain centers of patchiness remain established over several tidal cycles (Fleeger et al. 1990; but see Burkovsky et al. 1994).
Mechanism:
Active
Passive
Meiofauna in water
Suspension by disturbance
Emergence
Reentry
Erosion
Resettlement
Meiofauna in the sediment
Flow :
Low
Aboveground structure : Present
High Absent
Taxa :
Copepods dominate
Variable
Disturbance :
Biological
Biological, physical
Habitats :
Seagras beds Vegetated marsh areas
Unvegetated mudflats Sandflats Beaches
Fig. 7.3 Modes of meiofauna distribution compiled. (After Palmer 1988)
254
7 Patterns of Meiofauna Distribution
Although far from understood, the role of water-bound dispersal, whether by passive erosion or active emergence processes, must be included in any consideration of meiofaunal abundance, population dynamics, colonization behavior and zoogeographical distribution (Armonies 1994; Commito and Tita 2002). These seem to depend on a complex of (1) anatomical and behavioral features of the animals, (2) light and current conditions, (3) population density, and (4) the hydrodynamic patterns of the sediment surface (Fig. 7.3; Service and Bell 1987; Walters 1991; Armonies 1994). Dispersal of meiofauna and re-entry into the sediment can be fairly quick processes, often requiring just a few hours or days. Thorough experimental work and time-lapse observations are needed to account for the effects of currents and active responses on the drift and dispersal of meiobenthic animals. Microcosms without the natural microturbulences of open water bodies are probably not very helpful in this respect. Along continental coastlines, on tidal shores or in streams and rivers, suspension and subsequent re-entry/settlement of meiofauna are probably important distributional factors. However, a free (not raftborn) drift is hard to envisage for dispersal over longer distances, especially across over deep-water regions and across oceans. Rafting. With our increasing awareness of transoceanic dispersal (Carlton and Geller 1993; problem of invasive neozoa in established faunas), and because of our attempts to maintain species diversity and fight extinction, the roles of natural and anthropogenic mechanisms of meiofaunal transport have become important topics in meiobenthology. Bottom bacteria, microalgae (diatoms), and their mucous excretions can form dense mats that cover the sediment surface. These can become suspended, forming rafts that carry epibenthic meiofauna (especially harpacticoids) at least over small distances, together with some bottom particles (Hicks 1988b). Rich assemblies of diatoms, copepods, ciliates and nematodes associated with “floating natural detritus” have been found in shallow coastal waters (Faust and Gulledge 1996). Various kinds of macroalgae, such as seaweeds, thickets of filamentous algae, or Sargassum floating in the water, have also been shown of transport a rich meiofauna (Arroyo et al. 2006b) consisting mainly of epibenthic and phytal species. Along exposed shores, clumps of detached kelp and other macroalgae drift regularly over long distances, and these contain a diverse meiofauna (and even some macrofauna) (Ingolfsson 1995; Ólafsson et al. 2001). Thus, algal rafting can be considered an important vehicle for dispersal, even over hundreds of kilometers. Small-scale experiments with phytal nematodes confirmed their considerable colonization potential, depending on the proximity of donor communities and the degree of “open niches” in the receptor assemblage (Derycke et al. 2007b). Armonies (1989c) found numerous meiobenthic organisms in the “sea foam” excreted by algal cells after blooming periods (Phaeocystis). In the warmer seasons these foam mats drift regularly on the surface of the sea and accumulate on beaches, where they form relatively long-lasting rafts that may assist meiofauna dispersal over wide areas. Large “floating islands” with a rich vegetation are known to be stable for years, and to drift along the coast. These rafts, consisting of parts of the shore line torn loose after severe storms or floods, have even been discussed as vehicles for mam-
7.2 Zoogeographic Aspects
255
mals to cross the southern Atlantic (e.g., Houle 1999); they would certainly be a habitat for a rich spectrum of meiofauna. However, repeated transport that maintains a regular gene flow is required to explain genetic identity at lower taxonomic levels. In polar areas, suspension is often followed by colonization of the interstitial channel systems in ice floes (see Sect. 8.1.1). Here, drift ice may represent an effective rafting vehicle for even long-distance transport. Other natural rafts that could plausibly allow transoceanic journeys are rare. The thick fibrous “fur” of coconuts, which are adapted for floating over long distances in the oceans, has been found to harbor numerous meiobenthic organisms (Gerlach 1977b). Whether this vehicle is of more general relevance for meiofauna dispersal remains unclear. The possibility of man-made vehicles for transoceanic transport of meiofauna should not be ignored, at least for some routes and some predisposed meiofauna forms. Historically, the wet ballast sand of sailing vessels has been considered a possible source of meiofauna transport (Gerlach 1977b), and perhaps an important one between the more frequented ports. However, probably of more global importance for long-distance meiofaunal dispersal is the sand that accumulates in ships when pumping ballast water (Kelly 1993). A diverse and highly viable meiofauna with ovigerous females (a total of 314 individuals per dm3 sediment) was recorded in sediment from the ballast tanks of a bulk carrier brought to a Polish shipyard (Radziejewska et al. 2006). Even more considerable may be the meiofauna in the thickets of macrofaunal colonies that foul the hulls of ships. This is a known transport vehicle for polychaete larvae and is certainly also a suitable substratum for meiofauna (Carlton and Geller 1993). Barnes (2002) accentuates the role of the ever-increasing amount of plastic debris drifting across the ocean as an effective long-distance vehicle for invasive distributions of macrofauna. In particular, macrofaunal colonies (e.g., barnacles, mussels) almost certainly harbor a rich meiofauna that is able to survive oceanic passages. Whether the slow, random drifting of rubbish on the one hand or the short travel and harbor times of modern sea traffic on the other alleviate or aggravate the survival of meiofauna is an important study issue that needs to be addressed. Will the global increase in the number of ships and amount of man-made debris become a major means of dispersal for marine organisms and contribute to the exchange of fauna (Carlton and Hodder 1995)? Perhaps the most significant and global natural transport mechanism for microand meiofauna is a phenomenon termed “aquatic snow,” which occurs in both the marine and limnetic realms. Micro- and macroaggregates in the oceans formed by combined physicochemical and microbial processes and mucus secretions (see Sect. 2.2.3) are called “marine snow” (note that such aggregates have later also been observed in lakes). These aggregates are so universal and potentially important that they are of relevance for the worldwide production of organic matter (Riebesell 1992; Kepkay 1994; Heissenberger et al. 1996; Azam 1998). However, these aggregates of detritus, bacteria and mucus have considerable significance not only for the organic production and transport of nutrients, but also for the drift, dispersal, and distribution of micro- and meiofauna (Shanks and Edmonson 1990). A diverse community of temporary and permanent meiofauna has been found to be associated with these tiny but visible aggregates (sometimes > 500 µm), which have been sampled
256
7 Patterns of Meiofauna Distribution
in traps and filters but can also be created experimentally. The meiofauna clinging to these particles not only take advantage of a nutritious substrate (many polysaccharides and diverse bacteria), but their movements also appear to support the coagulation, compaction and growth of the aggregations (Walters and Shanks 1996; Shanks and Walters 1997). The meiofauna of marine snow consists mainly of nematodes and foraminiferans as well as harpacticoids. Equipped with various adaptations for adhesion and clinging, suspended meiofauna are well adapted to drifting in and feeding on this substrate, thus allowing them to survive even longer-term excursions in the open water. In collections from oceanic sediment traps high above the sea bottom, 80% of all trapped nematodes, polychaetes and larvae plus 20% of the harpacticoids were observed to cling firmly to marine snow particles (Shanks and Edmonson 1990). Based on microbiological studies, Simon et al. (2002) point to the need to include aggregate-associated processes in ecosystem analyses. In the context of meiobenthology, it is proposed, though it has been insufficiently studied, that micro-aggregates are also of the great significance for production, nutrition and dispersal of meiobenthos and for bentho-pelagic coupling processes. The permanence and natural dimensions of marine snow could make this phenomenon of great evolutionary relevance, even beyond those of classical geotectonic processes (see the following section).
7.2.2
Geological Structures and Processes
Similar or identical meiofaunal taxa, both marine and limnetic, are often reported from distant and isolated habitats. Vicariant occurrence of the same taxon on both sides of the Atlantic has been postulated, although we often lack knowledge of the intermediate deep-sea fauna (Sterrer 1973; Stock 1994). This similarity between many species or genera of meiofauna separated by vast global distances could imply a relatively recent genetic exchange (Fig. 7.1). This would exclude the drifting of continental plates as a mechanism of dispersal unless the speciation rates were extremely slow or zero after the continents had drifted apart. In the case of uniform amphi-Atlantic gnathostomulid species (Sterrer 1973), this would be through 120–150 million years. Conservative biological characters (low number of offspring and restricted means of dispersal) combined with constant habitat and climatic conditions, especially in anchihaline or troglobitic biotopes (see Sect. 8.7.2), serve as a basis for the “relict refuge model” that is used to explain the similarities between many continental groundwater fauna. The frequent relict characters of some meiofauna require as an explanation minimal speciation rates or evolutionary stasis over long geological periods. However, this extremely slow speed of speciation contradicts the paleontological experience that few recent species branched off before the Eocene period, i.e., date back more than 55 million years. Calculations based on molecular clocks of genetic change also point in the same direction, confirming a shorter existence of many species. This scenario of a rapid radiation is also postulated by paleontologists. At the
7.2 Zoogeographic Aspects
257
species level, this would refute hypotheses that plate tectonics and a Gondwandabased common origin are responsible for similarities between Europe and America (see above; Hartmann 1986). Recent findings on the rich meiofauna of the Galápagos Islands also contradict the idea of a particularly slow speciation in the meiobenthos, because in just 3 million years (the age of the islands) the pristine beaches of the Galápagos archipelago have been colonized by a rich and radiating meiofauna (Westheide 1991). In turn, genetic analyses of many seemingly identical, disjunct species have yielded molecular differences that are large enough to dispense with the dogma of evolutionary stasis. However, aside from plate tectonics, other geological/geographical pathways should also be considered. Brackish water species may have taken the circumpolar route along the Iceland–Greenland–Northern Canada path (Ax and Armonies 1990). A more detailed knowledge of shallow shelf connections, extinct landbridges (e.g., the legendary “Thule Landbridge” along the Scotland–Iceland Ridge) or island belts used as stepping stones during periods with low sea levels would reveal potential distributional pathways that have not been adequately considered previously. In the scenario of extinct geographical patterns, the impacts of the ice ages and the concomitant massive lowering of the sea level must be included. In freshwater meiobenthos, a clear link between the distribution of harpacticoids and glaciation has been shown, such that the unglaciated South European and North American areas without such habitat disruptions maintain a higher species richness (Strayer et al. 1995; Rundle et al. 2000). In recent studies, the numerous sea mounts (about 100,000 worldwide!), with their summits of highly complex, biogenic sediments, have attracted special meiobenthological studies (George and Schminke 2002). Do sea mounts support similarly high levels of meiobenthic biodiversity as known for macrofauna and fishes? Are sea mounts “trapping stones” for the dispersal of meiofauna, with a long history of isolated speciation and disjunct fauna that are separate from those of the surrounding deep-sea mud? Or are they instead in distributive contact with the deep-sea and serve as “stepping stones” for dispersal across the oceans? Relationships to meiofauna from neighboring continental coasts would resolve the potential stepping stone function. The occurrence of taxa that were identical at the genus or even species level (e.g., the harpacticoid Laophonte bicornis) on various sea mounts (stepping stones). Recent harpacticoid studies from some Atlantic sea mounts have indicated an absence of a typical sea mount meiofauna and a low degree of endemism. On the other hand, the fact that the sea mount meiofauna is very different from that of the surrounding deep-sea would support the isolation aspect. Among the 56 species of harpacticoid copepod species found on the Great Meteor Sea Mount, only two species were known from other regions, and so a faunistic connection between the deep-sea and the sea mount summit was not evident (George and Schminke 2002). This high biodiversity was combined with low dominance values; each species was represented by only a few individuals. These results favor a long-lasting geological isolation with independent radiation (trapping stones). The apparent variability in the harpacticoid situation between seamounts probably depends on the degree of faunistic separation by local geological and hydrodynamic
258
7 Patterns of Meiofauna Distribution
conditions, which are difficult to generalize. Perhaps an assessment of the taxonomic distinctiveness (sensu Warwick and Clarke 1995, see Sect. 8.8.1) would better reflect the degree of isolation rather than counting species? Ascertaining the degree of genealogical concordance by employing (several) genetic markers would also be an appropriate way to determine the zoogeographical role of sea mounts in relation to meiofauna. Which factors drive the dispersal of meiofauna? Not, it seems, the “geological vehicle” of plate tectonics, nor stagnant speciation, nor anthropogenic transport. So what could be responsible for so many closely related yet disjunct meiofauna taxa; for the high amphi-oceanic similarity in meiofauna? What means of dispersal could explain present day distribution patterns of meiofauna on isolated islands or in anchihaline coastal refuges? Did natural rafts drifting across the oceans provide regular transport to suitably adapted micro- and meiofauna? All of these factors combined will account for some exchange of meiofauna across all oceans. This permanent exchange is more intensive in some areas and taxa, and more limited in others. Ubiquitous or large-scale distributions will characterize species predisposed to colonization by their ecological flexibility, eurytopic occurrence and high invasive capacities. Candidates for short-distance transport along coastlines by tidal and bottom currents will be shallow-water and epibenthic species. They exemplify a pattern of dynamic bentho-pelagic coupling within a diffuse boundary layer (Boudreau and Jørgensen 2001). Natural rafts, whether marine snow, sea ice or drifting islands, could have been responsible for repeated transoceanic long distance transport, especially at geological time scales. Such transportation events could support the colonization of remote or pristine islands by meiofauna without any previous contact with continents, as in the cases of the islands of Galápagos or New Zealand. A single contact event would then initiate the evolution of separate phyletic lines, i.e., radiating species from one founder population. The high numbers of monospecific genera among the nematode or gastrotrich fauna of some Pacific atolls (54 nematode species belonging to 47 genera) have been explained by isolated, discontinuous genetic importation from different original faunas and a low speciation rate (Gourbault and Renaud-Mornant 1990). In contrast, a regular gene flow between different areas would be accomplished by repeated genetic exchange and result in a high level of meiofaunal similarity. Natural long-distance dispersal could have provided this regular gene flow, causing little differentiation between, for example, foraminiferan species from the Arctic and Antarctic regions (Pawlowski et al. 2007). As anthropogenic activities and objects in the marine environment continue to increase (ship traffic, floating debris, shore maintenance works, large wide-range bottom dredging), additional potential means of meiofauna transport gain in importance. Both molecular analyses and experimental work offer us tools for specifying “similarity,” and estimating the relative impacts of and the timescales associated with the various distributional pathways for the dispersal of meiofauna. More detailed reading: Sterrer (1973); Palmer (1988); Hicks (1988a); Butman (1986); Armonies (1990, 1994); Eckman (1990); Shanks and Walters (1997); Simon et al. (2002).
7.3 Ecological Aspects of Distributional Importance: Horizontal Patterns
259
Box 7.1 Wide Taxonomic Occurrence vs. Local Individual Range: The Enigma of Meiofaunal Geographic Distribution What are the underlying processes, the distributional pathways that lead to identical taxa with restricted individual mobilities in amphi-oceanic shores or in the groundwaters of different continents? Avoiding the competitive marine shore conditions, many meiofauna evolved in the deeper regions of the sea, but were also displaced into continental aquifers after crossing the brackish transition zone. Relict taxa among the meiobenthos could survive in these refugial biotopes and retain archaic structures. A high degree of specialization (food, reproductive patterns) would support the trend towards isolation and enable sympatric speciation. Other taxa adapted by undergoing slight, initially cryptic mutations to the changing environment. They often formed morphologically closely related sister species or species flocks whose diversity has only been revealed by molecular analyses. And yet, the continental, even cosmopolitan, distributions of many taxa appear to result from a variety of dispersive mechanisms. Passive erosive suspension combined with active emergence may account for the fairly homogeneous meiofaunal distribution along continental shorelines; in freshwater habitats frequent flooding events will distribute many meiofauna. Similarity over transoceanic distances requires repeated contact, which is potentially accomplished by rafting. Natural rafts (floating islands, plants, sea ice) and anthropogenic rafts (vehicles, debris) may be of more local relevance, while the ubiquitous and natural flakes of “marine snow” are considered of global and continuous distributional relevance. Continental plate drift rarely explains genetic cohesiveness at low taxonomic levels. Many enigmatic relations require detailed analyses of genetic similarity in order to reveal the degrees of similarity, the routes and the timescales of dispersal.
7.3
Ecological Aspects of Distributional Importance: Horizontal Patterns
Are there any general patterns of meiofaunal distribution aside from those reflected by geographical and geological conditions? General patterns are difficult to conceive; the characteristic problems involved in assessing meiofaunal distribution have been outlined (for sandy sediments) by Fleeger and Decho (1987). The spatial distribution of meiofauna is notoriously patchy and unpredictably variable. Which factors control these patterns? It appears that we must discriminate again between abiotic and biotic factors. Large-scale meiobenthic distribution seems to be mainly related to physical and chemical parameters with sedimentary and hydrographic heterogeneities as modifying factors. The tidal zones of the oceans provide a good example. Along the slopes of shores, tides account for the large-scale zonation of meiobenthos (Hulings
260
7 Patterns of Meiofauna Distribution
and Gray 1976), since they control grain size composition, water content, salinity and permeability, which factors such as oxygen supply secondarily depend upon. A factor for which it is more difficult to differentiate the various ecological consequences is the general instability of environmental parameters. In sandy shores of the North Sea, with their rigorous physiographic regimes, the meiofaunal abundance is highest close to the mid-tidal line, while in the less exposed muddy sediments the greatest meiofauna abundance and species richness is recorded near the low tide level. The swash zone, with its marked physical and biological variations, displays low meiofaunal densities under strong tidal oscillations. However, it is a preferred zone in microtidal shores (e.g., Mediterranean, Moreno et al. 2006). In the most extreme upper areas of the eulittoral and supralittoral, meiofaunal abundance and diversity decreases. Only some annelids have their preferred habitat in this zone (Fig. 5.46). Perpendicular to the water line of these rather homogeneous shores, the deviations in meiofaunal abundance and composition are small. Small-scale distribution at the centimeter scale appears to be mostly due to biotic interrelations (Li et al. 1997; Snelgrove and Butman 1994; Somerfield et al. 2007), so that it is influenced by a complex factorial combination of attraction (e.g., as a result of reproductive activities) or avoidance reactions (e.g., predation). Aggregations a few centimeters in size are common in meiofauna and often even persist through tidal cycles. Many studies support the view that most of this patchy distribution is determined by aggregations of microorganisms (see Sect. 2.2), selective feeding preferences and direct or indirect trophic interactions (Findlay 1981; Fleeger et al. 1990; Blanchard 1991). However, larger patches of food such as decaying macrofauna can also structure the distribution of many meiofauna through selective attraction (Ólafsson et al. 1999). Small-scale heterogeneities of the sediment, as a biogenic bulk parameter, is generated by animal activities (e.g., bioturbation), become especially important for meiofaunal aggregation in non-tidal shores and in deeper, less disturbed areas (deep-sea) (Thistle et al. 1993). In any case, direct comparisons demonstrate that the horizontal distribution pattern of meiobenthos markedly differs from that of macrobenthos (Fig. 7.4). The divergent size scales of the various components of the benthos appear to be controlled by systems of factors that act differently in each case (see Sect. 9.2). The organic content of the sediments, as a biogenic bulk parameter, is another decisive factor and seems to play a key role in meiofaunal density and distribution. Along the coasts of South Africa, meiofaunal abundance was positively correlated with the detritus content of the sediment (McLachlan et al. 1981). A similar correlation was found in a Mediterranean beach, but the generally lower content of organic matter supported only a relatively poor meiofauna (between 14 × 103 and 715 × 103 ind. m−2) (Moreno et al. 2006). In contrast, the eulittoral of the North Sea, with its high amount of organic matter, especially in the finer sediments, is more densely populated: up to 16 × 106 ind. m−2 (McIntyre 1969) is not an unrealistic level of abundance. Other meiofauna data from a North Sea survey indicate, however, that the general decrease in both density and diversity of meiofauna from the south to the north cannot be simply attributed to grain size conditions and/or organic content alone (Huys et al. 1992). In many cases food overrides the abiotic parameters in distributional importance. In sublittoral and deep-sea bottoms, the reduced concentration of food accounts for the
7.4 Vertical Zonation of Meiobenthos
a
c
Macrobenthos
Ciliates
261
b
d
Nematodes
Ciliates after 14 days
Fig. 7.4 Different distribution patterns of three faunal size groups in the same area (40 × 80 m) of a mid-tide bay (fine sand) of the White Sea (after Burkovsky et al. 1994). Group clusters according to similar species composition.
reduced meiofaunal abundance (see Sect. 8.3). Despite the rather stable physiographic conditions, meiofauna in the subtidal zone are generally 3–4 times scarcer than in tidal bottoms. Towards greater depths, the general decrease in abundance and biomass is clearly attributable mainly to food scarcity: the rain of organic particles resulting from the primary production in the surface layers often controls meiofaunal numbers (Vanreusel et al. 1995a) which often show a clear seasonal fluctuation, with peaks in summer periods (Grove et al. 2006). Since the macrobenthos is even more hampered by food scarcity in deep-water sediments, the meiobenthos often gains in relative importance with respect to both abundance and biomass. Locally, for example in the deep-sea and in muddy fjords, the quantity of the meiobenthos can equal that of the macrobenthos, even attaining a competitive significance. However, even here there is still a close coupling of meiobenthic dynamics with food supply (Pfannkuche 1992).
7.4
Vertical Zonation of Meiobenthos
Demonstrating the parameters that control the general vertical distribution patterns of meiofauna is intricate because of the huge variety of sediments, from sandy beaches to deep-sea muds, that are inhabited by meiofauna. In all vertical sediment
262
7 Patterns of Meiofauna Distribution
profiles, the upper few centimeters have a richer supply of oxygen and food particles and so usually harbor more meiofauna than the deeper horizons. In sandy beaches, Kotwicki et al. (2005a) found that about 70% of the meiofauna was concentrated in the uppermost 5 cm. In finer sediments, with their steeper physiographic gradients, this general pattern is particularly apparent (e.g., Vanreusel et al. 1995a). In a core of silty sediment, Yingst (1978) found that the uppermost two centimeters contained 71% of all the meiofauna present. Mud flats, because of their rich food supply (organic matter), contained about as much meiofauna in the uppermost centimeter as were present in a 10-cm column from a sandy bottom, making the more exposed meiofauna in muds the preferred prey for macrobenthos (Smith and Coull 1987). A general preference of the surficial layers is apparently also valid for protists, which are 2–3 times more abundant in the surface layer than at a depth of 5 cm (Bak and Nieuwland 1989); an exception being Foraminifera, which seem to occur at dysoxic and even anoxic depths (Bernhard 1996). However, there are other, more marked, exceptions to this general pattern. Thiobiotic meiofauna prefer the chemocline at the oxic/sulfidic interface and are only exceptionally encountered in the surficial layer (Giere et al. 1991; Ott et al. 1991; see Sect. 8.4). In mangrove muds and salt marshes, specialized nematodes and harpacticoids have been found down to depths of 15 or 20 cm in layers with reducing Eh values. Macrofaunal burrows and plant roots seem to be major reasons for this deep vertical distribution, although the physiological background of this occurrence remains largely unclear. Detailed distribution studies have also uncovered vertical migrations along a gradient system, with preference and avoidance reactions to tidal change. Tidal currents, with their powerful “tidal pumps,” have a massive influence on fluctuations in the vertical distribution (Joint et al. 1982). In tidal flats, when the sediment is exposed at ebb tide, most meiofauna move closer to the surface, but they migrate rapidly downwards as soon as the tide water reaches the surface (Boaden and Platt 1971). In tidal beaches, the nematode and harpacticoid populations have been shown to change positions by up to 25 cm during one tidal cycle (Harris 1972). In sandy beaches at ebb tide, where fluctuations in water content, temperature and salinity are unfavorably high at the surface, meiofauna react by migrating downward (Figs. 2.4 and 7.5). These vertical migrations of meiofauna into the sediment column suggest an active reaction to the hydrodynamic conditions and vibrational stimuli (Boaden 1968; De Bovée and Soyer 1974; Meineke and Westheide 1979; Foy and Thistle 1991). A preference for deeper layers is especially pronounced in the erosive swash zone. However, any avoidance reaction must “compromise” with the oxygen-related disadvantages of living at a greater sediment depth. The temperature regime can also influence diurnal and particularly seasonal variations in the vertical occurrence of meiofauna. In the summer, most animals of boreal shores live closer to the surface, and in winter they migrate further down and tend to live in closer aggregates (Fig. 7.6). Differentiated by taxon, the more phytobenthos- and phytodetritus-linked harpacticoids are usually more closely associated with the surface layers, where they sometimes even dominate in number, while the more detritus-linked nematodes occur in high numbers further down. This is valid in both littoral and deep-sea sediments from
50
40
30
20
10
1
= 25 ind. 2
High Water
3
Low Water
4
73
Experimental Column
Fig. 7.5 Migratory reaction of meiofauna to a tidal wave. Graphs 1–3 are field samples taken during one 18-h tidal cycle beginning with low water; graph 4 is for an experimental core that simulates high water conditions without wave action. (After Meineke and Westheide 1979)
Depth (cm)
0
Low Water
7.4 Vertical Zonation of Meiobenthos 263
264 cm 0
7 Patterns of Meiofauna Distribution Total Meiofauna
Turbellaria
Nematoda
Copepoda
10 20 Summer
30 40 0 10 20
Winter 30 40 50 0 20 40 60 80100 %
Fig. 7.6 Vertical occurrence of eulittoral meiofauna under summer and winter conditions (After Harris 1972)
polar and tropical regions. However, there are exceptions: In the sediment column of tropical Brazilian beaches, harpacticoids (about 35% of the total meiofauna abundance) outnumbered nematodes (up to 30% of the total meiofaunal abundance). This distinct exception in abundance of the two dominating meiofauna groups was also obvious in North Sea shores, according to the comprehensive study by Huys et al. (1992). Again, in the vertical distribution pattern harpacticoids contrasted with nematodes. The schematic outline of meiofaunal occurrence described above may vary considerably when local distributional patterns are considered. For instance, in eroded shelf sediments, the surface-linked harpacticoid copepods did not show any avoidance reactions to the hydrodynamic forces involving vertical migration into the deeper layers. They tended to prefer suspension rather than living deeper in the sediment (Thistle et al. 1995b). Taking into account the patchiness, variability and frequent water-column transport of meiofauna, the existence of meiobenthic “communities” or consistent “associations” that characterize a particular zone or habitat has only been rarely confirmed, at least in shallow marine sites. When Remane (1933, 1952a) proposed the establishment of interstitial coenoses as a basis for an ecological grouping, this was influenced by corresponding ideas in early benthology, where depth and sediment type were thought to define stable communities. A “Halammohydra coenosis” was defined for coarse sublittoral sands (H. octopodites, some stenoecious archiannelids, turbellarians and harpacticoids), a “Turbanella hyalina coenosis” in sublittoral fine sediment, an “Otoplana coenosis” in exposed eulittoral beaches (otoplanids, some gastrotrichs) or, under rather constant groundwater conditions, a “Bathynella–Parastenocaris coenosis.” Today, with our more dynamic conception of the sediment realm in which many biotic and abiotic factors interact (see Chap. 2,
7.4 Vertical Zonation of Meiobenthos
265
Box 2.8; Snelgrove and Butman 1994), the occurrence of some characteristic species denotes temporary aggregations rather than stable communities. In some studies, nematodes, their more stable numerical and spatial distribution (compare also Armonies 1990), enable relatively significant communities characteristic of a particular sediment structure or organic content to be denominated (see Sect. 5.61; Bongers and Haar 1990; Vanreusel 1990; Vincx et al. 1990). Similarities between the nematode faunas in fine shelf and coastal sediments also led Heip et al. (1985a) to establish “parallel nematode communities.” Another example of the formation of distinct assemblages that can be statistically discriminated (by relative abundance of species) is the ostracods in a specifically structured Californian intertidal phytal habitat (Frame et al. 2008). For littoral harpacticoids from various European sites, Chertoprud et al. (2007) suggested six assemblages with fairly consistent species composition regardless of the geographic area. More detailed reading: McIntyre (1969); Heip et al. (1985a); Alongi (1990a)
Box 7.2 Patterns of Meiofauna Distribution and Their Determinants Examining patterns is always a matter of scale. Large-scale distribution of meiofauna seems to be mainly related to physical and chemical parameters, especially under unstable, extreme conditions. Hence, in coastal areas subjected to tidal fluctuations, the physiographic regime—primarily fluctuations of salinity, water permeability and water supply—are good determinants of the zonation patterns of meiofauna. When we leave the tidal zone and move towards the more stable subtidal reaches, biogenic factors—such as food supply or animal interactions— gain in importance. Factors such as predation, bioturbation or competitive interactions form an interconnected complex of high relevance, especially in minimally exposed environments. However, the most relevant and ubiquitous biogenic factor that determines meiofaunal microdistribution is the supply of organic matter: detritus, bacteria and protists. This becomes particularly evident when comparing exposed with sheltered sediments or shallow with deeper sites. Despite the rather uniform physiographic milieu, meiofaunal distribution is typically patchy. The effective food partitioning of most species differentiates life conditions and determines small-scale (a few centimeters or meters) patterns. The interdependent balance between “organic matter supply” and “oxygen depletion” is the major determinant of most vertical distribution patterns of meiofauna. It demonstrates how closely the biotic and abiotic factors are interlinked: Low organic content keeps oxygen consumption low and even allows meiofauna to live in subsurface layers; rich organic content will attract many meiobenthic organisms to the upper centimeters of the sediment, but increases the risk of oxygen depletion. The consequence are marked preference/avoidance migrations of the meiofauna. Since the structure of the sediment often reflects the oxidative and trophic conditions, certain assemblages of meiofauna (especially of nematodes) can be associated to granulometry.
Chapter 8
Meiofauna from Selected Biotopes and Regions
The ongoing rapid global decline in species number tells us that assessments of natural diversity (“species richness” or “biodiversity”) are necessary not only for scientific reasons, but also to substantiate conservation or sustainable regulation. Thus, studies have been initiated to file our knowledge of meiofauna diversity in numerous marine and freshwater biotopes (see international programmes and projects such as the Global Biodiversity Assessment, UNEP 1995; Convention of Biological Diversity, UNEP 2001–2005; BIOMARE, MarBEF, CenSeam; overview in Costello et al. 2006). Due to methodological problems, estimates of marine biodiversity lag behind the numerous terrestrial census studies. The book by Queirago et al. (2006) provides general information about marine biodiversity, and recently a comprehensive assessment of freshwater species has been published, edited by Balian et al. (2008; see also the earlier compilation by Segers and Martens 2005). Meiofauna were initially not included when comparing marine with terrestrial species richness. About 10 million macrobenthic marine species and between 10 and 100 million meiobenthic species are assumed to exist (Lambshead, pers. comm.). For assesment of the biodiversity of (marine) meiobenthos (both species richness = alpha diversity, and assemblage richness between habitats = beta diversity), many previously neglected regions such as the polar seas, tropical beaches and deep-sea bottoms have been studied in greater detail over the last decade. These have increased our knowledge considerably and some of these sites have turned out to be “hot spots” of meiofaunal diversity. The following chapters will characterize the ecological conditions and diversity of meiofauna in some relevant biotopes from different latitudes. The resulting questions regarding the latitudinal gradient concept (Pianka 1989) and its validity for marine meiofauna will be discussed in Box 8.3. More detailed reading: Lambshead (1993); Gaston (2000); Gray (2000); McCann (2000); Warwick and Clarke (2001).
O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
267
268
8.1
8 Meiofauna from Selected Biotopes and Regions
Polar Regions
Since the 1990s research activity in the polar regions has greatly increased and ecologically oriented meiobenthic long-term studies have been performed in Antarctica (compiled in Vanhove et al. 1998, 2003), and in the Arctic (mainly by Szymelfenig et al. 1995). There are also numerous Russian papers on polar meiofauna of the White Sea shores based on long-term collections, but these papers are not readily available. The more accessible arctic coasts of Svalbard (Spitsbergen) are relatively well studied (see below). Interestingly, the deeper sediments are often better covered by sampling campaigns than the eulittoral (e.g., Hoste et al. 2007; Fonseca and Soltwedel 2007). The increase in the number of taxonomic papers is demonstrated by the monograph of Scott and Marchant (2005) on Antarctic protists, and by numerous descriptions of nematodes provided by Tchesunov (e.g., Tchesunov and Portnova 2005). Therefore, a more detailed picture of the meiofauna in these remote areas can now be provided. The polar and subpolar meiobenthos do not seem, in both abundance and composition, to be essentially different from those found at boreal (resp. southern boreal) latitudes from corresponding depths and sediments. Kendall et al. (1997) ascribe the low endemism of (shallow) Arctic meiobenthos to the fact that they originated at lower latitudes. The similarities of the vertical temperature ranges of Arctic shallow and deep-sea sites may explain the fairly similar abundances of polar meiofauna found at different depths. The eulittoral is characterized by strong seasonal oscillations of two rigid factors to which the meiofauna is apparently well adapted: (1) temperature (an ice cover is present for many months, which is aggravated by destructive ice scouring), and (2) food input (there is a rich microphytobenthos in spring and summer and short pulses of phytodetritus from dense plankton blooms). Due to the strong impact of physical factors, eulittoral meiofaunal populations are relatively scarce, and show high seasonal and inter-annual fluctuations (Table 8.1). In less exposed or deeper muddy bottoms with their rich organic matter, the density of meiofauna is one or two orders of magnitude higher. The sediments under marginal ice zones harbor especially rich meiofauna communities, which feed on the rich sedimented phytodetritus (Fonseca and Soltwedel 2007; Hoste et al. 2007), indicating a close bentho-pelagic coupling (see below). Even the polar deep-sea beneath the ice margins is governed by a rich amount of surficial phytodetritus which provides the basis for a diverse and rich meiofauna. The transition is smooth from this ice margin towards the typical deep-sea basins (Sect. 8.3). Towards the deep-sea and in sediments containing less organic matter, the meiofauna decreases in abundance, despite a physiographically less aggravating milieu. A general difference between the meiofauna densities of Arctic and Antarctic sites does not seem to exist. However, the considerable local variations only permit gross generalizations. Regarding community composition, the polar and subpolar meiofauna correspond to those from boreal latitudes, with nematodes clearly dominating (> 60%,
Sand and gravel Medium-to-fine sand Fine sand Coarse sand Sheltered beach sand Exposed beach sand Muddy sand flat
*Includes Foraminifera; **without Foraminifera
Alaska Muddy sand flat King George Island, Coarse sand and gravel Antarctica Shallow sublittoral South Orkney Islands Fine sand Magellan Strait, Beagle Mud with some sand Channel Admiralty Bay, King George Fine sand Island, Antarctica Brazilian Station, Martel Inlet, Sand, org. enriched King-George Island Deeper shelf, cont. margin Off Spitsbergen Mud? Arctic Laptev Sea Mud Deep-sea Central Arctic Ocean Mud Central Arctic Ocean Mud
Barents Sea
Mid-tidal Low-tidal Bear Island
Spitsbergen (Svalbard)
Eulittoral
Prevailing sediment
Nematodes Nematodes
Nematodes Nematodes Nematodes Nematodes
Nematodes Nematodes Nematodes** Nematodes
6,200 >3,000 3,500–4,000 10,000–15,000
1,150–3,450 500–1,000 70–250* 150–3,400
Nematodes Nematodes Oligochaetes Nematodes Turbellarians Turbellarians Nematodes
Main taxon
500–5,000 400 per 10 cm3
10–110 0–10,000; mostly <1,000 100–900 <50 to >600 250–480 10–270 860–1,300
Abundance × 10 cm–2
Table 8.1 Abundance and main composition of the polar meiofauna from various locations
Location
Schewe and Soltwedel (2000) Hoste et al. (2007)
Pfannkuche and Thiel (1987) Vanaverbeke et al. (1997)
Skowronski and Corbisier (2002) Gheller and Corbisier (2007)
Vanhove et al. (2000) Chen et al. (1999)
Gal’tsova and Platonova (1980) Schizas and Shirley (1996) Arlt and Bick, unpubl.
Urban-Malinga et al. (2004)
Bick and Arlt (2005) Szymelfenig et al. (1995) Urban-Malinga et al. (2005)
Reference
8.1 Polar Regions 269
270
8 Meiofauna from Selected Biotopes and Regions
often 80–90%), followed by harpacticoids, ostracods and/or turbellarians (Chen et al. 1999; Hoste et al. 2007). In sand samples this meiofauna is supplemented by a typical interstitial fauna with turbellarians, gastrotrichs and polychaetes (Arctic: Rysgaard et al. 2000; Antarctic: Skowronski and Corbisier 2002). In (Arctic) eulittoral sediments, enchytraeid oligochaetes were surprisingly numerous and, aside from nematodes and turbellarians, were the dominant taxa (Szymelfenig et al. 1995; Urban-Malinga et al. 2005), while harpacticoids attained only low densities. In muddy sediments of the sublittoral Arctic, rich diatom populations allowed harpacticoid copepods to dominate during the summer season, while in the deeper sediment layers nematodes prevailed (Rysgaard et al. 2000). Towards greater depths, foraminiferans, if included in the evaluation, often surpass nematodes in abundance (Schewe and Soltwedel 2000). The community structure of the littoral meiofauna in polar and subpolar regions follows the typical rules for extreme biotopes: strong variations related to season, substrate and food supply (Vanhove et al. 2000), and occasional mass populations of a few opportunistic species (Ólafsson 1991). Diversity was also low in the eulittoral of Svalbard: for example, only three taxa (Szymelfenig et al. 1995) or six nematode genera comprised 97% of the overall nematode abundance (Vanhove et al. 2000). On the other hand, along the Antarctic continental margin between 15 and 80 nematode genera were found per station by Vanhove (pers. commun.) along the Antarctic continental margin, representing a high genus diversity. The deep-sea areas of the Antarctic and Arctic regions embody a much higher meiofaunal diversity with a high degree of endemism, at least at the species level. However, it seems that in the polar regions this increase with depth does not unequivocally follow the typical exponential curve. Often the intermediate, bathyal depths have a higher diversity than the more oligotrophic abyssal, resulting in a hyperbolic diversity curve (Lambshead 1993). In other studies, the diversity (of nematode genera) decreased with depth (Vanaverbeke et al. 1997) or fluctuated irregularly, apparently caused by changes in sediment composition. The application of different methods may also influence calculations of diversity (see Sect. 8.8). The reduced abundance reported by Hartmann (1990) for Antarctic ostracods does not generally apply to meiofauna judging from the rich stocks of nematodes found in deep water sediments of both the Arctic (Vanaverbeke et al. 1997) and the Antarctic oceans (Vanhove et al. 1995).
8.1.1
Sea Ice
Sea ice with its exotic, highly structured channel system, has gained increasing attention as a habitat for a rich “sympagic” meiofauna, in conjunction with augmented studies of polar, mostly Arctic regions. Sea ice covers, depending on the season, about 35 million km2 of the ocean surface with ice floes that are meters thick. The porous summer sea ice contains an especially huge and complex
8.1 Polar Regions
271
internal channel system: 1 kg of sea ice has an internal surface area of 0.6–4 m2 (Krembs et al. 2000). The channels of varying widths and structures (Weissenberger et al. 1992) originate from the underside of the ice and extend deep into the median and upper parts of the ice floe. This interstitial, lacunar system is filled with brine of that undergoes extreme changes in salinity depending on freezing and melting processes. The texture of sea ice varies greatly depending on how it was formed, its location and its fate. Ice floes formed on the open sea differ from shallow-water pack ice or landfast ice; firstyear ice differs in compactness from many-year ice or ephemeral ice with daily melting and freezing processes. Pennate diatoms and flagellates, provided with permanent light during the summer season and ample amounts of particulate or dissolved nutrients, densely cover the narrow channel walls. With some 200–300 species, diatoms are the most prominent microalgae, both in species richness and biomass (some 100 × 103 cells cm−3 ice). They can grow under the very low light and temperature conditions encountered in winter and at brine salinities of up to 95 PSU. Almost one-third of the total southern primary production is provided by algae associated with sea ice (Spindler and Dieckmann 1994). These microalgae are the nutritional basis for a rich sympagic meiobenthos consisting of a peculiar blend of (1) autochthonous, mostly endemic species that remain exclusively in the ice throughout their life cycles, (2) allochtonous taxa from the ice-adjacent benthos, or even (3) the plankton beneath the ice cover (calanoid copepods and their nauplii). Temporary meiobenthos are represented by numerous polychaete and mollusc larvae. Compared to benthic substrates, high numbers of turbellarians seem to be characteristic of both the Arctic and the Antarctic sea ice and of ecological relevance (Gradinger et al. 1993; see below), since they often dominate in biomass (e.g. 45%; Schnack-Schiel et al. 2001). Interestingly, the turbellarian species in Arctic areas are closely related, if not identical, to those in boreal latitudes. The presence of foraminiferans and ciliates varies widely, depending on the extraction methods used, on the nature of the sea ice, and on the region. In the Weddell Sea (Antarctica), foraminiferans numerically dominated (75%) the sympagic meiofauna, while ciliates dominated in Arctic sea ice by 53% (Gradinger 1999). Nematodes, mainly Monhysterida, are characteristic of Arctic sea ice, especially in coastal fast ice (Nozais et al. 2001). Here, they develop populations of many thousands per m2 (Riemann and Sime-Ngtando 1997). Nematodes from Antarctic sites have not yet been found in sea ice, although they are dominant in eulittoral sediments and many of them tolerate complete freezing. Also, rotifers, which are frequently a major taxon in Arctic sea ice (see Table 8.2), have not been reported from Antarctic ice samples. In the sea ice, foraminiferans, ciliates and harpacticoids—typical grazers of diatoms and flagellates—are at the base of the heterotrophic food web, and many of the unusually frequent acoel turbellarians feed on diatoms too (Gradinger et al. 1993). The food spectrum of the monhysterid nematodes most likely includes
272
8 Meiofauna from Selected Biotopes and Regions
microalgae, bacteria and dissolved organic matter, which is fairly rich in the brine (Gradinger et al. 1992). Cyclopoid copepods, together with predatory turbellarian species, live on other meiofauna, while the network of narrow brine channels prevents access by most (larger) predators, so that in the inner narrow channel system meiofauna are protected from macrofaunal predation. Hence, the utilization of meiofauna by macrofauna (scavenging amphipods, shrimps) is limited but seems to occur at the lateral or lower edges of the sea ice floes, where meiofauna are especially abundant. The most aggravating ecofactor for the sympagic meiofauna are the drastic changes in interstitial salinity caused by the freezing and melting of the seawater. Salinity in the channels can range from 0 to > 200 PSU, with steep gradients in the mm range. The authochthonous ice species Drescheriella glacialis (Harpacticoida) and Cyclopina gracilis (Cyclopoida) tolerate salinities of up to 80 PSU. For the fauna in the brine of the deeper channel system the constraints must be intense, while the peripheral ice portions, which have temperatures and salinities similar to the open water, offer more benign conditions. Is the relatively high proportion of rotifers in Arctic sea ice (22%, Gradinger et al. 1993), normally a group that is common in brackish and freshwaters, related to the occasionally low brackish salinity? Rotifers also have a physical advantage here: their flexible bodies favor life in the narrow ice channels, and they can squeeze through passages 50% smaller than the diameters of their bodies. Turbellarians also match their body dimensions to the varying channel widths by adjusting their osmotic pressures (Krembs et al. 2000). Further physiological and experimental studies that analyze the living conditions in the intricate web of ice channels are needed. Within a single ice core the meiofauna are mostly concentrated in the bottom layers, but even the upper layers harbor a few meiofauna (Dahms et al. 1990; Schnack-Schiel et al. 2001). Terrestrial meiofauna have also found access to polar ice: rich numbers of the enchytraeid Mesenchytraeus solifugus (Oligochaeta) live as “ice worms” or “glacier worms” in the fissures of glacier ice, mainly on the Alaskan coast. They feed on microalgae and pollen, and their strikingly blackish coloration can stain the ice dark (Shain et al. 2001). The changes in the abundance and production of sympagic meiofauna are as irregular as the changes in composition. Table 8.2 demonstrates these variations by region, season, and even in replicate samples. When appropriately processed (salinity-buffered melting of the sample), one core of 1,000 cm3 sea ice can contain 9,000 specimens. Densities from a few thousand up to several hundred thousand per m2 have been recorded (Carey Jr. 1992; Gradinger 2001); nematodes alone accounted for 24,000/m2 (Canadian sea ice, Riemann and Sime-Ngando 1997). However, comparisons are problematic: sympagic fauna in land ice are completely different from those in adjacent pack ice. Drastic changes in meiofauna abundance and composition may occur at the same station from month to month, with abrupt quantitative fluctuations ranging from more than 100 × 103 m−2 to less than 20 × 103 m−2. Rotifera, usually a rare group, can locally dominate; acoel turbellarians can comprise > 60% of the total sea ice fauna (Gradinger et al. 1993), and sometimes even
8.1 Polar Regions
273
Table 8.2 Abundance and main composition of “sympagic” meiofauna in sea ice from various locations (ciliates and foraminiferans are not considered) Area
Abundance (ind. 103 m−2)
Arctic regions Baffin Bay (both Apr–May 98): Pack ice
1.5
Land-fast ice
18.0
Greenland Sea, pack ice, summer 94, 95 Barents Sea, Aug 93 Frobisher Bay: Feb 81
31.7 68.7 17.3–110.3 105.0
May 81 June 81
110.3 17.4
Beaufort Sea: Apr 80 June 80
11.1–48.2 11.1 48.2
Stefansson Sound:
4.5–8.0
Mar 79
8.0
May 79
4.5
Composition (% ranked)
Nozais et al. (2001) 90 cop + naup/8 nem/ 1.1 pol 97 nem/3.3 cop + naup/ 0.1 pol 68.1 nem/5.5 turb/2.4 cop/6.9 rot 25 rot/25 nem/4 turb
Gradinger et al. (1999) Friedrich (1997) Grainger et al. (1985)
88.9 cop + naup/10.1 nem/0.4 pol 98.6 nem/1 cop + naup 92.3 nem/4.7 cop + naup/ 2.7 rot Kern and Carey (1983) 45.9 pol/45.7 cop/3.2 nem 51.9 nem/31.4 turb/ 14.5 cop Carey and Montagna (1982) 67.3 pol/32.2 cop. + naup/0.7 nem 76.9 nem/23.1 cop + naup/0 pol
Central Arctic Ocean: Aug–Sept 91
Reference
Recalculated from Gradinger (1999) ca. 15
46.9 nem/28.5 rot/15.4 cop. + naup/15.4 turb
Northern Fram Strait:
Schünemann and Werner (2005)
Summer
0.6–34.1
Winter
3.7–24.8
45% cop/33% rot/16 nem/3 turb 93% naup/4 rot/2 cop/ 0.5 turb/0.1 nem
Antarctic regions Weddell Sea:
Recalculated from Gradinger (1999)
Sept 89
1.5
Apr–May 92
78.1
34.4 cop + naup/31.2 turb/7.3 others 51.8 cop + naup/48.3 turb/5 others
274
8 Meiofauna from Selected Biotopes and Regions
temporary meiofauna can form the bulk of the sea ice fauna. Arctic sea-ice fauna differs profoundly from its Antarctic equivalent. Compared to the biomass of flagellates and diatoms (100 mg C m−2), meiofauna (5–7 mg C m−2 sea ice) represent only a few percent (Friedrich et al. 1996), while the ciliate contribution is 20% (macrofaunal amphipods represent less than 1 mg C). In terms of biomass, the intensive primary production performed by microalgae (although varying with season, region and texture of the ice) exceeds that of the sea ice meiofauna more than tenfold. This is higher than required by the meiofauna, so rates of diatom ingestion by the sympagic meiofauna are only between 3–4% (Arctic) and 16% (Antarctic) (Gradinger et al. 1999; SchnackSchiel et al. 2001), even during the dark winter season even included. For Arctic sea ice in Baffin Bay, Nozais et al. (2001) noted a similar rate of consumption of the primary production (6%). These low values correspond to daily grazing rates of only 1% of the algal production. These calculations suggest unlimited feeding conditions for the sympagic meiofauna in most areas (Gradinger 1999; Nozais et al. 2001; Gradinger et al. 2005). Within meiofauna, turbellarians, which are often the dominant taxon by biomass, are the main predators and have considerable grazing rates (Gradinger et al. 1993). Sea ice meiofauna is not an isolated and stable biota. It closely interacts with the benthic and pelagic environments depending on the water depth, its proximity to land and its age (Werner 2005). Shallow water ice (water less than 10 m deep) predominantly contains a selection of meiofauna from the underlying benthos, whose populations are much larger than those in the overlying ice (Carey 1992). The frequently suspended harpacticoids and cyclopoid copepods and turbellarians (see Sect. 7.2.1) easily colonize the ice habitats. Each bottom contact of an ice floe, of which 80% is below the surface, provides benthic meiofauna with the chance to access the ice interstitial. With each melting or freezing period, seasonal ice floes release and take up meiofauna and thus contribute not only to the coupling of bottom fauna and ice fauna, but also by long drift passages to the distribution of the meiobenthos via long drift passages (Carey 1992; Schnack-Schiel et al. 2001). During ice-free periods, land ice represents a retention habitat and provides a “stepping stone” that enables recolonization when new ice is formed in winter. Figure 8.1 illustrates some seasonal processes in sea ice colonization. Additionally, during freezing periods ice platelets are formed and lifted from the bottom, with some sediment underneath. This may contain meiofauna which are then incorporated into the ice system, and this would explain why coastal fast ice usually harbors more meiofauna than remote sea ice (Gradinger et al. 2005). The “benthic-oriented” Arctic sea ice in particular permits an intensive interaction between the benthic and sympagic meiofauna (Carey 1992). However, only about one-third of all sympagic meiofauna seem to be of benthic origin. Oceanic sea ice contains a blend of autochthonous meiofauna, meroplanktonic larvae and true plankton organisms. It is not clear whether they are enclosed while the floe is freezing or they actively colonize the channel system.
8.1 Polar Regions
Winter
275
Spring
Summer
Autumn
ICE ???
Ice algae & phytoplankton Naupliar larvae Sympagic amphipods
Sympagic meiofauna (copepods, turbellarians, nematodes, rotifers) Pelagic copepods
Fig. 8.1 Seasonal pattern of sea ice colonization by meiofauna, emphasizing the intensive cryo-pelagic coupling (After Werner 2005)
In any case, freezing and melting processes contribute to their uptake and their release to and from the ambient sub-ice plankton (Fig. 8.1). During seasonal ice-free periods in the open water, the remaining areas of fast ice in coastal areas serve as refugia for ice fauna and as seeding grounds for the colonization of newly formed ice floes. The origin and fate of meiofauna in old ice floes drifting across the open polar seas and their pathways of colonization are often unclear. The various endemic species (e.g., the harpacticoid Dreschierella glacialis, the cyclopoid copepod Cyclopina gracilis, perhaps also the nematodes Cryonema and Hieminema) must have had long evolutionary lines with authochthonous, ice-bound life histories. Chunks of ice originating from multiyear ice floes may have provided appropriate rafts for (long-distance?) transport and further distribution. In all polar regions, sea ice acts as an important retention substrate and transport vehicle that contributes to meiofauna distribution. The extent to which the sea ice meiofauna represent truly benthic or sympagic fauna may vary with the texture of the ice, and with region, season and climate; quantitative data remain to be assessed. However, the ecological role of meiofauna in/on sea ice is certainly important. A universal ice melt resulting from global warming would point to a bleak future for polar ecosystems and their fauna, including their meiofauna. More detailed reading: Carey (1985); Spindler (1994); Gradinger et al. (1999); Gradinger (2001); Nozais et al. (2001).
276
8 Meiofauna from Selected Biotopes and Regions
Box 8.1 Life Under Icy Conditions: Meiofauna in Polar Regions Despite the harsh conditions, meiofauna seem to thrive in Arctic and Antarctic regions. Their abundance and diversity are often no lower than those in temperate climates, and many boreal species also populate polar bottoms. Polar meiobenthic life depends on pulses of phytodetritus from the blooms of microalgae during the polar light season and is characterized, as in other extreme biotopes, by strong seasonal, interannual and local fluctuations. In the shallow bottoms mechanical disturbance by ice scouring adds to the physical stress, but under more favorable conditions densities of several 1,000 meiofaunal individuals per 10 cm2 are common. Nematodes and harpacticoids are normally the dominant taxa. In eulittoral sediments oligochaetes are common, while towards greater depths foraminiferans become increasingly dominant. Here diversity is often highest and many endemic species are found. A particular compartment of polar meiofauna lives in the channel system of sea ice. On the walls of this interstitial web filled with water of changing salinity thrive masses of microalgae, the food basis of a rich “sympagic” meiofauna, dominated by ciliates, foraminiferans and copepods, turbellarians and rotifers are unusually frequent. Nematodes are common in the Arctic ice but they, as well as rotifers, have not yet been found in Antarctic sea ice. Sea ice meiofauna are patchily distributed and show strong temporal fluctuations. This depends, apparently, on the processes forming the channel system, on the position and size of the ice floe, its distance from land, and on changing light conditions. Donor populations are from the bottom or from suspended meiofauna. A few species live their whole life cycles in the ice. They are often endemic and highly adapted to the special conditions in this exotic biotope, e.g., extreme salinities and temperatures. The meiofauna of sea ice is a food-unlimited community of specialists with reduced predation pressure. Only the rich meiofauna populations on the underside of sea ice are accessible and grazed upon by various macrofauna (particularly amphipods). Sea ice is a nutritious sheltered biotope for many meiofauna and a nursery ground for macrofauna. Linking the benthic and planktonic polar food webs, sea ice is of considerable ecological relevance. The consequences of its accelerated melting in the global warming process for oceanic life are by no means understood.
8.2
Marine Subtropical and Tropical Regions
In the last decade, studies of the meiofauna from subtropical and tropical regions have greatly increased following the pioneering studies on Indian beaches (McIntyre 1968), various Pacific islands (French researchers around Salvat and RenaudMornant) and the Bermuda Platform (Coull 1970). In a review of tropical meiofauna,
8.2 Marine Subtropical and Tropical Regions
277
Alongi (1990a) outlined a picture with large geographical, biotopical and seasonal variations. The tropics have a great range of habitat types for meiofauna, with carbonate sands on beaches and shelf regions of carbonate sand, estuarine muds, mangrove thickets, and enclosed lagoons. The present account will first focus on some general data and then present the meiofauna of some characteristics tropical biotopes. Despite the usually oligotrophic tropical seas, the abundance of littoral meiofauna in the tropics is very similar to that in temperate coastal areas: from several hundred to several thousand specimens per 10 cm2 (McIntyre 1968; Gourbault and Renaud-Mornant 1990; Alongi 1990b; Vanhove 1993). Dense populations beyond this range (e.g., >10,000 per 10 cm2 on the Malaysian coast; 17,000 nematodes per 10 cm2 in an Indian salt marsh) represent probably local aggregations and should perhaps not be generalized. In samples from the subtidal and the continental shelf, mean densities gradually decrease. Even the large local variations correspond to conditions in temperate regions and do not allow generalization. Unexpectedly, under tropical conditions the density fluctuations often also exhibit a seasonal pattern (Bermuda: Coull 1970; Galápagos: Westheide1981; Philippines: Faubel 1984; Red Sea: Arlt 1993). However, in contrast to temperate climates, the richest populations often develop in the cooler parts of the year (at least in habitats in the “dry tropics,” see Alongi 1990b). Breeding and reproduction are also mostly seasonal, and are often adjusted to avoid climatic extremes such as torrential rainfalls in the monsoon season. During their reproductive phases, even those meiofaunal species adapted to tropical conditions are limited in their tolerance to changing physical conditions (see Sect. 8.6). Monsoonal floods, with their high mud loads of river run-off, hurricanes (typhoons) and cyclones, which are characteristic of the tropical girdle, can cause a sudden and sometimes complete turmoil of the ecosystem, with severe destruction of meiofaunal assemblages, especially in the eulittoral (Suresh et al. 1992). A similarly negative effect has been recorded in South Africa after the seasonal flooding of estuaries (Nozais et al. 2005). However, these are natural disturbances to which meiofauna seem adapted, since the populations normally recover fairly rapidly, albeit often with an altered community composition (Alongi 1990a; Ansari and Parulekar 1993). Typical of certain tropical areas are oscillations such as “El Nino” or seasonal upwellings. These hydrographic events also have a marked impact on coastal meiofauna. By enriching the nutrient supply they augment meiofauna populations. Where river inputs provide rich organic matter, meiofauna will develop into particularly dense populations, so long as the load of fine sediment does not lead to anoxic events. In estuaries and coastal lagoons the great salinity and temperature stress reduce meiofauna to densities of around or less than 100 ind 10 cm−2 (Alongi 1990b); this is lower than observed in corresponding temperate or boreal brackish waters. As seen in non-tropical meiofauna, grain size composition, the complex indicator of abiotic factors (see Sect. 2.1), seems to influence meiofauna composition: community structure in tropical sediments mostly corresponds to the typical pattern, with nematodes as the most abundant and diverse group. However, in coarse sands with a low silt content harpacticoids may prevail. This relation to grain size and silt content has been found in both tropical quartz and biogenic coralline sands. In contrast with most studies from temperate regions, polychaetes and oligochaetes play a substantial
278
8 Meiofauna from Selected Biotopes and Regions
role in the tropics and can represent important taxa (Ansari and Ingole 1983; Faubel 1984; Ingole et al. 1997; Westheide 1991; Sasekumar 1994; Villora-Moreno 1997; Netto et al. 1999). On the other hand, a local scarcity of annelids may suggest the presence of turbellarians, their main predators, which occasionally exhibit considerable abundance in tropical studies.
8.2.1
Tropical Sands
The littoral fringes of many tropical seas, atolls and islets consist of calcareous biogenic sands. These splintery biogenic sediments are structurally complex, relatively little sorted and of high porosity. In experimental sediment columns, permeability of calcareous sand was often markedly higher than in corresponding siliceous sand. The “open” grain surfaces of calcareous particles favor adsorption of nutrients that are the basis for rich organic matter and large stocks of microorganisms (Suess 1973; Rasheed et al. 2003a,b; Wild et al. 2005). This means an attractively rich food supply for meiofauna (Dahms et al. 2007). Their assemblies consist of many different taxa, albeit with just a few individuals each, they have an unusually high biodiversity. This richness contrasts with the often low number of individuals per taxon (RenaudMornant & Serène 1967; Coull 1970; Gúzman et al. 1987). Gourbault et al. (1998) identified in their study on the beaches of Guadeloupe 122 spp of nematodes belonging to 112 genera, and the 42 spp of harpacticoids found by Villiers and Bodiou (1996) in Polynesia belonged to 21 genera! In somewhat coarser sediments (unprotected beaches and reef slopes), the dominant group is often the harpacticoid copepods; the nematodes here are relatively large, epigrowth-feeding or predacious types. Many other, often rare, taxa (foraminifera, interstitial ostracods, polychaetes, molluscs, priapulids, and tardigrades) also occur. Overall, in the coarse, exposed sites meiofauna densities are relatively low (< 500 ind. 10 cm−2; see Guzmán et al. 1987). In contrast, the finer sands and muds of more sheltered sites (lagoons, pools) harbor a more monotonous meiofauna with numerous small, deposit-feeding nematodes dominating (Coull 1970; Vanhove 1993; but see contrasting conditions in Gourbault and Renaud-Mornant 1990), (e.g., 30–540 × 10 cm−2: Grelet et al. 1987 (80–90% nematodes); about 1000 inds × 10 cm−2: Netto et al. 1999; several thousand × 10 cm−2: Guzmán et al. 1987); the turbellarian fauna is also rich here. Biomass at the surface was almost 4 g m−2 (wwt, 0–5 cm), twice that of corresponding records from North Sea sands (Grelet 1985). This increase in meiobenthic abundance with the reduction in the permeability of the interstitial system occurs mostly at the expense of taxonomic diversity. The switch from copepod to nematode dominance can occur within short periods of time (<1 year), depending on the fluctuations in currents. Thus, the rule that the numerical density of the meiobenthos increases with the proportion of detritus and silt (up to certain limits) remains valid also in biogenic sediments. The high taxonomic diversity in more exposed habitats is often reflected by a high statistical diversity (H’), especially in the surficial layers. H’ values (for nematodes) of >5 in the Great Barrier Reef or even up to 11 for the Red Sea coast have been reported.
8.2 Marine Subtropical and Tropical Regions Nematodes
Harpacticoids
Jacc. 1.0
0.5
279 Polychaetes
Others
Renk. % 20
1 0
Jaccard Index Renkkonen Number
Fig. 8.2 Comparison of similarity (Jaccard index) and dominance (Renkonen index) in meiofaunal samples from calcareous and siliceous sands in two adjacent Mediterranean sites. Low Jaccard values indicate marked differences between the species at the sites, while Renkonen percentages indicate the degree of similar dominance relations in the compared sediments (Giere et al., unpubl.)
More sheltered lagoonal ecosystems in the Pacific had a lower diversity of >3 (Gourbault and Renaud-Mornant 1990). On the other hand, Coull (1970) reported that the highest species diversity of Bermudian copepods occurred in the muddy, not the sandy, substrates. Hence, a simple relationship between granulometry and species richness does not seem to exist (Boucher 1990). In a direct comparison between the meiofauna of calcareous and siliceous sands from Italian shallow sites (Giere et al. unpubl.), indices of similarity and dominance differed profoundly, while the abundance, taxonomic richness and H’diversity remained comparable (Fig. 8.2). These differences in meiofaunal community structure despite similar granulometry became most evident at the species level, not when higher taxa were compared. Many calcareous tropical sediments harbor a typical thiobiotic and sulfidedependent meiofauna at depths below 10 cm (Ott et al. 2003; Bright and Giere 2005; Van Gaever et al. 2004; see Sect. 8.4.2). How can anoxia/sulfide develop beneath the upper few oxic centimeters in sediments of high porosity and in hydrodynamically exposed areas? As evidenced by the mostly low degree of sorting, between the calcareous fragments of shells and skeletons, a large amount of fine, powdery abrasion accumulates which tends to clog the pores, causing low permeability in the deeper strata. In addition, mucilage sheaths, which are not present to such a large extent on silicate grains, coat the particles (Suess 1969, 1973; Rasheed 2003a,b). These biogenic films and the rich organic content of the sediments may cause the rapid oxygen depletion in the deeper layers. The zonation of meiofauna on tropical shores corresponds to that in temperate climates: the highest diversity but not necessarily abundance occurs near the lowwater mark, a lower diversity is found in the higher shore. The intermediate zone often harbors fairly high meiofauna abundances. In the shelter of tropical algal assemblages, rich in habitat structure and food supply, a particularly abundant and taxonomically diverse epiphytic meiofauna is commonly observed (Faubel 1984; Arlt 1993; De Troch et al. 2001). Microhabitat complexity, exposure and sediment transport are major determinants of meiofauna abundance and composition in these phytal communities (Muralikrishnamurty 1993).
280
8 Meiofauna from Selected Biotopes and Regions
Box 8.2 Calcareous Sands: A Bonanza of Fascinating Meiofauna The rubble in tropical coral reefs, the sediments in many marine caves, the tops of many sea mounts, the beaches of atolls, the shoals along limestone shores: these all are of biogenic origin, and their sands have fascinating structural complexity. A look through a microscope shows a colorful sedimentary world: white coral rubble, fancy foraminiferan tests, reddish pieces of calcifying algae, crenulated fragments of echinoderm spines and sculptured parts of mollusc shells—and it shows a wonderful meiofauna, often with marked body structures and color patterns in white and red. Rare taxa like tiny holothurians and bryozoans, bizarre tardigrades and loriciferans, agile isopods, reddish harpacticoids or white bacterial symbiotic annelids and nematodes occur here. What causes this high meiofaunal diversity and endemism? Is it just the irregular, often splintery particle shapes? Are the porous surfaces of calcareous particles the clue to their high absorptive capacities? They bind many organic substances; their surfaces are coated with biofilms. Are specific biofilms the reason for the rich colonization with microorganisms and meiofauna? Whatever the case, their specific attractiveness to a diverse, often unusual meiofauna has been shown in direct comparisons. Contrasting in their origin, texture and probably also microbial composition, calcareous sands differ profoundly from their siliceous counterparts in their meiofaunal content. Irrespective of latitude, water depth and biotope, to the zoologist they represent a bonanza of new and interesting animals. The characteristic nature of these attractive habitats requires more comprehensive research.
8.2.2
Mangroves
One of the most characteristic, biologically rich and ecologically important habitats of the tropics, the mangrove girdle, has also attracted studies on its meiofauna. As in other tidal shores, differences in exposure generate various zones. The high shore with the most rigid fluctuations is usually the least populated, while the mid-tide or low shore harbor the densest populations of mangrove meiofauna. Although the sediment compositions in these zones vary slightly (the low shore has somewhat coarser particle fractions), in general mangrove sediment consists of fine mud that is rich in organic matter. With their low porosity and water percolation, mostly brackish and fluctuating salinities and limited oxygen supply, mangrove sediments are comparable to the tidal flats of temperate regions. In contrast, though, mangroves are characterized by different highly adapted and partly submerged mangrove trees. Their dense prop roots, the pneumatophores and the leaf litter create a much higher structural diversity than seen in temperate tidal flats. The horizontal distribution of meiofauna is determined by tidal changes in inundation, salinity and temperature. Tropical seasonal monsoons cause extreme salinity fluctuations and, along river mouths, flooding by riverine silt loads. Intense microbial degradation restricts the
8.2 Marine Subtropical and Tropical Regions
281
oxygen supply in the sediment to the uppermost millimeters, aggravating meiobenthic life. In addition, the tidal currents and torrential rainfalls can erode the silt, creating patches of somewhat coarser sediment, at least in the lower reaches. A specific chemical factor for mangroves seems to be the high tannin content in the mangrove litter and pore water (Alongi 1987a,b). This habitat variability plus seasonal fluctuations (pre-monsoon vs. post-monsoon) may contribute to the extreme differences in meiofauna associations. Population densities and depth of vertical zonations differ in many studies (Alongi and Sasekumar 1992; Castel 1992). Thus, the mangrove meiofauna, like that found in silty tidal mud flats, must not only be adapted to temporary oxygen depletion and changing salinities, but it must also tolerate the adverse effects of tannin. It is thought that the low ability of meiofauna to break up mangrove litter may reduce meiofauna abundance (Coull 1999). Records on the vertical distribution of meiofauna in mangrove sediments are contrasting. Some authors have encountered meiofauna, mainly nematodes, way beyond the oxic/anoxic interface in the reducing layers, down to 15 or even 20 cm depth (Nicholas et al. 1991; Vanhove et al. 1992; Ansari et al. 1993); others claim that the presence of meiofauna at depths below a few centimetres is negligible (Sasekumar 1994). For an account of the problems associated with meiobenthic life under conditions of low/no oxygen, see Sect. 8.4 on thiobios. In almost all mangrove studies, nematodes, especially Daptonema and Microlaimus, represent 80–90% of the meiofauna. However, they do not form a biologically uniform group. In the different mangrove zones the variations in the silt content of the sediment relate to the nematode feeding type (Schrijvers and Vincx 1997). In the coarser sediments of the low shore and on roots and leaf litter, epistrate feeders (e.g., Spilophorella, Ptycholaimellus in a Kenyan mangrove) are more common (Nicholas et al. 1991; Ólafsson 1995; Chinnadurai and Fernando 2007). Their occurrence coincides with the increase in benthic microalgae and phytodetritus (Alongi 1990b). In Australian mangroves, the mid-level, particularly the silty zone, harbored numerous deposit-feeding nematodes, while omnivores and predators prevailed in the high water zone (Alongi 1987a; Nicholas et al. 1991), resulting in proportions of 50, 28 and 22%, respectively. In the deeper layers, the typical “sulfide nematodes” (Metalinhomoeus, Astomonema or Catanema) can prevail (see Sect. 8.4.2). However, the proportions of feeding guilds are not static. They can change rapidly depending on the seasonal input of microalgae and phytodetritus and on the silt loads that are often flushed in by monsoonal rains. A clearer relation to sediments was evident in harpacticoid copepods. Low species numbers are typical of muds, and much higher ones of sand. Harpacticoids are usually second in abundance (around 5–10%) after the nematodes. The lower, more sandy mangrove girdle not only harbors richer populations of these copepods than the higher shore, but it has also the highest taxonomic diversity (Alongi and Sasekumar 1992; Ólafsson 1995). In very soft mangrove mud, members of the Canuellidae, Ectinosomatidae and the genus Stenhelia were the dominant harpacticoids in Queensland, Australia (Coull et al. 1995). Favored by detritus-rich silty bottoms and often brackish water, burrowing macro- and meiobenthic tubificid and naidid oligochaetes are common in mangroves (5–19% in African mangroves). Although not often
282
8 Meiofauna from Selected Biotopes and Regions
included in studies of meiofauna, ciliates can be quite common among mangrove meiofauna. In a South African mangrove they were the second most abundant (6.4%) after nematodes (80%) but before oligochaetes (4.5%, Dye 1983). In the more seaward Avicennia muds, kinorhynchs also attained considerable densities (up to 5%). Although both northern tidal flats and mangroves are detritus-based decompositional ecosystems, in mangroves the overall densities and diversities of meiofauna are often lower than those found in corresponding intertidal mud flats of temperate/boreal shores. Particularly because of their higher structural complexity (vegetation) is the epibenthic meiobenthos in mangroves more abundant (and more diverse) than that in northern tidal flats. About 1,000–7,000 individuals (mostly nematodes) per 10 cm2 have been documented in mangrove samples from different areas and continents (Dye 1983; Nicholas et al. 1991; Castel 1992; Vanhove et al. 1992; Vanhove 1993; Ólafsson 1995). Structural complexity would also explain why American salt marshes, that lack the extreme aggravations of mangroves, can harbor enormous meiofaunal densities (Wieser and Kanwisher 1961; Teal and Wieser 1966; Bell 1979). However, there are also records of much lower densities from non-estuarine and unpolluted mangrove forests (Alongi 1987a; Alongi and Sasekumar 1992; Chinnadurai and Fernando 2007), reflecting the high variability of local conditions and climates. The seasonal impact varies depending on the geographical location (the distance from the equator). In the dry tropics, the hot temperatures in spring and summer seem to reduce meiofauna, while in subtropical areas, (late) summer produces the highest densities. In general, the torrential rainfalls of monsoons have a negative impact on meiofauna, so that lowest densities were found in the post flood period (India). Biomass values from mangrove meiofauna are rare. Varying seasonally and locally, they may range (mean biomass) between 0.2 and 2.3 g C m−2, which, on the basis of an assumed turnover rate of 8 or 9, would give an annual production of as high as 1.5–8.4 g C m−2 (Dye 1983; Vanhove 1993). Similar to temperate tidal flats, in mangroves the dynamics of bacteria/microalgae communities and protozoan/meiobenthos assemblages are tightly linked (Schrijvers et al. 1995; Schrijvers and Vincx 1997): In the African mangroves, exclusion experiments on epibenthic grazers (gastropods) increased the stock of microalgae (chlorophyll a concentration) as well as the prevalence of epistrate-feeding nematodes. Exclusion of macro-epibenthic detritivores resulted in an increase in silty debris as well as a parallel increase in the population of detritivorous meiofauna (deposit-feeding nematodes and oligochaetes). Hence, a link between two different decompositional chains was revealed in the meiobenthos, one based on microalgae, the other based on detritus. Stable-isotope analyses (13C) showed that phytoplankton-derived seston was also an important food source in the upper centimeters of the sediment, which calls into question the prevailing trophic role of autochthonous mangrove litter. The rich epibiota in the thickets of prop roots and trees is another assemblage in mangrove shores that influences the mangrove meiobenthos by their structural diversity. Areas densely overgrown with mangrove vegetation often embody higher meiofauna densities. On the other hand, areas with many burrowing crabs have been found to be less densely populated (Dye and Lasiak 1986). This decline was interpreted as being more caused by disturbance and competition than by predation. Conversely, dense populations of epibenthic gastropods in a tropical mud flat led to the destabilization of the sediment surface and increased meiofaunal fluctuations
8.2 Marine Subtropical and Tropical Regions
283
(Carlén and Ólafsson 2002). Whether this negative effect was caused by competition for food or by physical disturbance is hard to discriminate in the field. However, in mangroves, with their abundant natural food supplies, why would competition among deposit-feeding snails and the much smaller nematodes exist at all? In all probability, the continuous fluctuations that characterize this habitat would rapidly reduce any long phases of resource limitation. Before we can generalize about interactive contacts between meiobenthos and other live compartments (Alongi and Sasekumar 1992; Schrijvers et al. 1995; Schrijvers and Vincx 1997), further experimental scrutiny is needed to gain a better understanding of the unique ecosystem of mangroves. More detailed reading: Alongi (1990a); Alongi and Sasekumar (1992); monograph, Alongi (2008).
Box 8.3 Tropical Plethora vs. Polar Purity: A Latitudinal Diversity Gradient in the Meiobenthos? The latitudinal gradient, derived from terrestrial studies, is one of the bestknown large-scale biodiversity patterns. The mechanisms governing this pattern, which have been disputed since the review of Pianka (1966), remain a challenge for the marine realm. But is the “classical” decline in species diversity from the tropics to the polar seas globally valid, and does it also apply to meiofauna? Has it only developed in the deep-sea or in shallow sites too? Many questions, but they have different answers. Recent and fossil Foraminifera seem to globally decline in species richness from the Equator to the North (Culver and Buzas 2000), However, nematodes in various deep-sea areas show a converse decline with increasing numbers of species towards the Arctic. This pattern was found to be linked to increased surface productivity and supply of organic matter (“food-driven gradient”), and is not related to a latitude gradient per se (Lambshead et al. 2000, 2002). In the southern oceans and the Antarctic deepsea, the highly diverse assemblages of meiofauna challenge the contention of a depressed diversity at higher latitudes (Brandt et al. 2007). Other studies and taxa did not demonstrate a latitudinal gradient, whether in nematodes or in harpacticoids. At shallow sites, meiobenthic species richness and diversity were as high as or even higher in temperate and boreal habitats than in the tropics (see Kotwicki et al. 2005b). An absence of any gradient between shallow tropical and temperate sites was also experimentally confirmed for nematodes (the main representative of meiofauna) using artificial collectors (Gobin and Warwick 2006). Also, in studies of freshwater meiofauna, a relationship between species number (mainly harpacticoid copepods) and latitude could not be discerned (Reid 1994; Rundle et al. 2000). A product of several interacting mechanisms, the latitudinal diversity index is often blurred by local geographic, historical and ecological variables. The lack of convincing evidence for a latitudinal diversity cline in the meiobenthos also may be partly due to our insufficient knowledge of regional diversification and the confounding effects of production and circulation patterns, but also due to differently sized samples, taken and evaluated by different researchers with different methods.
284
8.3
8 Meiofauna from Selected Biotopes and Regions
The Deep-Sea
Deep-sea research is largely instrument-limited. Since the first study of the deep-sea meiobenthos (Wigley and McIntyre 1964), the development of suitable corers (e.g., the multiple Aberdeen corer, Barnett et al. 1984) has enabled quantitative evaluations. Today, scientists on numerous research cruises are equipped with sophisticated remotely controlled instrumentation, and can even perform experiments on the deepsea bottom, deploy and retrieve automatic devices such as “bottom landers,” and record environmental variables on-line. The application of new analytical methods (e.g., analysis of proteins, chloroplastic pigments, adenosine nucleotide content, and electron transport system activity) has refined our knowledge of the deep-sea meiofauna. These biochemical parameters that are related to biomass often render more reliable results than those obtained by direct counting and weighing.
8.3.1
The Habitat
Compared to the shallow benthic zones, the bathyal and abyssal bottoms are rather static and monotonous; wide regions of the muddy bathyal and abyssal plains represent a fairly uniform “desert environment.” However, interspersed with the widely prevailing mud plains are hydrodynamically complex areas where water currents are strong enough (5–10 cm s−1) to suspend the silty deep-sea sediment and form sandy patches (Thistle 1988). Upwelling regions can create oxygen minimum zones (OMZ), while mountainous ridges cause complex and little-explored smallscale near-bottom currents. We know of big river outflows, steep canyons and disastrous turbidites that influence the assemblages of deep-sea fauna. In addition to these hydrodynamic and geological patterns, manganese nodule fields, cold-water reefs and aggregations of sessile macrofauna result in a benthic structural heterogeneity that is greater than previously assumed and is important for the diversity and distribution of meiofauna. However, perhaps the most biologically important and unexpected factor is the seasonality of deep-sea processes (Gooday 2002), because this vast depth is coupled to the phototrophic surface, inferring that the ocean is one interactive biome that is under continuous change. While the temperature of the deep bottom is usually 1–2 °C, it is considerably elevated in the Red Sea (21 °C), the Mediterranean (about 10–14 °C), and in sediments associated with volcanic and hydrothermal activity (5–35 °C). Oxygen content is one of the most important abiotic factors in shallow sites, but wide areas of the deep-sea bottom are well oxygenated, often down to 10 cm depth. This results from the small amount of degradable organic carbon present in most deep-sea sediments. Conversely, in upwelling areas, which have a rich input of sedimenting debris after plankton blooms, oxygen can become limiting for many meiofauna, at least in the subsurface sediment layers. Low oxygen contents also occur in warm deep-sea regions (Red Sea, Sulu Sea in the Pacific Ocean) and
8.3 The Deep-Sea
285
remind us that complete anoxia prevailed in the deeper oceans for millions of years (e.g., during the oxygen crisis at the Permo–Triassic transition or the Mid-Cretaceous). Decaying plankton blooms cause seasonal pulses of organic matter that largely sustain the deep-sea ecosystems. Fluffy masses of phytodetritus reach the bottom after weeks and settle there as a greenish (from chloroplastic pigments) unconsolidated surface layer of high nutritive value (Pfannkuche and Thiel 1987). Since most meiofauna prefer this layer, it is imperative to prevent the “green fluff” from being flushed away by inadequate sampling if reliable results are to be obtained (Bett et al. 1994). Thus, variable deep-sea meiofauna recordings do not necessarily reflect merely local and seasonal fluctuations in sediment structure; it may be that they are attributable to inadequacies in sampling methods. There are three main factors that control the structure of meiofauna assemblages in the deep-sea. 1. Sediment characteristics (mud vs. sand). Changes from muddy to sandy areas caused by hydrographical properties and different sedimentation rates are paralleled by a change in community structure. In sediments with a high content of fine particles (silt), the meiofaunal abundance is usually relatively high and dominated by the large community of nematodes that can even live in deeper, often anoxic layers. Especially in cores with calcareous ooze, meiofauna can attain high densities (Shirayama 1984). In sandy areas a relatively diverse meiofauna is dominated by harpac ticoids in the upper few centimeters. These typical changes in community composition can be best demonstrated on neighboring flanks of ridges with different exposures and sedimentation rates (Jensen et al. 1992a). 2. The supply of organic matter (measured as chloroplastic pigments or adenylates), which is usually reflected by the silt content of the sediment, regulates the abundance of deep-sea meiofauna in all oceans. Every degradable organic particle added, whether it comes from seasonal phytodetritus or from horizontal advection by currents, influences the composition and abundance of the meiobenthos in this precarious nutritive environment (Thiel et al. 1988/89; Gooday and Turley 1990). However, the effect largely depends on the quality of the settling particles (larger, fast-sinking, fresh aggregates compared to lighter, more degraded ones that have drifted for longer; see Soltwedel 1997). About 3% (in other estimates 5–10%) of the surficial primary production reaches the deep-sea bottom despite the biological degradation that the particles undergo when they sink down through the water column at a speed of about 100 m per day. The response to the addition of phytodetritus is most obvious in the temperate and northern oceans, especially underneath the ice margin, but the reasons for these regional differences are not always clear (Pfannkuche and Thiel 1987; Lambshead and Gooday 1990; Tietjen 1992; Vincx et al. 1994; Gooday 2002; Witte et al. 2003). A lesser food source of the deep-sea is terrigenous detritus, which often accumulates near big river mouths and at the foot of the continental slopes. Remains of dead large animals and plants (“food falls”) provide large food packages that contribute about 10% of the energy input into the deep-sea.
286
8 Meiofauna from Selected Biotopes and Regions
These are particularly used by nematodes, which aggregate despite often high oxygen deficiencies around sunken carcasses (Debenham et al. 2004). The last food source to be mentioned here are the bacterial films that colonize the mucus strands and excretions of the ubiquitous benthic foraminiferans. Bacteria and protozoa react rapidly to the periodic pulses of phytodetritus and fecal pellets and, thereby, can increase their production by an order of magnitude. In contrast, the metazoan meiofauna exhibit a delayed reaction (by some weeks to months). 3. Habitat heterogeneity on a small scale plays an important role in both species richness and diversity of deep-sea meiofauna. Sessile macrofauna protruding from the surface (sponges, coelenterates), worm tubes, agglutinated foraminiferan shells, komokiacean “mud balls” and “manganese nodules” (see Bussau et al. 1995) represent small-scale structures that increase the habitat complexity and also the species richness and functional diversity of meiofauna. These favorable structures persist for long periods in the quiescent deep-sea conditions (Thistle et al. 1993). A structuring effect is also ascribed to the bioturbative activities of megafauna (e.g., holothurians and enteropneusts; see Meadows and Meadows 1994) or to fragments of cold-water corals and sponges that accumulate on the deep-sea bottom (Raes and Vanreusel 2006). Increase in sediment complexity attracts a community of specialized meiobenthic epistrate feeders and contributes to local variations. The ameliorating effect of protruding tubes can be attributed to a complex interaction of favorable factors, such as establishment of hydrodynamically favorable sheltered zones with the accumulation of debris, increased downward transport of solutes, enhanced growth of bacterial stocks and better protection from predators (Thistle and Eckmann 1990, Eckman and Thistle 1991). Bioturbation by macro-epifauna can interact adversely with meiofauna, particularly with harpacticoid copepods (Thistle et al. 2008). Moreover, when macrofauna is experimentally excluded, the concentration of chloroplastic pigments increases and subsequently the meiofaunal densities increased too. Increased structural heterogeneity with strong hydrodynamic variability and the downward transport of large amounts of food material are also thought to be the reason why steep submarine canyons that cut into the continental margin represent “hot spots” of diversity (Ingels and Vanreusel 2006). The geological heterogeneity of the bottom (sheltered and exposed sides of submarine mountains, slopes and faults) may also contribute to variations and increased diversity in the pattern of meiobenthos colonization (Alongi 1990b; Grove et al. 2006). Recently, another geological phenomenon has received particular attention in meiofauna studies: hydrothermal vent areas. Here, an enhanced abundance of the meiobenthos—compared to the non-vent surroundings—results more or less directly from the enormous biomass and production of “sulfur and methane bacteria” (van Harten 1992). On the other hand, the species spectrum is restricted to the few forms that can tolerate the temporarily hostile sulfidic conditions (see Sect. 8.4). On the negative side, collapsing sediments of turbidites represent huge physical disturbances that have long-lasting and highly adverse effects on natural communities (Lambshead et al. 2001).
8.3 The Deep-Sea
8.3.2
287
The Meiofauna
The typical deep-sea meiobenthic organism is highly adapted in biological and ecological terms to the scarcity of food. Favored by the prevailing low temperatures, it grows slowly and has a long life span with low maintenance expenditure. Its metabolically costly reproduction is also affected by the need for energy conservation, resulting in a low number of eggs, often combined with brooding and asexual multiplication (protozoans). Hermaphroditism is frequent and reduces the energetic costs of finding a partner. Sometimes, especially at lower latitudes, reproductive activity in the depth changes in tune with the seasonal supply of surface-derived organic matter. Additionally, the predominant mode of nutrition in the deep-sea, passive deposit feeding, is energetically more favorable than the more active selection of food particles. All of these features characterize deep-sea meiofauna as specialized K-strategists, with a remarkable degree of trophic partitioning and evolutionary diversification. The structuring influence of a scarce food supply is also evident in analyses of size spectra, which have been best studied in the dominant deep-sea taxon, the nematodes. Thiel’s (1975) general rule for deep-sea fauna is valid for various deep-sea regions: with decreasing food supply (chlorophyll content) and mostly parallel with increasing depth, the average body size of nematodes decreases (Soetaert and Heip 1989; Tietjen 1992; Schewe and Soltwedel 2000; Kaariainen and Bett 2006; Rex et al. 2006). Only in the nematode fauna inhabiting the deeper sediment layers with hypoxic or sulfidic conditions has another trend evolved. Here, a long, filiform body size seems to be more favored, perhaps because it is better adapted to the uptake of dissolved organics or to higher mobility in the semi-liquid mud (Jensen et al. 1992a; Soetaert et al. 2002). The trend towards small body size also seems to hold for harpacticoids: in a deep-sea area of the South Pacific, more than 50% of the copepod specimens were less than 200 µm in length (Schriever, personal communication). Gigantism relative to the average representatives of a taxon also occurs in the deep-sea among meiofauna, e.g., in Loriciferans, but it is probably ecologically irrelevant. High specialization and slow biological processes in the deep-sea make meiofauna, especially the rarer species, sensitive to disturbances and much more vulnerable than their relatives in shallow sediments. Processes of recovery from a major disturbance have been found to last for years, a fact that should be considered when planning the economic exploitation of deep-sea bottoms (Ingole et al. 2005), e.g., the mining of manganese nodules (Radziejewska 2002) or the deposition of CO2 (Carman et al. 2004). In this context, the possible role of meiobenthos in the formation of polymetallic (manganese) nodules in the Pacific Ocean should be mentioned briefly. Although the chemical processes involved in the massive accretion of valuable heavy metals are not yet understood, each nodule is densely covered with and colonized in its internal interstices by a diverse meiobenthic community of protozoans and also meiofauna. It may be of relevance that Foraminifera are known to selectively excrete manganese, iron and other metals as xenobiotic particles (xanthosomes). Perhaps these excretions serve as the initial granules for the formation of new nodules, since mineral centers
288
8 Meiofauna from Selected Biotopes and Regions
have been found in all of them (Riemann 1985)? Thus, it is conceivable that meiobenthic organisms influence the growth of these structures of high economic potential in one way or another (Shirayama and Swinbanks 1986). However, the expected largescale exploitation of polymetallic nodules by deep-sea mining would certainly massively threaten the slow-growing deep-sea meiobenthos. Meiofauna community composition. The ubiquitous Foraminifera (Protozoa) play the dominant ecological role in the shelves, the slopes and the plains of abyssal deep-sea bottoms (Fig. 8.3). Usually 50% of all individuals (maximally 90%) and
3m
295 m
> 2000 m
Foraminifera Nematoda other meiofauna Polychaeta (macrofauna) other macrofauna
Fig. 8.3 The relative increase in Foraminifera with ocean depth among the benthos. (After Shirayama and Horikoshi 1989)
8.3 The Deep-Sea
289
about 30% of the meiofaunal biomass consists of foraminiferans (Shirayama and Horikoshi 1989; Moodley et al. 2002). This group alone is as abundant as all remaining metazoan meiofauna. Aside from the foraminiferans, other rhizopods such as Amoebina and the large Xenophyophoria have also been found to richly populate the deep-sea bottoms (Levin 1991). It seems that there is no square centimeter of the deep-sea bottom that is not interwoven with rhizopod pseudopodia. While many predator species among foraminiferans directly affect the metazoan populations by predation, Foraminifera also have a considerable indirect impact on the metazoan meiofauna, consuming about 50% of the incoming phytodetritus. Among the metazoans, nematodes dominate in almost all deep-sea studies, with a share of between 80 and 90%—an even higher value than in shallow reaches. In the muddy abyssal plains and often in the deeper layers below the sediment surface, the meiobenthos is essentially a community consisting primarily of Desmodoridae and Microlaimidae. In a Mediterranean deep-sea canyon, Comesomatidae (with Sabatieria) were the prevailing (up to 40%) nematode family (De Bovée and Labat 1993). Other typical deep-sea nematodes are Acantholaimus, Molgolaimus, Microlaimus, Thalassomonhystera and Halalaimus. Bacterivorous deposit feeders seem to prevail, followed by predators and omnivores, e.g., Sphaerolaimidae, while epistrate feeders usually only represent a low percentage. A 7% share was assigned to gutless nematodes by De Bovée and Labat (1993). Occurring in much lower numbers and particularly in the surficial layers of somewhat coarser sediments are harpacticoids (e.g., Pseudomesochra, Zosime, Malacopsyllus) and polychaetes. Juvenile bivalves and some kinorhynchs are encountered in muddy samples. In studies from the northern Atlantic, polychaetes ranked second after nematodes. The meiofauna of sea mount sediments differs from that of general deep-sea bottoms (see Sect. 7.2.2). Meiofauna (e.g., harpacticoids) even exists in the hadal troughs at depths beyond 10,000 m, although in reduced abundance. Interestingly enough, even representatives of oligochaetes, a group purportedly of limnogenous/terrigenous descent and lacking propagatory stages, have been found at > 7,000 m depth. Nemertines, reported by Ingole et al. (2005) to rank second after nematodes in greater depths of the Indian Ocean, certainly represent a local exception. Diversity. The deep-sea is renowned as a “hot spot of biodiversity.” Based on their high patch dynamics, meiofauna confirm this general rule. The alpha-diversity, “weighted” or “expected species richness” or “Shannon diversity” are unexpectedly high, especially in the bathyal and abyssal depths and around Antarctica. Foraminifera alone can exist at an abundance of 40 different species per cm2. On average 25–50 distinct species of nematodes or harpacticoids can be discriminated per 100 individuals of meiofauna sampled; the equatorial and southern Pacific seems to have a particularly speciose nematode community (Snelgrove and Smith 2002). Based on nematode data, the average taxonomic diversity increases with water depth, although the number of genera may decline approaching deep-water sediments. In samples from the deep-sea bottom under the Arctic ice margin, 300 species of nematodes were discriminated, most of them new to science (Hoste et al. 2007). In a manganese nodule field, an assemblage of 2,022 nematodes consisted of 250 distinct species belonging to 110 genera (Miljutina et al. 2006). Also, Arctic
290
8 Meiofauna from Selected Biotopes and Regions
deep-sea bottoms structured by a rich sponge assemblage exhibited a correspondingly high alpha-diversity (Hasemann and Soltwedel 2006). On a small scale, the diversity of species is extremely high in the deep-sea while comparisons of diversity between large biogeographic regions (gamma-diversity), especially at the genus level, often yield low values. A comparison of nematodes on the generic level between southern and northern abyssal sites revealed minor differences only (Sebastian et al. 2007). Compared to abundance, biodiversity (alpha-diversity) seems to be more intricately influenced by food supply and habitat heterogeneity. Whereas in some studies both diversity and abundance decreased with the depth-related losses in particulate organic matter (POM), many other reports, especially those from bathyal and abyssal sites, measured an inverse relation between POM flux and diversity, i.e., diversity was higher in bathyal and abyssal depths than along the shelf slopes (Rex et al. 1993; Boucher and Lambshead 1995). A comprehensive literature evaluation (Mokievskiy et al. 2007) confirmed, regardless of the methodology applied, an increasing dominance of nematodes, with a maximum at depths of below 1,000 m. Again, this pattern appeared to be controlled by the habitat heterogeneity and the food supply. Only in the hadal regions did both of these parameters decline in parallel. Here a rich meiobenthic life apparently cannot be sustained. The resulting hyperbolic (“hump-shaped”) bathymetric gradient of biodiversity recorded from shallow reaches to the deepest parts of the ocean, with a maximum occurring around 2,000 m, has been corroborated for (nematode) meiofauna in the North Atlantic (Lambshead 1993). A multiple regression re-analysis of numerous nematode data sets using latitude, areas scale, sampling effort and depth as independent parameters (Mokievskiy and Azovsky 2002) also resulted in a hyperbolic diversity curve for depths >100 m, with the highest values obtained at latitudes of between 30 and 60°N. However, the evaluation of another comprehensive data set of nematode genera from numerous sites at depths of between 200 and >8,000 m from a wide spectrum of regions did not show the expected hyperbolic curve of species richness (Gambi et al. 2007). The critical point when looking at diversity patterns seems to be the choice of an appropriate scale that separates local from regional data sets. This would perhaps explain the problematic and partly contradictory results (Lambshead et al. 2000). Why is the deep-sea meiofauna (and macrofauna) so diverse? Depth per se is probably not the causative factor involved per se. The limited food supply promotes niche partitioning, which results in strong functional and morphological specialization and spurs the process of evolutionary diversification. Additonally, structural heterogeneities influence the trophic interrelationships and contribute to biodiversity. This is demonstrated in studies of meiofauna from submarine canyons and from sea mounts. Both of these regions are increasingly recognized as being “hot spots” of meiofaunal diversity. Their variable and complex hydrodynamic patterns create sediment heterogeneity, sometimes combined with a favorably rich flux of organic particles. This scenario differs considerably from the uniform deep-sea plains and enhances meiobenthic diversity. The coral and sponge rubble in cold-water coral reefs enhance the habitat complexity as well and create islets of high diversity in the deep-sea bottom. They are preferred by
8.3 The Deep-Sea
291
robust nematode taxa that mostly act as epistrate feeders on microbial films (Epsilonematidae, Draconematidae). Epsilonema multispiralum is particularly common in North Atlantic deep coral sediments (Raes and Vanreusel 2006).
Box 8.4 Meiofauna in the Deep-sea: Diversity in Scarcity Scarcity of food and low temperatures are the key factors that determine the life of meiobenthos in the deep-sea. The most distinctive aspect is the high biodiversity, which exceeds that at shallow sites. Hydrodynamic patterns and macrofaunal activities structure the bottom and provide heterogeneity; sufficient oxygen and pulses of phytodetritus as food are usually available. This combination results in a deep-sea ecosystem that is far from monotonous, sustaining a meiofauna of particular composition, low abundance and biomass, but high diversity. Foraminiferan protists dominate, and the sediment is interwoven with their pseudopodia. The other main component is nematodes, while harpacticoids remain relatively rare. The number of specimens per species/taxon is extremely low: any random selection of 100 meiofaunal individuals would usually include 25–50 distinct species of nematodes and harpacticoids. What causes such a high diversity? Scarce and patchy food pulses support energy-efficient K-selection in most deep-sea animals. Reduction in average body size reduces consumption and trophic specialization avoids competitive displacement. Moderate fluctuations in the hydrodynamic and oxygen regimes, small disturbance effects in a mosaic of variables, and narrow niches create a factorial complex that maintains subtle ecological disequilibrium processes. Today, these together with the classical time-stability effects are considered the main promoters of deep-sea biodiversity. With electronically controlled instruments, robots and submarines, deep-sea research is providing fascinating contributions to meiobenthic research, and many future discoveries remain to be made in this vast biotope.
Distribution pattern. At a large geographic scale, the horizontal distribution pattern of deep-sea meiobenthos is rather monotonous (see Box 8.1 for an assay of the “latitudinal diversity gradient”). In areas of little sedimentation, the ubiquitous muds harbor a rather evenly distributed association of deposit feeders of high conformity at the genus level. However, changes in sediment structure, e.g., local accumulations of phytodetritus, seem to correspond with a non-random meiofauna distribution and differing composition. The Northeast Atlantic, with its high primary production, harbors a particularly rich deep-sea meiobenthos compared to other large ocean basins and the Mediterranean. In the hydrodynamically more complex sandy areas of the deep-sea, suspension feeders with mucus filtration become more frequent.
292
8 Meiofauna from Selected Biotopes and Regions
Vertical core profiles also demonstrate that deep-sea meiofauna is mainly controlled by the restricted food supply. About 90% of all meiofauna are concentrated in the upper 2–5 cm, where the nutritive detritus accumulates. Especially foraminiferans, harpacticoids and polychaetes aggregate near the surface, while the deeper strata are the domain of nematodes. In contrast to shallow sediments, this concentration at the surface does not result from decreasing oxygen levels at greater depths. As a result of the extremely low input of organic matter, the biological oxygen demand is reduced so much that the upper 5–10 cm of the deep-sea bottom remain oxic. Only areas with richer organic input suffer from low oxygenation, which then becomes an additional key factor. In regions intensively bioturbated by macrobenthos, meiofauna (mainly nematodes) also occur in deeper layers beyond the usual 5–10 cm threshold, since the oxygen penetration is enhanced in this case and the detritus is buried deep down. In the deep-sea of the Indian Ocean, Ingole et al. (2005) reported that only 16% of the meiofauna was present in the upper 2 cm, while meiofauna was recorded down to a sediment depth of 35 cm. The density of deep-sea meiofauna is largely determined by three sediment factors (Shirayama 1984): calcium carbonate content; heterogeneity of the substrate (low sorting coefficient); and organic matter, indicating food availability. Abundances of 100–1,000 meiobenthic organisms per 10 cm2 (without foraminiferans) are quite typical (Table 8.3; Tietjen 1992). At greater depths, meiofaunal density often declines to 10–100 per cm2 (Shimanaga et al. 2007). Sediments with a high abundance of shell remains (calcareous ooze) harbored up to 1,300 ind. ml−1 of metazoan meiofauna (Shirayama 1984). Even Protozoa (Foraminifera) do not markedly exceed this range (maximally 2,000 ind 10 cm−2 or 150–200 ind. ml−1 sediment). Compared to surface values, these figures document the often limiting effect of the organic particle flux on the existence of a deep-sea meiobenthos (Tietjen 1989). Since the amount and the nutritive value of this flux decrease with increasing depth, the meiofauna in most deep-sea regions follow a negative hyperbolic abundance/depth relation (Fig. 8.4; Pfannkuche and Thiel 1987; Tietjen 1992). Only where adverse hydrodynamic patterns dominate shallow sites, will meiofauna abundance and diversity be greater at deep-sea sites. This can cause an exceptional increase in the vertical density curve. Even deeper down, the usual decrease towards hadal depths begins, so that the resulting abundance/depth curve here is parabolic (Rex et al. 2006).
Table 8.3 Meiofaunal composition (%) and abundance in samples from increasingly deep marine sites (Coull et al. 1977) Taxon 400 m 800 m 4000 m Foraminifera Nematoda Harpacticoida Unidentified Polychaeta Organisms per 10 cm2 (average values)
30.8 45.1 10.7 5.1 2.8 442
33.1 59.7 2.4 1.5 1.6 892
65.2 30.2 2.0 1.1 0.5 74
8.3 Deep-Sea
293
Metazoan meiofauna (number x 103 x m−2) 2800
2000
1000
200
0
Depth (m)
1000
5000
9000
Fig. 8.4 The decrease in meiofaunal abundance with ocean depth. (Tietjen 1992)
Population densities of more than 1,000 ind. 10 cm−2 have been recorded in deep-sea areas with a high input of food particles. In the deep-sea of polar regions, especially beneath the ice margins with their rich supply of phytodetritus, meiofaunal abundance can even exceed these values (up to 5,000 ind. 10 cm−2 have been sampled) (Pfannkuche 1992; Soltwedel et al. 2003; Vanhove et al. 1995). Also, in upwelling areas, along the foot of the continental slope or in sites where currents accumulate debris, the richer nutrient supply gives rise to a generally higher meiofauna density (Alongi 1990b). Whereas in temperate and boreal waters phytodetritus from plankton blooms drives seasonal variations in deep-sea meiofauna, in some tropical areas, monsoon-driven fluxes in surface production seem to cause seasonal variations in the deep-sea. In Indian Ocean samples, the deep-sea meiofaunal density at 5,000 m depth was unusually high, about 45 × 103 m−2 (Ingole et al. 2005). Compared with the macrobenthos, the deep-sea micro- and meiofauna are generally more responsive to nutritional changes arising from the surface (Vincx et al. 1994; Gooday 2002). However, this close correlation between food supply (measured as chloroplastic pigments) and meiofaunal abundance was not found in the warm deep-sea of the Sulu Sea, where (corresponding to the conditions at the bottom of the Red Sea) low rates of degradable organic matter and low-oxygen conditions act as stressors (Shimanaga et al. 2007). Biomass and production. Towards the deep-sea, all animal groups experience significant exponential decreases in both abundance and biomass. Smaller size classes replace larger size classes. Since the general decline with depth is particularly rapid for macrofauna, larger animals, their fraction of the total community biomass compared to the fraction represented by the meiofauna shows a steeper
294
8 Meiofauna from Selected Biotopes and Regions
decline towards the deep-sea and favors the meiobenthos. This can result in a biomass ratio between the macro- and meiobenthos of 1:1 (Tietjen 1992). With the benthic protozoans included, the preponderance of the macrobenthos is even more attenuated (see Sect. 9.3.3). The strong relationship between the meiofaunal abundance and biomass and the supply of organic matter, which in turn depends on depth, allowed De Bovée and Labat (1993) to suggest a linear regression with sample depth as the main variable. Meiofaunal biomass from the abyssal plains is often only 50–100 mg fresh wt m−2 (equalling approximately 4–8 mg C m−2). These low values underline the oligotrophic character of wide abyssal regions, although in the areas that receive a high flux of organic matter, values of around 1 g fresh wt × m−2 or more have been recorded. The meiobenthic biomass in Arctic regions was 3–10 times lower than that in richer East Atlantic bottoms (Pfannkuche and Thiel 1987). Canyons and the foot of the continental slopes, with their larger influxes of organic matter from the shelf, are usually more productive. 37 mg C m−2 were measured for the meiofauna of a Mediterranean canyon (De Bovée and Labat 1993). At various Pacific stations, Shirayama (1984) found unusually high biomasses (calculated from ash-free dry weight): metazoan meiofauna was between 103 and 4,120 mg wwt m−2, depending on depth and sediment type (corresponding values for foraminiferans were 130 and 3,414 mg wwt m−2). The few biomass data from the Indian Ocean are not sufficient to provide a consistent picture. In general, productivity rates in deep-sea bottoms are 2–3 orders of magnitude lower than in shallow-water sediments. Shirayama (1995) calculated that a generalized deep-sea meiofauna specimen will ingest about 10 ng C per day, while the same value can be ingested per hour in shallow sediments (Tietjen 1980a). Nevertheless, 80% of all metabolic processes in the deep-sea refer to meiobenthic organisms (Shirayama and Horikoshi 1989). An annual mean consumption rate for the meiobenthos of about 10 g C m−2 has been calculated (De Bovée and Labat 1993). Locally (e.g., in the Mediterranean) this rate can be lower (about 2.9 g C m−2). This demand is largely supplied by seasonal pulses of organic matter from the surface and by horizontal advective currents, sometimes influenced by the tides. The specific role of the meiobenthos in the biological productivity of the deep-sea is still rather difficult to assess, since the problems involved in measuring production (see Sect. 9.3.2) are aggravated in the deep-sea. The following promising methods have been suggested: - Recording community respiration in a limited sediment area under a bell jar (Pfannkuche and Lochte 1990). - Quantitative analysis of the most important nutrient input, the chloroplastic pigments. This parameter, which depends directly on the production of phytoplankton in the euphotic zone, couples the meiobenthos in the deep-sea with the ocean surface (Thiel et al. 1988/89; Fig. 8.5). However, there may be disadvantages to this approach, since the nutritive value (palatability) of phytodetritus can vary (fresh vs. aged detritus: Soltwedel 1997; Witte 2005).
8.3 The Deep-Sea
295
No. x 10 cm−2
2000
1000
0 0
5
10
15
AFDW mg x 10 cm−2
25
CPE (ng x cm−2)
a 1.0
0.5
0 0
b
20
5
10
15
20
25
CPE (ng x cm−2)
Fig. 8.5a–b The correlation between phytodetritus (chloroplastic pigments) and meiofaunal abundance (a) and biomass (b) in the deep-sea. (Pfannkuche 1985)
- Measuring the ATP content as well as the activity of the electron transport system (ETS), both of which reflect the (meio)benthic metabolic rate. The deep-sea is the largest biotope but the one that is least explored biologically. We do not yet have a reliable picture of the composition, diversity, distribution and ecological role of the deep-sea meiobenthos. Why were the composition and the abundance of the deep-sea meiobenthos, with its dominance of protozoans, not recognized earlier? Apart from problems with accessibility and inadequate equipment, the dominant taxon—the benthic foraminiferans with their irregular shells that are covered by agglutinated detritus—has often been overlooked. Considering the global extent of the deep-sea, a representative exploration seems impractical. Large parts of the southern oceans in particular must still be considered “terra incognita” in terms of their meiobenthos. More detailed reading: Thiel et al. (1988/89); Tietjen (1992); Vincx et al. (1994); Gooday (2002a); manganese nodules, Mullineaux 1987; Foraminifera, Gooday et al. 1992; Snelgrove and Smith (2002); latitudinal gradient, Clarke (1992); Rohde (1992); Rosenzweig (1995); Gaston (2000); Lambshead et al. (2000); monograph on the deep-sea, Gage and Tyler (1991).
296
8.4
8 Meiofauna from Selected Biotopes and Regions
Dysoxic, Anoxic, and Sulfidic Environments: Discussing the Thiobios
Life began in an anoxic and sulfidic world. Even in today’s oxic world there are large environments where suboxic (dysoxic) or even anoxic conditions prevail. They originate under the influence of bacterial degradation wherever low oxygen supply meets high organic enrichment. The bacterial consumption of oxygen is often accompanied by the formation of toxic hydrogen sulfide. Hence, hypoxic/anoxic sediments develop naturally and worldwide, although human activities also continuously increase the number and persistence of anoxic/sulfidic sites. Anthropogenic eutrophication and deposition of organic matter accompanied by rising temperatures often create anoxic and sulfidic conditions, as evidenced by the frequent appearance of (black) sediment patches in the summer months. Sheltered hydrographic conditions are most affected, while the oxygen depletion is less severe in places with rich plant growth and intensive bioturbation. As general interest in ecology has increased, the fauna adapted to sulfidic conditions have also attracted more attention. This living system of the sulfide biome (Fenchel and Riedl 1970) was named “thiobios” by Boaden and Platt (1971). The fauna consisted mostly of ciliate protozoans along with some meiobenthic metazoans. Fenchel and Riedl (1970) had realized that there is a distinct, definable community of eukaryotic life that is specifically adapted to reducing conditions. Referring to nematodes, Ott (1972) had also emphasized that “the sulfide system has a homogeneous and stable … fauna of its own right”. Since complete anoxia had not been considered a prerequisite for the existence of a thiobios, it was an unfortunate and ecologically unrealistic approach to link the existence of a thiobios to the absence of oxygen. Reise and Ax (1979) questioned the term “thiobios,” since they could not find “a specific meiofauna confined to oxygen-deficient horizons of the sediment,” regrettably without measuring the microgradients of oxygen or hydrogen sulfide. In the modern understanding of “thiobios,” it seems adequate to base the definition on an etymological root (ϑηιon, theion, Greek for “sulfur”), emphasizing hydrogen sulfide (and/or other reduced substances) as the controlling factor: “Thiobios represents a diverse community of organisms characteristic for biotopes where hydrogen sulfide and other reduced substances are regularly dominating ecofactors. The thiobios is directly or indirectly linked to sulfidic habitats.” The existence of a meiobenthic thiobios is neither questionable nor ecologically irrelevant considering the trend towards more hypoxic/sulfidic environments. Among the thiobios, soft-bottom meiobenthic forms play a substantial role.
8.4.1
Reducing Habitats of the Thiobios
Oxygen depletion is often connected with the development of hydrogen sulfide. The distribution of this important ecofactor is mostly antagonistic to that of oxygen (see Fig. 2.11), and part of a complicated and changing system with intermingled
8.4 Dysoxic, Anoxic, and Sulfidic Environments
297
processes and differentiated gradients of O2 and H2S. The few general features that exist in this dynamic web are (a) ubiquitous transitions between oxic/anoxic and anoxic/sulfidic microenvironments, often in the range of micrometers or even just cell diameters (see Fig. 2.8; Revsbech et al. 1980; Bock et al. 1988; Fenchel 1996), and (b) their continuously changing dynamics based on microbial metabolism (Reichardt 1989). Since these transitions are often suboxic (also termed “dysoxic”), from an ecological perspective they belong to the reducing habitats, and so will be considered here too. Areas where sulfidic biotopes can be encountered are as numerous as the processes associated with them, and include deep-sea hydrothermal vents, gas and oil seeps, and silled marine basins such as the Baltic and the Black Sea. Large suboxic to anoxic and highly sulphidic areas also prevail in upwelling zones off the coasts of Peru and Chile, where organically rich muds are blocked from taking up oxygen by upwelling currents. Oxygen minimum zones (OMZ) are widespread at the continental margins of the oceans (Levin 2003). Basins filled with anoxic brine that has elevated levels of hydrogen sulfide and methane occur in the Mediterranean and in seep areas in the Gulf of Mexico. The subsurface layers of tidal mudflats and mangroves and coastal sites polluted by sewage represent large sulfidic habitats. Stratified deep lakes develop completely anoxic, often sulfidic, layers in their profundal depths. Reduced sediments develop dissolved hydrogen sulfide in two general ways: (a) The deeper layers of organically rich marine shore sediments (tidal flats, mangroves, estuaries, enclosed seas) become oxygen-depleted. Then, through the microbial reduction of oxidized sulfur species (mainly sulfates), sulfide develops. If produced in excess, not all sulfide becomes bound chemically, so that free sulfide will accumulate as toxic hydrogen sulfide. Because of the rich concentrations of sulfate in seawater, marine habitats reach higher (up to millimolar) concentrations of dissolved hydrogen sulfide than limnetic ones, where the sulfide mainly originates from the degradation of the proteins contained in organic matter. (b) Tectonic activities often lead to the venting of anoxic, volcanic, geothermal waters and gases rich in hydrogen sulfide, often combined with rich amounts of methane and ammonium (geothermal reduction). Particularly frequent in the deep-sea, hot smokers or diffuse venting sites characterize these hydrothermal activities (see Sect. 8.4.7). The structures of the ubiquitous reducing/sulfidic areas and their relevance as biotopes for animals can only be understood by considering microbiological and geochemical aspects, which, in turn, leads to a more dynamic understanding of their biota (e.g., Jørgensen and Bak 1991; Watling 1991; Diaz and Rosenberg 1995; Levin 2003). Some important aspects to be considered in this scenario are: - Microbial sulfate reduction, which produces sulfide, is possible under both anoxic and low-oxic conditions (see Jørgensen 1977a; Jørgensen and Bak 1991). - In natural marine sediments, thiosulfate is the predominant sulfur species, and it breaks down to a large extent into sulfide and sulfate (Jørgensen 1990; Fossing and Jørgensen 1990).
298
8 Meiofauna from Selected Biotopes and Regions
- Microchambers (often only 50–200 µm ø) with reduced conditions amid oxic sediment can form a three-dimensional network of oxic and sulfidic microgradients (Jørgensen 1977a; Wilson 1978; Gowing and Silver 1983; Ramsing et al. 1993; Fenchel 1996; Förster 1996). Bioturbation and plant growth complicate their continuous dynamics and alteration (Fenchel 1996; Wetzel et al. 1995; Lee 2003). - The narrow light-colored haloes around tube structures in the sulfidic depths contain oxygen that is mostly in a chemically fixed form; free oxygen is available only temporarily (Jørgensen and Revsbech 1985; Watling 1991; Wetzel et al. 1995; De Beer et al. 2005a). - Oxidized sediment layers can lack free oxygen, but they can still have positive redox values, see Sect. 2.1.4). Oxygen respiration by most animals requires dissolved free oxygen (Sikora and Sikora 1982; Jørgensen and Revsbech 1985; Watling 1991). The following chapter will show that all of these diverse sulfidic biotopes are, at least temporarily, habitats for a characteristic, specialized meiobenthos, and it will challenge their alleged azoic nature.
8.4.2
Thiobiotic Meiobenthos
What is the composition of the typical meiofauna thiobios? Which groups are regularly encountered in the sulfide biome? Foraminifera: In temporarily oxygen-depleted and sulfidic sediments, Foraminifera prevail in abundance. Numerous species have been found in dysoxic or even completely anoxic sediments (Bernhard 1996; Bernhard et al. 2000; Moodley et al. 1997). Their quantitative prevalence under temporarily anoxic conditions has been tested experimentally and is not restricted to the deep-sea (see Sergeeva and Gulin 2007). Especially under conditions of high organic enrichment, they can form high-abundance but low-diversity communities. Ciliata: In many beaches, most of the ciliates are probably migrating between anoxic and oxic strata (Fig. 5.6; see Berninger and Epstein 1995). A rich ciliate fauna can be found mainly around the RPD layer (e.g., Kentrophoros), and many of these thiobiotic ciliates maintain a veritable “kitchen garden” of prokaryotic symbionts (Fenchel and Finlay 1989). Other ciliate species (Metopus, Plagiopyla) harboring methanogenic bacteria as symbionts live as true anoxybionts deep in the black layers (Fenchel et al. 1977; Fenchel and Finlay 1991). They lack normal mitochondria but possess hydrogenosomes (Hackstein et al. 1999). The hydrogen excreted by the ciliates is coupled by methanogenic bacteria to the reduction of CO2 and the resulting methane is released. Platyhelminthes: Among the turbellarians, numerous representatives of the Acoela (e.g., Solenofilomorpha, Oligofilomorpha, Parahaploposthia) and Catenulida (Retronectidae, Paracatenula) are found in the deeper horizons around or underneath the oxic/sulfidic interface (see Figs. 8.7, 8.8; Sterrer and Rieger 1974;
8.4 Dysoxic, Anoxic, and Sulfidic Environments
299
1
2
3
4
ind.x 10 cm3 2000
1500
1000
500
1
June
May
3 April
Feb.
Jan.
Dec.
Nov.
Oct.
Sept.
Aug.
July
June
0
March
2 4
Fig. 8.6 The preferred occurrence of meiofauna (here mainly turbellarians) along the tube walls of biogenic structures. Diagram showing the distribution and annual fluctuation around the tube of the lugworm, Arenicola marina. Numerals in the sediment block stand for subsamples equidistant from the tube into the surrounding sediment. They correspond to those indicated in the lower distributional graph. (After Scherer 1984)
Boaden 1975; Crezée 1976; Powell 1989). Their preference for hypoxia has been known for quite a while, and has been experimentally documented by Meyers et al. (1987, 1988). Scherer (1985) found most thiobiotic turbellarians in the close vicinity of macrofauna burrows, where they probably take advantage of an enriched food supply (Fig. 8.6). Gnathostomulida: While in most meiobenthic groups the thiobiotic species are exceptional specialists, it seems that all of the gnathostomulids prefer low oxic to anoxic or mildly sulfidic biotopes (Müller and Ax 1971). They are regularly encountered along the tube walls of endobenthic burrowers and they have been
300
8 Meiofauna from Selected Biotopes and Regions 0
Ptycholaimellus sp.
1 2 Daptonema fallax Theristus roscoffiensis
Neochromadora trichophora
Viscosia franzii
Rhinema sp.
Sabatieria celtica
Cyartonema sp.
000
Karkinochromadora lorenzeni
425
Monoposthia sp.
0 1 2
25 000 ind. x m−2
10
Paralinhomoeus sp.
Rhabdocoma riemanni
Pomponema sp.
Daptonema sp.
Scraptella tenuicaudata
Prochromadorella paramucrodonta
Nannolaimus fusus
Leptonemella aphanothecae
Sabatieria longispinosa
Spirinia sp.
20 cm
Pomponema tautraense
Odontophora rectangula
10
20 cm
Fig. 8.7 The vertical distributions of various nematodes in sediments of the Kattegat (Baltic Sea). Separate oxibiotic and thiobiotic faunal assemblages are apparent. (After Jensen 1987b)
found confined (and present with considerable diversity) to the reduced fine sand underneath cyanobacterial mats or the roots of surf grass (Westphalen 1993). They even dominate all other meiofauna in the permanently anoxic sediments underneath deep-sea brine seeps (Powell and Bright 1981; Powell et al. 1983). This phylum of primitive Bilateria is a typical component of thiobiotic meiofauna. Gastrotricha: The close relation of some gastrotrich genera to the reduced milieu of the thiobiota is documented by their scientific names: Thiodasys, Turbanella thiophila, T. reducta (Boaden 1974, 1975). Frequently, these and other gastrotrichs can be found in the black layers of sand underneath the chemocline. Urodasys is represented by several species in the anoxic Santa Barbara Basin (Balsamo et al. 2007). Kinoryncha (indet.) have been reported from the permanently anoxic bottom of the Black Sea (Sergeeva 2003). Nematoda: The most abundant animals in suboxic and reduced sediments are particular groups of nematodes (Moodley et al. 1997; Gooday et al. 2000; see Fig. 8.7). Most of them are unusually slender or threadlike (Jensen 1987b), and belong to Siphonolaimidae and Linhomoeidae (Monhysterida), but typical inhabitants of sulfidic biotopes also occur in the enoplid family Oncholaimidae (Pontonema), the chromadorid families Comesomatidae (Sabatieria), Xyalidae (Daptonema),
8.4 Dysoxic, Anoxic, and Sulfidic Environments
301
Macrostomum Turbanella
Solenofilomorpha
Preapha nostoma
Kuma
Parahaploposthia
OXIBIOS
THIOBIOS
normoxic
anoxic, no sulphide
microoxic
low sulphidic
zero-oxygen line
high sulphidic
Fig. 8.8 The occurrence of various oxibiotic and thiobiotic turbellarians in a depth profile of a tidal flat and around a worm burrow. (After Powell 1989)
the Desmodoridae (subfamily Stilbonematinae: Leptonemella, Eubostrichus), and in some Epsilonematidae (Glochinema). Desmodora masira regularly carries bacteria of unknown function in its cuticular furrows (Bernhard et al. 2000). Sabatieria pulchra tends to increase in abundance under hypoxic conditions (Modig and Ólafsson 1998). For faunistic details of the nematodes from the anoxic depths of the Black Sea, see Zajcev et al. (1987), Sergeeva (2003) and Sergeeva and Gulin (2007). Their counterparts in the limnic thiobios are members of the genus Tobrilus and some dorylaimids such as Eudorylaimus andrassy. Corresponding to their taxonomic divergence, they exhibit a wealth of structural peculiarities, of which only a few have been closely investigated: reduction or absence of the gut (Astomonema, Parastomonema, Rhabtothyreus with a “trophosome,” Miljutin et al. 2006); accumulation of large globular granules in the
302
8 Meiofauna from Selected Biotopes and Regions
intestinal cells (Siphonolaimus, Sphaerolaimus, Sabatieria, Terschellingia); inclusion of crystals in the muscle cells (Tobrilus, Sabatieria). It has been assumed by various authors that the furry covers exhibited by the Stilbonematinae, which consist of symbiotic sulfur-oxidizing epibacteria, are related to a thiobiotic lifestyle (Ott et al. 1991, 2004). Oncholaimus campylocercoides from shallow hydrothermal vents has been experimentally found to accumulate globules of polysulfur underneath the cuticle if exposed to hydrogen sulfide. This metabolic capacity is considered rare among aposymbiotic metazoans (Thiermann et al. 2000). Chemoautotrophic bacterial mats on the surface of a freshwater pool in Movile Cave (Romania) are inhabited by several nematode genera, among them the endemic Chronogaster troglodytes. These species have been experimentally shown to be well adapted to the methanic and sulfidic water and the extremely low oxygen concentrations in the mats (Riess et al. 1999; Muschiol and Traunspurger 2007). Oligochaeta: There are two marine genera, Inanidrilus and Olavius (Tubificidae), with numerous species that all are typically thiobiotic (Giere 1981; Giere and Langheld 1987; Erséus 1984, 1990b; see below). Polychaeta: A few adapted species occur in anoxic bottoms of the Black Sea (Nerilla, Protodrilus, Victorniella). In some nerillids, endo- and ectosymbioses with bacteria of unknown function are established (Tzetlin and Saphonov 1995; Bernhard et al. 2000). Crustacea: A few specialists among the normally oxygen-demanding ostracods and harpacticoid copepods (e.g., Cletocamptus confluens) were experimentally found to not only tolerate low oxygen concentrations but also unusually high rates of hydrogen sulfide (e.g., Cyprideis torosa). Among copepods, the Dirivultidae (Siphonostomatoida) dominate around hydrothermal vents (see Sect. 8.4.7). The harpacticoids Ectinosoma melaniceps, Parastenhelia spinosa and some others are regularly encountered in anoxic muds of the Black Sea; here copepod resting eggs were also quite common (Sergeeva 2003). The same study also reports that malacostracan crustaceans, widely held to be sensitive to any oxygen deprivation, occur under complete anoxia in the depths of the Black Sea (tanaids, amphipods). Various crustaceans such as the cephalocarid Lightiella live regularly under lowoxic conditions in marine caves (Schiemer and Ott 2001).
8.4.3
Survival of Thiobios Under Anoxia and Sulphide – Mechanisms and Adaptations
Survival and even persistence for months and years under anoxic/sulfidic conditions have been recorded in various meiofauna, but the underlying physiological pathways are yet to be elucidated. In protozoans, a broad spectrum of Foraminifera are tolerant to complete anoxia, even in combination with the presence of hydrogen sulfide (Bernhard 1996; Bernhard et al. 2000; Bernhard and Sen Gupta 2002; Moodley et al. 1998a,b; 1997; Moodley et al. 2008). A subset of hard-shelled
8.4 Dysoxic, Anoxic, and Sulfidic Environments
303
foraminiferans was viable for three months under anoxic conditions, surviving even longer than the nematodes species in the same sample, which lived for up to two months in complete anoxia (Moodley et al. 1997). The authors therefore suggested a foraminiferan/nematode ratio as a bioindicator of prolonged anoxia. The capacity of many ciliate species to live under suboxic to anoxic conditions formed the basis of Fenchel’s conception of a sulfide biome (see above, Fenchel et al. 1977). Among the metazoa, nematodes almost always prevail in suboxic to sulfidic biotopes (Cook et al. 2000). Some sampling sites in Kenyan mangrove muds were dominated by the bacterial symbiotic Astomonema sp. The nematode Eudorylaimus andrassyi and the tubificid Euilyodrilus heuscheri were retrieved from the permanently anoxic Lake Tiberias (Por and Masry 1968) and subsequently kept for several months in a sealed jar under complete anoxia. Even the symbiotic oligochaete Inanidrilus leukodermatus could survive in a sealed jar with its original sediment and in the presence of H2S (notable for its smell) for five months. Gnathostomulids and nematodes occurred in reduced sediments cut off from oxic seawater by a thick-layered seep of brine (Powell and Bright 1981; Jensen 1986a). Various other species of nematodes, turbellarians and gastrotrichs have been reported from the black depths of various tidal flats. In the oxygen minimum zone (OMZ) off the coast of Peru and Chile the anoxic/suboxic sediment interface is covered with large bacterial mats (mainly Thioploca) and harbors, besides the dominating nematodes, considerable numbers of rotifers, annelids and nemertines (Aramayo et al. 2007). Compared to these rather isolated reports of survival under anoxia, the meiofauna found in the world’s largest anoxic basin, the bottom layer of the Black Sea, undoubtedly proves the existence of a fairly diverse and rich benthos that lives under anoxia and in relatively high methane/sulfide concentrations. Nematodes of the genera Desmoscolex, Tricoma and Cobbionema and some tubificid oligochaetes (?Tubificoides sp.) have been reported from the permanently anoxic depths (> 300 m) of the Black Sea (Zajcev et al. 1987). Confirming these reports, Luth et al. (1999), Sergeeva (2003), and Sergeeva and Gulin (2007) found a meiofauna with an abundance of 50–75 ind 10 cm−2 (dominated by nematodes and foraminiferans) living below the oxycline in permanently anoxic and highly methanic/sulfidic conditions. They also found crustaceans (harpacticoids, tanaids and ampipods), kinorhynchs, and acarids. Even polychaetes (Nerilla, Protodrilus, Victorniella), mites, and juvenile molluscs were encountered. In addition, the muddy sediment contained numerous (> 200 10 cm−2) dormant eggs of copepods and cladocerans. A considerable portion of this peculiar anoxic fauna was not attributable to known taxa, even to phylum (Sergeeva 2003). The sea floor beneath brine basins in the Gulf of Mexico harbored an aberrant meiofauna dominated by gnathostomulids (Powell and Bright 1981; Powell et al. 1983); also in anoxic and sulfidic brine basins of the Mediterranean lived a remarkable meiofauna that was not dominated by nematodes (Lampadariou et al. 2003). As yet, we cannot conceive the physiological pathways that lead to this survival capacity, since most physiological methods are still inappropriate for animals of such minute size. In contrast to macrofauna, the small body diameter of meiofauna
304
8 Meiofauna from Selected Biotopes and Regions
(below 1–2 mm), would allow maintaining an oxygen supply even under low oxygen concentrations by diffusive gradients alone (Powell 1989; Fenchel and Finlay 1995; Fortey et al. 1996). Theoretical considerations and recorded oxygen consumption rates of different meiobenthic groups led Powell (1989) to infer that meiofauna can use an aerobic metabolism at oxygen concentrations as low as 0.1 µmol l−1, which is below the detection levels of many oxygen electrodes. Even when physiologically adapted to low-oxygen conditions, meiofauna must nevertheless be protected from toxic and highly permeable hydrogen sulfide in order to exploit the ecological potential of sulfide-exposed habitats (Giere and Langheld 1987; Ott and Novak 1989; Powell 1989; Schiemer et al.1990; Vopel et al. 1996; Grieshaber and Völkel 1998; Bernhard et al. 2000). The underlying physiological pathways remain largely unexplored. Extreme metabolic specialization of whole communities was demonstrated by studies in the oxygen minimum zone (OMZ) off Chile. Here, sites with very low oxygen concentrations (0.8 ml l−1) harbored even higher meiofauna abundances than more oxygenated sites (Veit-Köhler et al. 2009). The toxicity of H2S is believed to be the main factor that controls the occurrence of thiobiotic meiofauna, since it (reversibly) blocks cytochrome c oxidase and thus the oxygen uptake required for “normal” ATP production. However, the physiology of the lugworm (Arenicola) demonstrates that organisms can develop other energy production pathways, e.g., gaining energy from the oxidation of sulfide using sulfide-insensitive cytochrome complexes (Grieshaber and Völkel 1998). “An ecological compromise between the food requirements of these organisms and their adaptations to the toxic influence of HS” (Sergeeva and Gulin 2007) describes the situation in the Black Sea, but does not reveal the underlying cellular physiology. Ecophysiological experiments by Wieser et al. (1974) not only confirmed the long-term survival of nematodes in anoxic sediment; for Paramonhystera wieseri, they even documented that the presence of oxygen impaired the ecophysiological capacity and viability of this nematode, which exists solely in deep, black sediments. The long persistence and even growth of nematodes under anoxia remains enigmatic, since the formation of cuticular and collagenous material requires oxygen. Jensen (1995) reported that juvenile Theristus anoxybioticus (Nematoda) from sublittoral muds died in oxic water while adults survived well, which corresponds to their field behavior where the long-lived juveniles preferred the deep, anoxic sediments. Schiemer and Duncan (1974) showed experimentally that the nematode Tobrilus gracilis stayed, metabolically, largely anaerobic, even in the presence of oxygen. In experiments performed under anoxia and slightly sulfidic conditions, Metachromadora vivipara even increased in abundance (Steyaert et al. 2007), while other nematode species in these experiments decreased in number and showed reduced activity. Many animals with a high tolerance for anoxia still try to maintain an oxic metabolism at extremely low residual oxygen concentrations (Powell 1989; Gnaiger 1991; Giere et al. 1999). Long-term survival under anoxic conditions was also reported for the turbellarian Parahaploposthia, which was found to be CN- and H2S-insensitive (Fox and Powell 1986, 1987). Thiobiotic animals are usually sluggish animals of low activity. They can considerably reduce their oxygen uptake rates compared to typical aerobic species
8.4 Dysoxic, Anoxic, and Sulfidic Environments
305
(Fox and Powell 1987; Schiemer et al. 1990). Many nematodes and gnathostomulids living below the chemocline can regulate their metabolic levels down to low rates (Schiemer and Ott 2001), and often fall into quiescence when exposed to prolonged anoxia and sulfide concentrations (Vopel et al. 1996). In many species the role of mitochondria seems to be important. In macrobenthic species, the mitochondria have been identified as the site of sulfide oxidation (Powell and Somero 1986; Völkel and Grieshaber 1996). Perhaps are unusually high numbers of these organelles in the tissues of many thiobiotic meiofauna or significant modifications of their structure compared to the normal appearance of adaptive relevance (Duffy and Tyler 1984; Giere et al. 1988a; Jennings and Hick 1990)? Balsamo et al. (2007) speculated that the absence of mitochondria in the sperm of some gastrotrichs might be related to the occurrence of suboxic/anoxic muds in the Santa Barbara Basin. Parallel to conditions in the macrobenthos (Powell and Arp 1989), in some specialized meiobenthos the properties of hemoglobin also seem adapted to scavenging the slightest traces of oxygen (e.g., Colacino and Kraus 1984 for the gastrotrich Neodasys). Boaden (1975, 1977) discussed the role of heme proteins as efficient oxygen scavengers in various red-colored species of Gnathostomulida and Turbellaria; without further explanation Tsurumi et al. (2003) related the presence of hemoglobin in dirivultid copepods of hydrothermal vents to their tolerance of reduced oxygen levels. Unusual organelles possessing sulfide-oxidizing activity (“sulfideoxidizing bodies”) have also been suggested for meiofauna, but have never been documented. Anoxia in nature is mostly accompanied by hydrogen sulfide, one of the most toxic natural agents. However, there are almost no studies that examine the combined effects of these often co-occurring ecofactors. This may be due to the complex instrumentation required to reliably work with defined concentrations of volatile solutions (Visman 1996). Despite some inconsistencies, there is usually a negative synergism for hydrogen sulfide and anoxia. The presence of sulfide triggers a switch to a fermentative metabolism at an earlier phase than would be induced under anoxia alone (Cyprideis torosa, Ostracoda: Jahn et al. 1996; Cletocamptus confluens, Harpacticoida: Vopel et al. 1996). In some typical thiobiotic meiobenthos the activities of oxygen-metabolizing, sulfide-insensitive enzymes were found to be higher than in oxybiotic and macrobenthic fauna, indicating a specialization to cope with the toxicity of oxygen radicals (Morill et al. 1988). It is conceivable that in an environment free of dissolved oxygen, oxidized substances such as nitrate or thiophosphate can become enzymatically activated to serve as oxygen donors. This pathway is known from prokaryotes, from the ciliate Loxodes (Finlay et al. 1983), from representatives of the bacterial symbiotic Stilbonematinae (Nematoda), and gutless interstitial oligochaetes (Hentschel et al 1999; Woyke et al. 2006). An increasing number of meiobenthic species from dysoxic or sulfidic habitats have been found to depend on the symbiotic association with sulfide- or methaneoxidizing bacteria (for reviews see Giere 1996; Ott et al. 2003). A large group of gutless interstitial oligochaetes completely depend on their subcuticular extracellular symbionts and rely on the integrated cooperation of a bacterial consortium, each
306
8 Meiofauna from Selected Biotopes and Regions
with a complicated and highly adapted metabolism (for an overview see Bright and Giere 2005; Dubilier et al. 2006). All of these oligochaete species have incorporated beneath their cuticles a thick layer of sulfide- or methane-oxidizing bacteria which continuously remove the toxic sulfide as they metabolize. This active protection enables the hosts to live in the subsurface layers of shallow sands. Obligate endosymbiosis with bacteria, combined with a reduction of a functional gut, has also developed in nematodes such as Astomonema that occur in sulfide-enriched sediments (Musat et al. 2007), and in Rhaptothyreus (Miljutin et al. 2006). The symbiosis of numerous stilbonematine nematodes (Leptonemella, Catanema) with sulfur-oxidizing ectobacteria, which decoratively ornament their cuticles, enables these worms to live in deeper layers of sand beneath the chemocline (Ott et al. 2003). There are several other meiofauna species within the turbellarians and the polychaetes that live in symbiosis with bacteria, but details about their symbionts’ nature and function are lacking. While further analysis of symbiotic bacteria is required, it should be noted that many thiobiotic animals can survive in extreme sulfidic environments without “bacterial metabolic help.” Beyond certain specific threshold values, they will switch over to an anaerobic metabolism that can sustain them for long periods of time. However, severely sulfidic habitats are avoided, a reaction that corresponds to those of numerous macrobenthic animals (for reviews see Somero et al. 1989; Fisher 1990; Vismann 1991; Bagarinao 1992; Grieshaber and Völkel 1998). The production of sulfur-containing granules or crystals and protection from sulfide by external precipitation (Giere et al. 1988b for Tubificoides benedii) is probably of limited importance, although the quantitative roles of these processes are still to be assessed. Iron has been found in various tissues of sulfide nematodes, and it has been suggested that it binds reduced sulfur (Nuß and Trimkowski 1984; Nicholas et al. 1987; Giere 1992). However, the exportation of the resulting precipitates needs to be demonstrated before this pathway can be verified as a means of detoxification. A more efficient option is the oxidation of sulfides into long-chained polysulfur, which can be stored as an inert but easily activated product in the tissues. This pathway is known from “sulfur bacteria,” but apparently also occurs in a thiobiotic nematode, Oncholaimus campylocercoides (Thiermann et al. 1994, 2000). However, considering the high diffusion rate of hydrogen sulfide through the minute bodies of meiobenthos, the efficiency of any protective method remains doubtful (Powell 1989). A pathway using accumulated carotenoids, common accessory pigments in photosynthesis, as energy-rich substances and oxygen reserves, has been suggested by Zajcev et al. (1987), but has not been demonstrated in physiological detail for invertebrates. Nonetheless, the multiple ecophysiological approaches that have evolved in meiobenthic animals to cope with life under anoxic and sulfidic conditions indicate the complexity of the ecological and physiological processes involved. The sulfidic ecosystem, which is dominated by the presence of reduced substances such as dissolved sulfide, methane and ammonium, is much too complex to allow for simple right-or-wrong opinions that arose in the early debate about the existence of a thiobios. It is the regular exposure to hydrogen sulfide in a spatially
8.4 Dysoxic, Anoxic, and Sulfidic Environments
307
and temporally changing combination of micro-oxic and microsulfidic niches that characterizes the world of the thiobios, and not an either/or situation. With today’s deeper knowledge, we have a better understanding of “sulfide habitats,” their ecological conditions, and their physiological demands on the “sulfide fauna.” Future discussions about the possibility of animal life under anoxic conditions would be substantiated if studies were always accompanied by careful microelectrometric oxygen and sulfide measurements (see Sect. 2.1.4), and where possible combined with physiological analyses.
8.4.4
Food Spectrum of the Thiobios
Areas of steep gradients between anoxic/oxic and sulfidic layers are the preferred habitats of rich bacterial stocks with densities of 109–1010 cells cm−3 (Jørgensen 1977b; Aller and Yingst 1978; Ramsing et al. 1993). The sea bottom beneath the upwelling area off Peru and Chile is covered so densely with mats of the sulfur bacterium Thioploca that the sediment has been termed “Thioploca mud.” It has frequently been pointed out that thiobiotic meiofauna graze on the rich stock of sulfur bacteria (Fenchel 1969; Fenchel et al. 1977; Yingst and Rhoads 1980; Grossmann and Reichardt 1991). This attractivity influences the microdistribution of the thiobiotic meiofauna. For some freshwater nematodes in an isolated thermomineral cave (Romania) bacterial mats provide the habitat and apparently the sole food source (Muschiol and Traunspurger 2007). Animal/bacterial symbioses are another means of utilizing the thiobiotic environment (see above). Many symbiotic animals take advantage of the ability of microbes to metabolize reduced substances such as sulfide, thiosulfate and possibly methane. Studies on the gutless Inanidrilus and Olavius (Oligochaeta), on Kentrophoros (Ciliata) and on Stilbonematinae (Nematoda) indicate that these species use their bacteria as a convenient food source. The mostly used pathway is controlled phagocytosis by the host (Giere and Langheld 1987; Fenchel and Finlay 1989). The stilbonematine nematodes feed by grazing on their symbiotic epibacterial fur (Ott and Novak 1989; Ott et al. 1991). The most complicated symbiosis that has been functionally examined so far is a complex syntrophic web of five different bacterial taxa that cooperate in their host, the oligochaete Olavius algarvensis from the Mediterranean (Woyke et al. 2006). Here, some of its bacteria serve the host as food, but others also remove its wastes by reducing nitrate. The food spectra of the thiobiotic turbellarians Solenofilomorphidae and Kalyptorhynchia as well as those of the specialists among the crustaceans and polychaetes have not yet been ascertained. Dissolved organic matter. Dissolved organics are present in considerable amounts in anoxic sediment horizons (Liebezeit et al. 1983; Sect. 2.2.2). Acetate, an important substrate for sulfate-reducing bacteria (Jørgensen 1977a; Gibson et al. 1989; Michelson et al.1989), is utilized by nematodes (Riemann et al. 1990) and is probably an important organic food source for the thiobios. Boaden (1977) even postulated that absorptive feeding on dissolved organics was a general feature of
308
8 Meiofauna from Selected Biotopes and Regions
the (primeval) thiobios. The thread-like body of many thiobiotic animals with a high length/width ratio and a large surface is believed to support the transepidermal uptake of dissolved organic substances (Boaden and Platt 1971; Giere 1981; Jensen 1986b; Wetzel et al. 1995). Combined with endosymbiosis, this nutritional pathway favors the repeated reduction or transformation of the intestinal tract in the thiobiotic (meio)benthos. The absence of mouth and anus has generated names like Astomonema (Nematoda), Inanidrilus (Oligochaeta) and Astomus (Polychaeta).
8.4.5
Distribution and Succession of the Thiobios
Thiobiotic species have a bacteria-controlled vertical distribution that is different from the usual, characteristic concentrations of oxybiotic meiofauna near the surface (Boaden 1977; Giere et al. 1982; Ott and Novak 1989). Benthic foraminiferans were found at or below the oxic/anoxic chemocline (Bernhard 1996), and ciliates from Scandinavian sites occured over a wide range of depths (Fenchel 1969; see Fig. 5.6). Berninger and Epstein (1995) even contended that most typical beach ciliates live under anoxic conditions, probably migrating up and down. An analysis of the nematode vertical distribution in the Kattegatt (Baltic Sea) revealed an assemblage of oxybiotic species that was clearly separated from the thiobiotic species (Jensen 1987b; Fig. 8.7). Jensen (1981) also demonstrated the presence of a very species-specific vertical pattern within the genus Sabatieria. S. pulchra, a typical thiobiotic species, lives deep in the anoxic sediment, while S. ornata occurs close to the surface. Among limnic nematodes, a similarly differentiated pattern occurs in the genus Tobrilus (Traunspurger 1997a). Powell (1989) summarized experimental studies on various turbellarians with different preference reactions to oxic and sulfidic layers (Fig. 8.8). In normal oxybionts, decreasing the oxygen content and increasing the sulfide content cause avoidance reactions in almost all meiobenthos, first through upward migrations and even emergence from the sediment into the bottom water layers. This avoidance behavior is barely developed in typical thiobios. A preference for layers around the oxic/sulfidic chemocline has been experimentally demonstrated in thiobiotic ciliates, turbellarians, gutless oligochaetes and stilbonematine nematodes (Fig. 8.9). However, this preference is combined with a dynamic positioning: thiobiotic animals typically migrate between oxic and sulfidic horizons, traversing the chemocline each time. They do not maintain a stable position at the chemocline. For the bacteria-symbiotic thiobios it appears that the bacteria recharge their energetically valuable reduced sulfur store while their host stays in the sulfidic layers and subsequently they gain energy from the oxidation of reduced sulfur when their host migrates into the oxic layers (Schiemer et al. 1990; Giere et al. 1991; Ott et al. 1991). Soon after the onset of dysoxic conditions, a marked succession of meiofauna community changes can be observed since the reaction to a hypoxic/sulfidic event differs depending on the meiofaunal taxon. Only when hydrogen sulfide rises to
8.4 Dysoxic, Anoxic, and Sulfidic Environments
309
a 0
2
4
6
8
10
0
b 0
10
20
30
40
50
0
2
4
10
6
20
8
30
Fig. 8.9a–b Migration experiments with thiobiotic oligochaetes (a) and nematodes (b) that prefer the chemocline (dark field in vertical bar). The more diffuse this transitional layer, the wider the migratory radius (a after Giere et al. 1991; b after Ott et al. 1991)
millimolar concentrations will most nematode and foraminiferan species disappear (Moodley et al. 1997). However, during periods of severe oxygen depletion even well adapted thiobiotic species such as Pontonema vulgare or Sabatieria pulchra (Nematoda) show avoidance reactions and are often found on the sediment surface (see Hendelberg and Jensen 1993; Wetzel et al. 2001). Some specialized thiobios can withstand up to millimolar levels of H2S in complete anoxia (Jahn et al. 1996; Vopel et al. 1996). Concentrations of H2S above 5 mM are usually lethal, even to the most tolerant thiobionts (but see Sommer et al. 2003). With the return of oxic conditions, the recolonization of the sediment by meiofauna proceeds rather rapidly, normally within a few weeks depending on the distance and status of the oxic donor assemblage. It is mainly the hydrographic regime that determines processes of recolonization, since drift through the water column is the main pathway for harpacticoids and also for nematodes. Usually the fauna is recruited by colonizers of the ambient oxic fauna, and in its first phases it is characterized by low diversity and high dominance indices (Wetzel et al. 2002).
8.4.6
Diversity and Evolution of the Thiobios
Compared to its oxic counterparts, the thiobiotic meiobenthos seems impoverished, at least in diversity, while abundance and biomass can be rather high. This is particularly evident in shallow sulfidic areas (Neira and Rackemann 1996).
310
8 Meiofauna from Selected Biotopes and Regions
In the polluted Baltic Sea basins and bights, during anoxic and sulfidic periods the meiobenthos is mainly represented by dense masses of the nematode Pontonema vulgare, which feeds on dead macrofauna (Lorenzen et al. 1987). Under similar circumstances, the nematode Terschellingia communis can also develop huge populations. In meiofaunal samples from the oxygen minimum zone off Peru, Neira et al. (2001a,b) found the proportion of nematodes to be extremely high (99%), with a remarkable dominance of one epsilonematid species (Glochinema bathyperuvensis). The authors related the nematode maximum in the OMZ to the high and barely degraded organic content of the sediment. A 10-ml sample of Thioploca mud from the same upwelling area contained about 400 nematodes belonging to more than 20 species of numerous genera (Riemann, pers. comm.). Below the OMZ, where the oxygen content was restored, the abundance of nematodes decreased and that of harpacticoids increased. A change in relative dominance from nematodes to harpacticoids was also recorded by Levin et al. (1991) when sampling from the OMZ down to better oxygenated depths. Also from sediments on the shelf and in canyons with frequently anoxic conditions a prevalence of specialized nematode species that developed considerable abundances was noted (Soetaert et al. 2002). These studies confirm the generally higher sensitivities of harpacticoids compared to nematodes. In oxygen-depleted and sulfidic biotopes a decreasing diversity is observed along with an increasing dominance of single species, and the relative abundances of major taxa change markedly from oxic to suboxic/sulfidic depths (Gooday et al. 2000; Cook et al. 2000). Anoxic/sulfidic events exert a considerable selective pressure. Does this pressure become significant from an evolutionary perspective? Is it possible to characterize thiobios by particular anatomical features (bacterial symbioses, malformations or degenerative trends, see above)? The phylogenetic relevance of the thiobios is debated (see Fenchel and Finlay 1995). A fauna that has existed since archaic times should have considerable evolutionary relevance and should represent primitive life forms. However, suggestions “that the sulfide biome would contain at least some primary (faunal) elements… of the oldest biosystem on Earth” (Fenchel and Riedl 1970) were vehemently refused (Reise and Ax 1979). Today it is widely accepted that the extant thiobios is derived from oxybiotic predecessors. Even the evolution of complex bacterial symbioses in some meiofauna must be considered a relatively recent development, despite the profound anatomical reorganizations involved. This does not refute that the earliest metazoan fauna did indeed evolve under anoxic/sulfidic conditions. The various hypotheses about the environment of archaic metazoans center around the question of whether primitive, meiobenthic metazoan life had already evolved in the Proterozoic. The appearance of oxygen at the lowest concentrations probably predates the Ediacaran period (Jenkins 1991; Mangum 1991; Runnegar 1991; Thomas, 1997). During the late Proterozoic the ocean water may have contained low concentrations of free oxygen. However, at the sediment/water interface, with its rich organic matter, degradation will have caused intense oxygen consumption. Hence, according to modern insights into sedimentary ecology (see Sect. 2.1.4), during that geological
8.4 Dysoxic, Anoxic, and Sulfidic Environments
311
period the sediment strata (the habitat of meiofauna) were probably largely anoxic (Revsbech et al. 1980a,b; Watling 1991; Giere 1992). If subsurface sediment layers in the late Proterozoic world were devoid of oxygen, but the water above the sediment was oxic, why should animals prefer the hostile conditions below the surface and live an endobenthic life? Compared to recent geological periods, organic particles serving as a nutritive basis were rare in the Proterozoic. This rare food would have accumulated at the surface and in the upper sediment layer, probably leading to a richer microbial life compared with the water column. These microbial aggregations plus the shelter from erosion and the elevated UV radiation would have favored the existence of small metazoans at the semisolid, fluffy sediment/water interface. Selective pressure to reduce predation might have been a later incentive to stay below the surface in the upper sediment layers. Only with the advent of bioturbation (early Cambrian) were oxic microniches formed, and additional migrations might have sustained a microaerophilic life. It is probably the parallel existence of adjacent oxic and sulfidic microniches in the sediment, especially in the fluffy surface layers, that enabled mobile microscopic animals to adapt to oxic environments. The small dimensions of the most primitive metazoans (see Fortey et al. 1996), today classified as meiofauna, were an important prerequisite for their utilization of the traces of oxygen present, via diffusion gradients (Powell 1989; Fenchel and Finlay 1995). Fenchel and Riedl (1970), in their initial paper on the thiobios, did not postulate the restriction of thiobiotic animals to complete anoxia. They included in their reasoning for an archaic thiobiotic fauna “the necessary assumption of a low-oxygen atmosphere in the Precambrian age.” Boaden (1977) also considered low oxygen concentrations to be important for the thiobios when he described a “metabolism adapted to very low levels of dissolved oxygen.” However, Boaden (1975, 1989b) also hypothesized an anaerobic, interstitial and holobenthic primitive “thiozoon.” Today, the possibility of the continuous existence of some “lower animal groups” belonging to the meiobenthic thiobios in a “plesiomorphic biotope” (Boaden 1975) is rejected by most authors (cf. Mangum 1991; Fenchel and Finlay 1995), although some of the animal clades that dominate among the thiobiotic fauna are considered primitive (Plathelminthes). On the other hand, Gastrotricha and Nematoda have attained a derived phylogenetic position and are not primitive descendants of an archaic low-oxic or anoxic biotope. Their thiobiotic representatives are instead metabolic specialists among the “normal,” oxic fauna. Hence, it remains speculative to link thiobios with an anaerobic metabolism and an evolutionary origin in the Proterozoic. As demonstrated above, there are many meiofauna that remain viable under anoxic/sulfidic conditions for extended periods of time or even permanently. The fact that aerobic respiration is much more frequent today than anaerobic fermentation reflects the dominance of present-day oxic environments. Among free-living benthos, one could consider strictly anaerobic life and completely aerobic life as extremes in the wide continuum of varying oxygen supply. Further research will discover many more highly interesting meiobenthic thiobios; perhaps more free-living metazoans from different biotopes (not only from the Black Sea and the OMZ
312
8 Meiofauna from Selected Biotopes and Regions
sediments). The ability of free-ranging animals to thrive permanently in anoxia provides a relevance far beyond the meiobenthic scope. However, based on the aforementioned definitions and considering the necessary differentiations, this does not really affect the discussion about the existence of a meiobenthic thiobios. More detailed reading: evolutionary aspects, Fenchel and Riedl 1970; Boaden 1975, 1989b; physiological aspects, Powell 1989; Meyers et al. 1987; Grieshaber and Völkel 1998; ecological aspects, Ott et al. 1991, Giere 1992; geochemical aspects, Jørgensen and Bak 1991; anoxic ciliates, Fenchel and Finlay 1991; Hackstein et al. 1999; reviews, Bryant 1991; Diaz and Rosenberg (1995); Fenchel and Finlay 1995; Wetzel et al. 2001; Wu (2002).
Box 8.5 Thiobios: An Old Debate and Modern Data “Does a thiobios exist?” The debate about this question arose from several discrepancies: ecological field observations vs. geochemical microscale measurements, physiological understanding vs. palaeoclimatic reconstructions, classical taxonomic ordination vs. molecular positioning. Meiofauna of (sheltered) sediments have been known to harbor species typically encountered in the black, H2S-smelling mud layers which traditional methods record as being anoxic. So these species live under sulfide and anoxia; they are “living systems of a sulfide biome,” or “thiobios” (Fenchel and Riedl 1970; Boaden and Platt 1971). Provocatively, it was claimed that some thiobios were primitive representatives of a Proterozoic anoxic fauna. In the light of modern molecular methods this contention cannot be upheld. On the other hand, our traditional ecological picture was also erroneous. Detailed geochemical microrecordings disproved the concept of a rather static two-dimensional layer (oxic conditions at the surface and anoxic/sulfidic below). Today we know of a continuously changing three-dimensional web of oxic and anoxic/sulfidic microhabitats through which the tiny, often elongate, thiobiotic animals move. It is just this alternating exposure to oxic/anoxic spaces that is ecologically favorable and physiologically required. The thiobiotic taxa are adapted in various ways to cope with the challenges and chances of the hypoxic/sulfidic world. Today new fossil records have demonstrated that a meiofauna existed in the Proterozoic sediments. In this period, oxygen in the sediment was minimal or absent and reduced sulfur probably ubiquitous. Hence, a thiobios with a “metabolism adapted to very low levels of dissolved oxygen” was (and still is) present, as postulated by Boaden (1977). Also, the recent discoveries of a meiofauna at chemoautotrophic vents and the anoxic deep Black Sea contribute to a more differentiated and more comprehensive view of the thiobios: “Thiobios represents a diverse community of organisms characteristic of biotopes where hydrogen sulfide and other reduced substances are regularly dominating ecofactors. The thiobios is directly or indirectly linked to sulfidic habitats.”
8.4 Dysoxic, Anoxic, and Sulfidic Environments
8.4.7
313
Chemoautotrophy-Based Ecosystems: Vents, Seeps, and Other Exotic Habitats
Life at the globally famous hot vents or seeps is based on chemoautotrophic processes and not on photoautotrophy. These unique ecosystems are characterized by a spectacular macrofauna. Mud volcanoes, oil seeps or decaying organic masses (large carcasses, sunken wood) also belong, from an ecological perspective, to these biotopes. What about their meiofauna? In contrast to the spectacular macrofauna observed at chemoautotrophic sites, their meiofauna are less exceptional and striking. Therefore, meiofauna have only recently been included in studies of chemoautotrophic ecosystems and detailed knowledge about them is still rare. Characteristic of an increased awareness is the “Meiovent” initiative (Bright and collaborators, Vienna). Meiofauna occur either in the thickets of mussel and tube worm colonies (Alvinella, Riftia, Bathymodiolus) or at the few sites where soft sediments accumulate. The higher the structural complexity of the epifauna thickets, the richer the meiobenthic life (Tsurumi and Tunnicliffe 2003): tightly interwoven tubes are more densely populated than bush-like structures; in mussel beds with young Bathymodiolus (small shells), copepods were more abundant than nematodes, but these prevailed in the older beds (Copley et al. 2007). Vents are characterized by irregular pulses of sulfide- and often methane-rich water alternating with influxes of ambient oxygen-rich seawater. Hence, the ecophysiological potential of the vent meiofauna is comparable to that of the typical thiobios in oxygen-deficient environments outlined above. Because of the accumulated detritus, meiofaunal abundances near vents can be elevated when compared to neighboring sediments. In the methane-dominated center of the Håkon Mosby mud-volcano, a mean density of 220 copepods per 10 cm2 (Tisbe sp.) is indicative of a rich but monotonous meiofauna (Van Gaever et al. 2006); cold seeps in Japanese waters harbored a total of around 400 meiobenthic individuals per 10 cm2 (Shirayama and Ohta 1990). Sediments from methane seeps off the Oregon coast (USA) often contained 1,000 or more individuals per 10 cm2 (Sommer et al. 2007). At one sampling site in the East Pacific Rise (EPR) up to 950 ind. 10 cm−2 were counted (Gollner et al. 2007). However, densities around 10,000 nematodes and more per 10 cm2 belonging to only a few species (Olu et al. 1997; Sommer et al. 2003; Van Gaever et al. 2006) may be exceptional. These figures make these deep-sea sites “chemosynthetic oases” for meiofauna (Soltwedel et al. 2005). However, at a hot-vent effluent in the South Atlantic, some samples (coarse sand) contained no meiofauna whatsoever (unpublished data)—a unique experience, without a readily apparent explanation (toxic composition of the effluents?). Surprisingly low densities and species richness (10–200 individuals per 10 cm2 belonging to 30 copepod species and 4 nematode species) were also recorded in evaluations of the hot-vent meiofauna among macrobenthic epigrowth from thickets of the tube worm Riftia in the East Pacific Rise (EPR), as well as from cold (oil) seeps in the Gulf of Mexico (Zekely et al. 2006a,b; Gollner et al. 2006, 2007). Also, the first compilation of hydrothermal meiofauna by Dinet et al. (1988) from the Guaymas mud seeps and West Pacific vents has reported relatively low mean densities.
314
8 Meiofauna from Selected Biotopes and Regions
What are the compositions of vent and seep meiofauna assemblages? Worldwide, more than 80 meiobenthic species have been identified at vents so far: Pacific samples average 24 species; Atlantic samples 15 species. Nematodes usually represent the bulk of the meiofauna in chemoautotrophy-driven habitats, around deep-sea and shallow hot vents or in seeps (Kamenev et al. 1993; Buck and Barry 1998; Debenham et al. 2004; Flint et al. 2006; Sergeeva and Gulin 2007; Copley et al. 2007). Since many species belong to separate genera, often in a 1:1 ratio (Vanreusel et al. 1997; Zekely et al. 2006a), the generic diversity is rather high. In Mid-Atlantic Ridge sites nematodes prevailed (63%), and in various macrofauna substrates of the East Pacific Rise (EPR) the nematode Thalassomonhystera dominated, accompanied by other monhysterids and a few other families (Draconematidae, Cythalaimidae, Leptolaimidae, Microlaimidae, and Desmodoridae) (Flint et al. 2006; Zekely et al. 2006a,b). The monhysterid nematode Halomonhystera disjuncta was the only abundant species around a deep-sea mud volcano (Van Gaever et al. 2006). Shallow vent sites mainly harbored species of the common genera Oncholaimus, Sabatieria Desmodora and Chromadorina (Thiermann et al. 1997; Van Gaeuer et al. 2006, 2008; Zepilli and Danovaro 2007). The bacterial symbiotic stilbonematines, common around shallow gaseohydrothermal vents (Kamenev et al. 1993), seem to be lacking at deep-sea vents. The second most common taxon among many hydrothermal meiofauna is copepods. Their diverse taxa represent about 15% of the total vent fauna, with most of them belonging to the siphonostomatoid families Dirivultidae and Ecbathyriontidae, while harpacticoids are of a subordinate rank. A good example of the dominance of dirivultid copepods are the meiofauna in hydrothermally active areas on the East Pacific Rise, with Aphotopontius and Stygiopontius being the most typical genera endemic to hydrothermal vents (Heptner and Ivanenko 2002a,b; Martinez Arbizu et al. 2006; Gollner et al. 2006; Zekely et al. 2006a). Among the subdominant or rare meiofauna taxa are polychaetes (often dorvilleids), turbellarians, halacarid mites, ostracods, solenogastres, and recently kinorhynchs and gastrotrichs (Van Harten 1992; Scheltema 2000; Flint et al. 2006; Katz et al. 2006; Sergeeva and Gulin 2007; Copley et al. 2007). The roles of hydrothermal foraminiferans and ciliates, which are certainly of relevance to the chemoautotrophic biota, have not yet been duly assessed. At Japanese vents, 13% of the hydrothermal fauna consisted of Foraminifera (Shirayama and Ohta 1990). In methane seeps of the Black Sea, other taxa, including polychaetes, hydroid polyps and turbellarians, have been found under anoxic conditions. At oil seeps, some non-specialized harpacticoids with a wide occurrence were encountered. Large food falls in the ocean or thermomineral freshwater caves with high concentrations of methane and sulfide represent special cases of chemoautotrophic environments. Specialized nematodes also prevail in these unusual biotopes (Debenham et al. 2004). The isolated thermomineral Movile Cave (Romania) contains a simple meiofaunal food web. The bacterial mats are grazed upon by various ciliates and five nematode species, among them the endemic Chronogaster troglodytes. At least in experiments, the populations of Chronogaster were devoured and controlled by the copepod Eucyclops subterraneus (Muschiol et al. 2008a). Different meiofauna taxa have been found in other chemosynthetic habitats:
8.4 Dysoxic, Anoxic, and Sulfidic Environments
315
sediments around methane hydrates with sulfide concentrations of up to 17 mmol harbored, besides the usual nematodes, two species of Lecane (Rotifera). As sulfide concentrations increased their populations became richer, such that they formed the dominant taxon (Sommer et al. 2003; Sommer et al. 2007). The adaptations that enable these specialists to survive under the toxic conditions of chemoautotrophic sites are not understood. They have this in common with the typical thiobiotic meiofauna (see above). Ovoviviparous reproduction, as found in Geomonhystera (Nematoda), might help to ensure the survival of the offspring (Van Gaever et al. 2006). Detoxification of hydrogen sulfide by oxidation into inert polysulfur might be another pathway that is realized in the dominant nematode around shallow gaseothermal fields in the Mediterranean Sea (Thiermann et al. 1994, 2000). Elemental sulfur was also shown to exist in thiobiotic turbellarians (Powell et al. 1980). Symbiosis with sulfide- or methane-oxidizing bacteria as a means of detoxification (as often found in other reducing environments) is rarely documented for meiofauna at vents and seeps. The significance of the biased sex ratios (clearly more females than males) found for both vent nematodes and copepods (Tsurumi et al. 2003; Gollner et al. 2006; Zekely et al. 2006a) is unclear; it is a phenomenon that is also common in “normal” deep-sea environments. A prevalence of slender filiform nematode species, interpreted as an adaptation enabling the effective uptake of dissolved organics in thiobiotic biotopes (see above), is not always seen at vent sites (Buck and Barry 1998). The significance of numerous dormant benthic eggs of copepods and cladocerans and also copepod nauplii in the permanently anoxic Black Sea mud remains unexplained. The absence of adults suggests long migrations (Sergeeva and Gulin 2007). The food web structure in hydrothermal and seep sites is relatively simple— another indication that this is an extreme habitat. Most of the dense meiofauna associated with chemoautotrophic sites are primary consumers that depend on the copious food supply from bacterial deposits condensed as mats or biofilms, resulting in a positive correlation between the abundances of the vent meiofauna and the (bacterial) debris (Limén et al. 2007). Hence, nematodes with weakly cuticularized buccal cavities, typical of deposit feeders (see Sect. 5.6.1), prevail (Dinet et al. 1988). Also, stable carbon isotope analyses suggest that bacteria provide a chemosynthetically derived food source (Van Gaever et al. 2006). Moreover, the considerably larger body sizes of hydrothermal nematodes (Vanreusel et al. 1997) may reflect an abundance of food compared to the normally food-depleted ambient deep-sea bottom. Because this trend is not generally observed at all sites (Buck and Barry 1998) its significance remains uncertain. The dirivultid copepods, frequent among mussel beds, crawl over the substrate and mostly graze on the rich bacterial films. They are also found associated with the tubes of polychaetes; some species of the genus Ceuthocetes may exist as parasites (Gollner et al. 2006), while some occur in the gill chambers of shrimp (Dinet et al. 1988; Heptner and Ivanenko 2002). Chemoautotrophic biotopes show highly variable fluctuations that are often associated with extreme environmental parameters. Judging from the macrobenthos, this should favor specific adaptations, resulting in an independent vent or seep community that is different from the neighboring “normal” fauna and rather isolated
316
8 Meiofauna from Selected Biotopes and Regions
in its distribution. The entirely benthic larval stages of dirivultid copepods found by Gollner et al. (2006) would suggest distributional restriction (but in other studies vent nauplii were caught in the plankton above the vent; Ivanenko 1998). Comparisons of the meiobenthos in different regions emphasize its high endemism and independence from the ambient fauna. If analyzed at the species level, we find a patchy community controlled by local conditions with low correspondence to neighboring sites (Fricke et al. 1989; Shirayama and Ohta 1990). In a comparison of Pacific and Atlantic deep-sea mussel beds, the overlap between meiofaunal species was zero (Zekely et al. 2006a). This emphasizes the local and patchy character of most vent meiofauna, and would also explain why comparisons of hydrothermal vent and seep fauna with other sulfide/suboxic habitats display a low affinity. However, this isolation only refers to the species level (Shirayama and Ohta 1990; Vanreusel et al. 1997, Flint et al. 2006; Zekely et al. 2006b). The compositions on the genus level are often similar to those in the surrounding normoxic environment rather than to those of other disjunct chemoautotrophic sites. The reported absence of predatory meiofauna in hydrothermal vent sites would be surprising in habitats that are rich in deposit feeders, and so needs further assessment. From corresponding photoautotrophic ecosystems with rich bacterial stocks and organic deposits we know the presence of a variety of carnivorous species. Summarizing, hydrothermal vents and seeps with their copious supply of bacterial food can be perceived as extreme habitats harboring a specialized thiobiotic meiofauna with typically local characteristics. A high density with a low species richness would characterize the extreme nature of these biotopes. The low density and diversity of meiofauna at Pacific hydrothermal vents and seeps, as found in some reports, is perhaps not a general feature. Occasional aggregations of almost 1,000 ind. 10 cm−2 (Sommer et al. 2003; Zekely et al. 2006b) suggest a population patchiness that is typical of extreme habitats. The meiofaunal results from hot vents in the eastern Pacific published so far seem to support a gradient pattern from the hot and toxic smoker walls (Alvinella community) with hardly any sediment to the diffuse venting sites (Bathymodiolus community), where detritus accumulates on the shells and in the crevices. Meiofauna among the tubes of alvinellids represent a highly specialized association of low abundance and high dominance of some dirivultid copepods. Towards more diffuse venting fields with “milder” conditions and more available sediment, nematodes gain in importance and the communities become more diverse and speciose. The small-scale spatial pattern of meiofauna distribution at vents is probably shaped by the varying hydrothermal fluxes and needs further scrutiny. While it appears that the meiofauna at vents along the East Pacific Rise is poorer in terms of both biodiversity and abundance compared to the “non-vent” deep-sea surroundings, the abundance of meiofauna from other vent regions seems to support vents as favourable habitats. Anyway, with respect to meiofauna, vents cannot be considered “oases in the depth” (Dinet et al. 1988; Shirayama and Ohta 1990; van Gaever 2006; Vanreusel et al. 1997; Olu et al. 1997; Soltwedel et al. 2005). More detailed reading: Mokievsky and Kamenskaya (2002); Levin (2005); Van Gaever et al. (2006b); Zekely et al. (2006a).
8.5 Phytal Habitats and Hard Substrates
8.5
317
Phytal Habitats and Hard Substrates
Remane (1933, 1940) characterized the phytal as a habitat populated by an abundant and diverse faunal community, which he termed the “phyton.” Wieser (1959b), in his comprehensive study of the phytal meiofauna from various shores, noted a relationship between the size, body structure and locomotion type of the inhabitants, the shape of the algae and the biotopical differentiation of the phytal. For example, on foliose, shrub-like algae of exposed sites, >50% of the often flattened harpacticoids have clawed clinging legs adapted for climbing (see below), while in tufted, fine-filamentous algae only 10% of the harpacticoids are armed with claws. The main reason why algal belts on hard bottoms are important habitats for vagile benthic animals is their structural complexity. This is modified by a set of (often local) abiotic and biotic conditions, such as water depth or exposure. This general conclusion about the phytal meiofauna emerged as early as the fundamental studies by Remane (1940) and Wieser (1959). In exposed phytal zones, the usual numerical dominance of nematodes and their productive role is reduced. Hence, unlike other meiobenthic habitats, the phytal is often characterized by a high percentage of harpacticoid copepods and ostracods. Compared with their low abundance in soft sediments (usually less than 10%), harpacticoids in phytal habitats regularly comprise more than 30% and often more than half of the total meiofaunal abundance and production (e.g. Danovaro et al. 2002; Hopper and Davenport 2006). A high percentage of copepods is also recorded from the blades of seagrass meadows, in strata under dense algal cover, and specifically on the fronds of algal thalli. In both tropical and boreal phytal areas, copepods seem to be the prevalent meiofaunal inhabitants (Wieser 1959; Jarvis and Seed 1996; Danovaro 1996; Danovaro and Fraschetti 2002), primarily represented by Tegastidae, Tisbidae (Tisbe, Scutellidium), Peltidiidae and Porcelliidae. Only in the deeper, hardly exposed phytal are the climbing meiofauna replaced by wriggling meiofauna (nematodes). The closer the phytal is to soft sediments and detritus, the higher the percentage of nematodes (very often oncholaimids). In addition to copepods and selected species of nematodes, other widespread members of the phytal meiofauna are ostracods, especially Xestoleberis spp., Paradoxostoma sp. and Loxoconcha sp. (Hull 1997; Frame et al. 2008) and halacarid mites, especially the plant suctorial Rhombognathinae (Pugh and King 1985b). These taxa are adapted to climbing in the thickets of densely branching algae. Commonly encountered taxa also include turbellarians, syllid polychaetes and tanaidaceans (peracarid crustaceans) (Arroyo et al. 2004). Which adaptive features characterize members of the phytal meiofauna? As we saw above, the fact that the plants are exposed to waves and tidal currents necessitates that the animals have highly developed attachment capabilities (Fig. 8.10): flattening of the body, development of “haptic” organs such as sucker-like structures, adhesion by mucus secretion. Most of the inhabitants of the upper strata are climbers. For this locomotory type, long, prehensile grasping legs and hairy spines are of high adaptive value (e.g., Porcellidium, Ectinosoma and Thalestris in harpacticoids, Paradoxostomatidae, Xestoleberidae, Bairdiidae in ostracods, Rhombognatidae in Halacaroidea).
318
8 Meiofauna from Selected Biotopes and Regions
a
b Fig. 8.10a–b Structural adaptations of harpacticoids to a phytal environment. a The flattened harpacticoid Porcellidium in lateral view, and its “sucker disk” (mouth parts) in ventral view. b Some grasping legs from various phytal harpacticoids. (Combined from Tiemann 1975; Hicks 1985)
Exposure and tidal stress are the main factors influencing the composition and microdistribution of the littoral phytal meiofauna (Hopper and Davenport 2006). Turbellarians were found most commonly on the less exposed underside and inner parts of the thalli (Boaden 1996), where retained water better prevents desiccation. The harpacticoid Porcellidium (see Fig. 8.10a) avoids tidal stress by vertical migrations in the algal canopy (Gibbons 1991). Algal branches and thalli and seagrass plants offer numerous microhabitats, mitigate harsh hydrodynamic forces, and provide shelter from predators (Coull and Wells 1983; Webb and Parsons 1991; Muralikrishnamurti 1993). The thalli/blades also accumulate sediment and detritus, which adhere to the exudations and the biofilms on the plants. These sediments seem to increase the overall structural complexity and favor meiofauna colonization. Some colonizers have been found in both the organic layer covering the surrounding rocky substrate and in the detrital film on the surfaces of the fronds. In densely branching, turf-forming, tufted or fine filamentous algae meiofaunal abundance and diversity is enhanced; these microhabitats become more rapidly colonized than the foliose, blade-like thalli (Gibbons 1991; Frame et al. 2008). On the other hand, if accumulations of fine deposits become too rich, the habitable area and structural complexity are reduced, yielding
8.5 Phytal Habitats and Hard Substrates
319
a parallel decrease in phytal meiofaunal abundance and diversity. Less complex, shrub-like algae that do not retain the sediment and often grow in more exposed areas are populated by a less abundant and less diverse meiofauna. The epigrowth of small, often ephemeral, filamentous algae on the thalli or well-developed biofilms particularly enhance the species richness and density of meiofauna (Hall and Bell 1993; Peachey and Bell 1997). Along boreal shores, there are three general macroalgal types, which are distinguished by their decreasing structural complexity: the Laminaria–Delesseria zone, the Fucus zone, and the Zostera zone Remane 1940. Correspondingly, in warmwater areas the delicately branched Cladophora, Corallina and Delesseria harbored a much richer “meiophyton” than Laminaria, Fucus and seagrass blades (Hall and Bell 1988). The complex pelagic Sargassum rafts are also regularly inhabited by a rich meiobenthic fauna. Even the smooth laminose thalli of the green alga Caulerpa taxifolia, an invasive neophyte in the Mediterranean, are colonized by a rich and diverse community of meiofauna. Here, epiphytic macrofauna are yet to be found (Travizi and Zavodnik 2004). Macroalgae (e.g., kelp) provide three subhabitats for meiofauna (Hicks 1985): the surfaces of the thallus fronds, the interstices of holdfasts (rhizoids), and the deposited sediment and detritus that accumulates at the bases of the stems. Many phytal meiofauna differentiate between these microhabitats. Algal fronds are dominated by harpacticoid copepods, especially by rarer taxa, while the holdfasts harbor the highest abundance of more eurytopic meiofauna, primarily nematodes (Arroyo et al. 2004, 2007). The meiofauna of the algal holdfasts and the stems of seagrasses (see below), which are often rich in sediment and detritus, are frequently recruited from local softbottom sediments and are not tightly linked to the phytal fauna proper. This lowest phytal stratum seems to represent an ecotone that combines the structural complexity of the phytal with the rich food supply of the sediment below, which would explain the high meiofaunal densities and the similarity to the bottom fauna. Experiments with artificial substrates (standardized bottle brushes, blade mimics) confirmed the positive relation between structural complexity and meiofaunal density and between reduced water flow and meiofaunal colonization (Gibbons 1991, Attila et al. 2005; Mirto and Danovaro 2004). These authors conclude that structural density controls the species richness of copepods and nematodes, whereas the surface area of the plants controls the abundance of copepods. Apparently, the “realizable niche” of the phytal, which structures meiofaunal colonization and determines its “value,” depends on a multiple factorial complex, the nature and position of phytal elements, their protective potential, hydrodynamic exposure, surface area and amount of surface with sediment cover with (Attila et al. 2005; Hopper and Davenport 2006). The structural complexity can be numerically described and compared by plotting outlines of the phytal fronds and calculating the fractal dimensions. Multiple fractal scaling seems an appropriate approach to describe the complex physical environment of phytal meiofauna (Gee and Warwick 1994). Aside from spatial differentiation into various strata, meiofauna also display a structured temporal occurrence. A number of phytal meiofaunal species are known to periodically leave their substrate for (nightly) excursions into the overlying water
320
8 Meiofauna from Selected Biotopes and Regions
column and can subsequently be caught in suspension traps (Kurdziel and Bell 1992). In the Black Sea phytal, Kolesnikova et al. (1995) found a differentiated diurnal behavior that varied depending on the meiofaunal taxon. While during the daytime harpacticoid copepods and nematodes both settled on macrophyte thalli, they separated at night, with harpacticoids ascending into the water column and nematodes heading for the bottom sediment. The authors interpret these migrations as avoidance reactions to fish predation. Because of the close structural and trophic ties of phytal meiofauna to the plants and their surface films, there is often a marked seasonal variation in the population dynamics of phytal meiofauna that depends on the cycles of growth and decay of the algal stocks. Even in the tropics, marked temporal variations were noted (Arlt 1993, Faubel 1984; Jarvis and Seed 1996). This seasonality becomes particularly apparent in seagrass beds with their decaying external blades in the winter (Novak 1992; Danovaro 1996). Growing in soft bottoms rich in detritus, seagrass beds usually develop in little-exposed subtidal zones with fairly stable conditions. They differ in their structural characteristics from algaecovered hard bottoms, i.e., the shape and the surface of the smooth sea grass blades are structurally rather simple. Here, the relevance of the biofilms, adhering detrital particles, and tiny algal epibionts on the blades becomes apparent. This epigrowth provides a microstructure of high protective and nutritive value, and has been found to control the density and diversity of phytal meiofauna (Hall and Bell 1993; Peachey and Bell 1997). Moreover, the fairly tall and rugged seagrass stems offer a wealth of microhabitats, especially when growing in dense stands. It is this density of plant patches per area that has been found to influence the meiofaunal abundance. In general, seagrass meiofauna are particularly rich in nematodes (often monhysterids and oncholaimids) and, hence, are more similar to the inhabitants of the ambient soft bottom than to those of algal belts. In Mediterranean Posidonia beds, the structurally complex and meiofaunally rich “stem stratum” has been distinguished from the more monotonous “upper leaf stratum” and “lower leaf stratum” (Novak 1989). The ease with which seagrass can be mimicked by artefacts allowed numerous manipulative studies of the impacts of structural complexity, biological aging of surfaces, protective effects and colonization potential on phytal meiofauna (Bell and Hicks 1991; Edgar 1999). In many cases the physical components are modified by the (micro)biological coating and the sedimentation of natural debris on the pristine artefacts. These components affect the rate, density and variety of meiofaunal colonization. Newly settled meiofauna came from the ambient sediment as well as from surrounding plants, thus demonstrating the high vagility of the phytal meiofauna (Bell and Hicks 1991). Comparable results have been reported for the meiofaunal colonization of artificial mangrove pneumatophors (see below; Gwyther and Fairweather 2002, 2005). One biotope typical of American tidal flats are the extensive Spartina salt marshes. The root system, the culms and the leaves of the plants provide the basis for a richly structured, well-protected habitat (Bell et al. 1978; Osenga and Coull 1983). With its abundant supply of organic matter, Spartina marshes harbor rich and genuinely adapted meiofaunal populations of ecological importance (Wieser and Kanwisher 1961; Rutledge and Fleeger 1993). In a complex, mutually regulating
8.5 Phytal Habitats and Hard Substrates
321
system, rich stocks of insect larvae, juvenile fish and shrimp seem to exert a severe predation pressure, acting as a top-down control for copepods in particular, but also for nematodes (Feller and Coull 1995). A peculiar subhabitat in salt marshes of the southern United States is the rich system of “gas passages” (aerenchyma) in the stems of Spartina alterniflora (Healy 1994; Walters et al. 1996). These are regularly inhabited by an amazingly rich and diverse meiofauna characterized mostly by specialized oligochaetes (Enchytraeidae) and epsilonematid nematodes. Together with the meiofauna found under the Spartina leaf sheaths, the inclusion of this stem biotope increased the overall abundance of salt marsh meiofauna by an order of magnitude! Another richly structured habitat of a particular character is provided by mangroves (see Sect. 8.2.1). The branched roots and the numerous pneumatophores, with their high degree of complexity, allow for comparisons with classical phytal habitats. Experiments with artificial mimics have demonstrated the relevance of natural surfaces, with their specific biofilms and mature epigrowth, for meiofaunal colonization, so that the degree of convergence and intrinsic patchiness remained rather low (Gwyther and Fairweather 2002, 2005). In their “taxo-ecological” structure, phytal habitats from tropical and south temperate shores correspond to those from temperate shores (Hicks 1985; Arlt 1993; Muralikrishnamurty 1993; De Troch et al. 2001). This underlines, despite climatic divergences, that comparable algal complexity and structural impact create meiobenthic “isocommunities” in geographically disjunct zones. An example of this is the dominance of enoploid nematodes in both the North Sea and Chilean phytal habitats. Also, many oncholaimid nematodes are trans-regional, characteristically inhabiting the phytal, and epsilonematids populate the stems of seagrasses in many areas. A similar trans-regional parallelism can be found in families of harpacticoids, ostracods and halacarid mites. The existence of isocommunities with far-reaching similarities in their fauna separated by continental distances with far-reaching similarities in their fauna (compare Wieser 1959; Arroyo et al. 2004) probably relates to the close adaptive parallels of the fauna to the phytal structure. This may also explain the similarity in phytal fauna between areas of varying exposure and water depth. Only a few members of the phytal meiofauna feed directly on their substratum, the plant’s tissue, by piercing the cells and sucking their cytoplasm. Some nematodes (Halenchus), tardigrades (Echiniscus), halacarids (Rhombognathidae), siphonostomatid cyclopoids and ostracods (some Xestoleberidae, Paradoxostomatidae) have adopted this specialized mode of living. Some phytal nematodes, harpacticoids and ostracods grasp and crack diatoms on the plants with specialized mouth parts. Other specialized phytal meiofauna may take up exudates secreted by the plants. However, the bulk of the meiofauna encountered in the phytal live on detritus and microorganisms that have accumulated on the plants. This organic film, and not the plant itself, is the grazing ground for the meiofauna and the basis of the phytal food chain. An example is decaying frond ends, which are densely populated by bacteria, and so attract a meiofaunal assemblage that grazes on the microbes. In seagrass beds, nematodes have been found to specialize on the brown, decaying parts of the blades (Moens and Vincx 2000a).
322
8 Meiofauna from Selected Biotopes and Regions
The other trophic line in the phytal is, in fact, detritus-based. Lower current velocity within algal thickets causes an accumulation of plant debris and fine sediment. Experimental work with artificial substrates has shown a better correlation of meiofaunal colonization with debris accumulation on the plants rather than with an enhancement of the structural complexity (Edgar 1999). In seagrass beds, experimental results have also suggested a close relationship between meiofaunal settlement (harpacticoids) and the deposited detritus layer (Meyer and Bell 1989). It is largely the detrital food accumulating on the surface that controls the meiofauna and correlates with their density and diversity (Hall and Bell 1993). Dense, tufted filamentous algae retain more detritus particles and thereby harbor mostly detritus-feeding meiofauna, while epistrate feeders prevail in coarse, branching algal bushes (Wieser 1959). Conversely, the mobilization and reduction of the detrital layers on the blades and thalli by feeding meiofauna may also be advantageous for the plants and enhance their growth. More sessile members of the phytal meiofauna regularly utilize the outer parts of plants as prominent substrates for filter feeding (e.g., rotifers, cladocerans). Rotifers are particularly frequent in the phytal fringes of brackish waters. Some harpacticoids (Diarthrodes sp., Amphiascoides sp.) live on macroalgae, where they mine the fronds and ingest the medullary tissues, producing algal galls (Hicks and Coull 1983). In relation to the successive alteration of the surface coating, Gwyther and Fairweather (2002) noted a succession in nematodes from epigrowth-feeders to deposit-feeders, while omnivores and predators followed later. The abundance of meiofauna exploiting the complex habitats in rocky algal belts has been repeatedly emphasized (Crisp and Mwaiseje 1989; Danovaro and Fraschetti 2002; Frame et al. 2008). Seagrass beds also harbor twice as many meiofaunal species as the adjacent sediments (Hicks 1986), and are considered “hot spots of meiofaunal production,” producing around 10 g C m−2 y−1; a value equivalent to the world’s most productive sediment sites (Danovaro et al. 2002). A phytal meiofauna of a million individuals per m2 of macroalgae is not uncommon and, in terms of biomass, may correspond to 10% of the macrofauna. In addition, hard substrate communities comprising crusts of mussel beds, barnacle, bryozoan and hydrozoan colonies, or thickets of worm tubes represent habitats with a rich yet poorly studied meiofauna (Somerfield and Jeal 1995). They provide shelter even under exposed littoral conditions, and are, in many respects, comparable to the belts of crustose, twisted or upright algae. The phytal biotopes, with their highly productive meiofauna, represent an important food source for higher trophic levels. While meiofauna are usually scarce on bare rocky shores, valuable meiofaunal food is available in algal belts and among seagrass beds, mainly for small fish and shrimps. However, experiments demonstrated that the accessibility to this food source is reduced by the vegetation and stems. In experiments with grass shrimps (Palaemonetes), this shelter function of the phytal resulted in a 40% decline in the consumption rate compared to unvegetated cages (Gregg and Fleeger 1998). In a New Zealand rocky intertidal Coull and Wells (1983) demonstrated that more complex algal habitats exhibited less predation on meiofauna by tide pool fish. Thus, in many phytal ecosystems meiofauna provide an important link to higher trophic levels (see Sect. 9.4). Studying phytal meiofauna poses some problems of quantitative sampling. The plants have to be carefully placed in sampling bags underwater so that any loss of
8.5 Phytal Habitats and Hard Substrates
323
fauna is avoided. An open cylindrical jar with a thick and softly attached rim of flexible silicone sealant has proven to be a simple and useful tool (Gibbons and Griffiths 1988), but solid enclosures easily cause currents which may displace some of the meiofauna. After the injection of some formalin in order to release the attached meiofauna, the water volume of the sample is siphoned off. For better reproducibility this procedure should be repeated several times. Since a good proportion of the phytal meiofauna are linked to not only the plants themselves but to the ambient sedimentary meiofauna and the overlying water column, the meiofauna of phytal habitats have an important function coupling benthic and pelagic fauna. Considering the often small sizes of the vegetated areas, the phytal meiofauna exhibits high diversity and species richness, which it achieves by exploiting the structural complexity of the phytal and its numerous microhabitats and ecological niches. A recent study (Bracken et al. 2007) shows that there is also a “reversed” mutualistic link between meiofauna and the algal community. The excretions of meiofauna (ammonium) easily provide the nitrogen needed by the algae (Cladophora). More detailed reading: Posidonia meadows, Novak (1989); production, Danovaro et al. (2002); reviews, Hicks (1985a), Gibbons (1991); monographs, Remane (1940); Wieser (1959).
Box 8.6 The Phytal Meiofauna: A Haven Depending on Structural Complexity The meiophyton, which lives preferably on and among plants (algae on hard substrates; seagrass on sand or mud), wonderfully exemplifies the influences of habitat structure and heterogeneity on faunal diversity and abundance. On foliose thalli or blades lives a meiofauna that is different in composition and abundance from tufted or crustose algae. The colonizers of the fronds and blades, with their frequent excursions into the demersal water layer, differ from the inhabitants of lower stems and holdfasts that merge with the sediment fauna. Regularly migrating meiofauna further blur the limits of both biotopes—yet, the phyton is understood to be a meiofaunal unit with specific adaptations and with a faunal composition of its own. Harpacticoid copepods often dominate, while nematodes often rank second. Ostracods and halacarids are other frequent members of the phyton. Climbing among the plant thickets is supported by a flat body surface and clinging legs. The phytal habitat with its numerous retreats provides shelter against strong currents and larger predators such as small fish and shrimps, while the microbial biofilm on the surfaces and a well-developed layer of debris provides an ample food supply. There are many parallels to the meiofauna populating clumps of barnacles, mussels, bryozoans, or dense patches of hydrozoans or annelid tubes. This suggests an ecological incorporation of substrates formed by animal epigrowth into a wider definition of the phytal. Favorable ecological conditions are the basis for one of the highest production rates in benthic habitats and afford the meiophyton a considerable ecological role.
324
8.6
8 Meiofauna from Selected Biotopes and Regions
Brackish Water Sites
Marine and limnetic sediments are connected by a brackish zone that is primarily characterized by its variable salinity. This is influenced by the tides, climatic, hydrodynamic and/or geographical factors. The future climatic change with increased salinity oscillations will globally affect the coastal soft-bottom zones with brackish conditions (Richmond et al. 2007). Coastal lagoons and tidal estuaries are typical brackish sites, as are large, semi-enclosed water bodies such as the Baltic Sea or the Black Sea. The salinity range between about 30 and 3 PSU can physiologically stress many fauna and can often limit their distributions. Another huge brackish water habitat, particularly for the meiobenthos, is the coastal subsurface zone where the marine and continental groundwater systems merge. The brackish coastal groundwater extends well above the supralittoral fringe of the seashore into the limnetic biome. Hence, it has always been pivotal for colonization and migration. Perhaps it was more shelter and less competition in this coastal groundwater system (see Sect. 8.7.2) that allowed meiofauna to adapt to brackish salinities. This long-term process, facilitated by biotopical stability, then led to a gradual transition from marine to limnetic conditions. Thus, this brackish zone has considerable relevance as an evolutionary pathway for the immigration of marine meiofauna into the continental groundwater system (see Chap. 7). This would also explain the relatively high percentage of genuine brackish water species among the meiobenthos (Fenchel 1978) and the high diversification of meiobenthic species adapted to all haline regimes. The weak separation of macrobenthic from meiobenthic biomass size spectra under brackish water conditions (see Sect. 9.2) may reflect this wide occurrence of meiofauna in various salinity zones (Drgas et al 1998; Duplisea and Drgas 1999). As a consequence, in contrast to the macrobenthos and the plankton, the meiobenthos is less constrained by a “brackish water species minimum” around salinities of 8–10% PSU (Remane 1940). Of course, in the most stressful intermediate range, between 5 and 10% PSU, even the number of meiobenthic species will decrease. However, the depletion of meiofaunal species in this critical range is relatively indistinct (Riemann 1966; Dauer et al. 1993; Yamamuro 2000; see Gerlach 1954 for nematodes). A significant size reduction, known from macrofauna in brackish areas, has also been found in marine nematodes from brackish lagoons compared to marine sites (Yamamuro 2000). Morphological aberrations, e.g., in the number, structure and position of amphids in populations of some Black Sea nematodes (e.g., Terschellingia longicaudata, Axonolaimus setosus, Sabatieria abyssalis), have also been interpreted as an indicator of extreme water conditions in combination with anthropogenic stress (Sergeeva 1999). In contrast with macrofauna, surprisingly many marine and freshwater meiofaunal species have developed a high tolerance capacity, usually to not only an unstable salinity but also to combined variations in temperature and oxygen supply. As a result, the distributional ranges of many marine and freshwater meiofauna in estuaries or in the Baltic Sea characteristically extend more widely into critical brackish zones than those of macrobenthos. Also in brackish water the influence of sediment characteristics as a controlling mechanism is apparent, evidenced by a comparison of meiofauna from various mesohaline regions in the Baltic Sea: Riga Bight, with its rich supply of
8.6 Brackish Water Sites
325
organic matter, is populated mainly by deposit feeders, in contrast with sediments with a low organic load around the Swedish east coast, where epistrat feeders dominated (Pallo et al. 1998). Estuarine sediments harbor many euryhaline marine species of nematodes and turbellarians, which co-occur with freshwater and brackish water forms (Riemann 1966; Heip et al. 1985a; Sopott-Ehlers 1989; Alongi 1990a; Ax 2008). Euhaline meiofauna even populate olighohaline reaches with considerable densities. On the other hand, the meiobenthic rotifer species that colonize the brackish region are limnogenic; in this taxon the percentage of marine species is very low. Therefore, under brackish water conditions a relatively diverse and rich meiofauna gains in importance when compared with the less resistant macrobenthos. Despite this high tolerance, experiments have demonstrated that meiofauna in salinity gradients display the characteristic discrepancy between the range of survival and that of reproduction known from macrobenthos. Meiobenthic animals survived over a fairly wide salinity range, but they only reproduced over a rather narrow range (Ingole and Parulekar 1998). In nature this means that in a gradient of salinities, e.g., along an estuary, the area of successful reproduction will be distinctly smaller than the area of a mere persistence (see Ekman 1953). In recent years the meiofauna in the brackish Black Sea and Sea of Azov have been examined in more detail by the teams of Sergeeva (1996) and Vorobyova (1999). The pertinent publications deal not only with the meiofauna of anoxic and sulfidic depths in the Black Sea (see Sect. 8.4.1) but also with the littoral sediments, mussel beds and algal epigrowth of this strongly stratified brackish water body. Some of these studies also cover the adjacent Sea of Azov (Vorobyova 1999; Sergeeva and Burkatsky 2002). Meiobenthic communities of high abundance but low species richness characterize the shallow sites of these brackish seas (5–13% PSU in the Sea of Azov) as extreme habitats. In muddy sediments foraminiferans and nematodes (plus some kinorhynchs) prevailed, while sandy areas harbored relatively numerous harpacticoid copepods. Other important groups were ostracods, polychaetes, and oligochaetes. Meiofauna accounts for about 38% of the total invertebrate species identified so far in the Black Sea. As average densities almost 300 ind. 10 cm−2 in the Black Sea and about 500 ind. 10 cm−2 in the Sea of Azov (corresponding wet weight biomass of >10 mg) are reported (Vorobyova 1999; Sergeeva and Burkatsky 2002). Thus, meiofauna are important for the overall productivity of these brackish sites. The increase in meiofaunal density upon increasing anthropogenic impact is interpreted as a compensational effect of meiobenthos for the decreasing macrobenthos. In estuaries, the astatic tidal regime, the salinity and the sediment characteristics are the main determinants of the meiofauna distribution (Soetaert et al. 1995; Coull 1999). Fine estuarine sediments support a less species-rich meiofauna than coarser ones (Dye and Furstenberg 1981). Because of the typically rich content of organic matter, food is rarely a limiting factor that facilitates meiobenthic life in the harsh oligo- to mesohaline zone (Austen and Warwick 1995; Heip et al. 1995; Ingole and Parulekar 1998). In a South African temporarily open estuary, the meiofauna was likely not food-limited, because the main food source, the microphytobenthos, was abundant despite intensive grazing rates …(Nozais et al. (2005). Typically, salinity gradients seem to control meiofauna distribution in estuarine regions. In Dutch North Sea estuaries, the outer polyhaline regions harbored a meiofaunal community rich in non-detritus feeding nematodes
326
8 Meiofauna from Selected Biotopes and Regions
(mainly deposit-feeding and predacious nematodes; annual average 3,200 ind. 10 cm−2), while in the mesohaline region the meiofauna were less numerous (average 2,300 ind. 10 cm−2) and more linked to a detritus food chain (Li and Vincx 1993). From South African estuaries, an average meiofaunal population density of 1,000 ind. 10 cm−2 was compiled (Dye and Furstenberg 1981), but locally up to 6,000 ind. 10 cm−2 were recorded. A seasonally varying meiofaunal abundance was noted by Coull (1999) in subtidal estuarine sites on the North Carolina coast of the US: during the first half of the year higher meiofaunal densities (max. 1,800 ind. 10 cm−2) were encountered in mud sites, and during the second half higher densities occurred in sand sites (max. 1,400 ind. 10 cm−2) (average values for a 22-year data set). A marked seasonality especially among harpacticoid populations was also observed in Belgian estuaries (Smol et al. 1994). Williams (2003) emphasized the aggravating role of a vertically increasing salinity gradient in the sediment of estuarine river mouths. Here, salinity and sediment characteristics appeared to be major factors influencing the distribution pattern, with a clearly negative relation to silt content observed. Since estuaries are often important traffic routes and are massively impacted by anthropogenic activities, Smol et al. (1994) calculated the influence of man-made hydrodynamic changes on the meiofauna. They predicted that increasing the tidal amplitude and current velocity due to shoreline regulations would decrease meiofaunal diversity but increase overall meiofaunal biomass. Since epibenthic species will replace interstitial species, the meiobenthos will be a more readily available and important food for the macrobenthos. The species composition of meiofaunal communities in brackish sites usually follows the general rule that nematodes are the main representatives. However, there are deviations from this rule that reflect locally varying conditions. In the Baltic Sea, the numerous nematode species, despite their high diversification, show a steady decline in species richness from the western Belt Sea to the eastern Bothnian Gulf (Arlt et al. 1982). Regarding taxon abundance, nematodes are not always followed by harpacticoids. In the rich meiofauna of an Indian estuary (average density 387 ind. 10 cm−2) turbellarians ranked second in abundance after nematodes (which contributed 60% of the total meiofaunal abundance); oligochaetes were third and harpacticoids fourth (Ingole and Parulekar 1998); ostracods are also fairly common inhabitants of estuaries. In some semi-enclosed brackish lagoons, the percentages were 50–85% nematodes, 8–33% turbellarians and only 7–25% harpacticoids (Escaravage et al. 1989). In the muds of Australian mangroves, turbellarians were the dominant taxon, followed by nematodes (Alongi 1987a; see Sect. 8.2.1). The high reproductive potential of many opportunistic turbellarians may explain this unusual prevalence. Among the Baltic Sea turbellarians, no endemic fauna were reported. All of the species also occurred in the supralittoral of the adjacent North Sea, where they represent a genuine brackish water fauna (Armonies 1988d). The surprising degree of faunal identity between Alaskan, Canadian and European coastal turbellarians has been ascribed to the well-developed brackish water tolerance capacity of meiofauna and the uniformity of brackish water biotopes. This results in a community of Platyhelminthes with a circumpolar distribution (Ax and Armonies 1990; Ax 2008). A lack of indigenous Baltic Sea representatives is also known for halacarids (Bartsch 1974), which are regular and frequent members of the brackish meiofauna. In a South African temporarily closed
8.6 Brackish Water Sites
327
estuary, astigmatid mites (Tyrophagus) play a considerable role after nematodes (Nozais et al. 2005). Here, the extreme abundance varies between almost 0 and 888 ind. 10 cm−2, indicating the high temporal and spatial variations that occur in estuaries. Lagoons appear to be brackish water bodies that are particularly favorable to meiobenthos, although here the salinity extremes seem to limit species richness and abundance of meiofauna (Castel 1992; Yamamuro 2000). In his review paper on the meiofauna of brackish lagoons, Castel (1992) emphasized the role of the nutrient-rich sediments in the abundance of meiobenthos. Since high abundances (often ranging from 3,000 to >5,000 ind. 10 cm−2) are often linked to a low species richness, he categorized lagoons as extreme biotopes. Meiofauna other than the usual nematodes and harpacticoids can make up the relevant population stock also in lagoons: high proportions of turbellarian (see above), ostracod and oligochaete species, well adapted to brackish conditions, are often found. In these shallow brackish systems, meiofauna can represent a particularly important nutritional resource for higher trophic levels, especially for juvenile fish. With their open accessibilities and limited species richness, coastal lagoons appear highly suited to population studies and experimental work on meiofauna. Freshwater tidal flats, like their brackish counterparts, are highly productive and ecologically important feeding grounds for macrofauna and juvenile fish. Their rich meiofauna appears to be an important trophic link (Yozzo and Smith 1995; Coull 1999). More detailed reading: Dye and Furstenberg (1981); Heip et al. (1995); Coull (1999); Ax 2008.
Box 8.7 Brackish Sites: Instability as a Criterion Salinity gradients of high temporal and local variability are the prominent character of brackish sea basins, estuaries and coastal lagoons. Interactions with other unstable factors such as changing temperature, solute composition and oxygen depletion in this stressful physiological milieu control the viability and restrict the distribution of species. Whereas in the critical mesohaline “minimum zone” macrofaunal assemblages are typically reduced in diversity, meiofaunal species often have a higher tolerance. Here, the meiobenthos becomes more important in species number and abundance compared to macrobenthos. The extreme nature of many brackish biotopes is often exemplified by huge populations comprising just a few adapted and widespread species. Even in the hardy nematodes, species richness declines in the intermediate haline ranges. The resistant nature of many brackish water species is demonstrated by the rich meiofauna in enclosed seas such as the Baltic or the Black Sea. Numerous meiofauna are found even in the anoxic sediments of the Black Sea. Brackish shallow lagoons with their rich nutrient supply and often intense primary production are particularly favorable sites for rich stocks of meiofauna with an important role in the trophic web. The more stable brackish groundwater horizons probably acted as important invasion routes for some marine meiofauna that gradually adapted to brackish and ultimately freshwater conditions. Hence, many subterranean continental aquifers and cave waters are populated by meiobenthic species of marine origin.
328
8.7
8 Meiofauna from Selected Biotopes and Regions
Freshwater Biotopes
When the first edition of this book was published, research on the small invertebrates of freshwater habitats was rarely considered in the context of “meiobenthology.” Links to the marine realm were rare and the terminology was different. The scattered literature was published in specialized journals not regularly read by the marine researcher. An early exception was the compendium Stygofauna Mundi (ed. Botosaneanu 1986a), a faunistic and zoogeographic compilation of the subterranean fauna: the stygofauna. Beside freshwater fauna, this volume also considered some marine taxa. Since that period, several comprehensive and competent reviews on the various freshwater biotopes have appeared (Allan and Castillo 2007; Gibert et al. 1994; Palmer et al. 2006; Hakenkamp and Palmer 2000; Robertson et al. 2000a; Wilkens et al. 2000), and improved methods have been developed in order to overcome sampling problems, especially in groundwater research (Mathieu et al. 1991; Malard et al. 1994). Detailed accounts of the various meiobenthic groups in sediments of running freshwater sites are given in the treatise on Freshwater Meiofauna edited by Rundle et al. (2002). The compiled knowledge of important taxa (e.g., nematodes: Traunspurger 2002; Eyualem-Abebe et al. 2006; ostracods: Horne and Martens 1994; turbellarians: Kolasa 2000; protists: Patterson 1996) now serves as a solid platform for further detailed studies. In groundwater research, the comprehensive publication by Hancock et al. (2005) and volumes on groundwater ecology (ed. Gibert et al. 1994; Wilkens et al. 2000) provided major progress. Supplementing the chapter on meiofauna in the second edition of Methods in Stream Ecology (Palmer et al. 2006), there is an electronic key to freshwater meiofauna (Strayer 2006) that can be downloaded to help the beginner in the field. Previously, various independent limnetic research lines had developed a diverging, specialized nomenclature and used different methods, originally with astoundingly little interconnection. Now, many of the new studies use a terminology that is similar to that in marine analogs, which had previously only been applied in the pioneering papers of Palmer (1990a,b). Bridging this traditional nomenclature gap indicates just how many aspects and trends freshwater and marine meiobenthos have in common, and fruitfully deepens our understanding of the structures and functions shared by both fields (e.g., Palmer et al. 1996). Groundwater research in particular demonstrates the numerous parallels and transitions between freshwater and marine meiobenthology. The recent emphasis on freshwater meiobenthology necessitates rewriting the pertaining section in this book. This section outlines the more general or summarized results without claiming complete coverage; for more details, the reader should consult the literature cited above. This increase in literature indicates that freshwater meiobenthology has gained considerable momentum. It is now apparent that: (a) much of the biodiversity in freshwaters is contributed by meiofauna …(Robertson et al. 2000b,c; Balian et al. 2008); (b) meiofauna are many times more abundant than the macrofauna, and; (c) their biomass could potentially be half that of the macrofauna, or may even exceed it (Poff et al. 1993; Stead et al. 2003). The biotopes of freshwater meiofauna are perhaps more heterogeneous than their marine counterparts, each harboring a
8.7 Freshwater Biotopes
329
specialized meiofauna. Groundwater aquifers, river shores, cave pools and lake bottoms, all are inhabited by a particular meiobenthos. Research on running waters was initially neglected, but recent research momentum (see above) has filled many ecological gaps and compensated for some of the previous lag in this field. The influential and early freshwater research on subterranean, mostly karstic fauna by Karaman and his school (e.g., Karaman 1935) as well as Chappuis (1942) and Rouch (1968) may have contributed to the notion that interstitial animals represented most of the freshwater meiofauna. Easier access and an apparent relevance also prompted early studies on the small bottom fauna of standing water bodies such as lakes and ponds (Wiszniewski 1934; Pennak 1940). Ecologically, the freshwater meiofauna often attains considerable importance, since together with bacteria and protozoa, meiobenthic animals are involved in the remediation of wastewater and the natural regeneration of our groundwater. The ecological constraints and the typical faunistic composition of the freshwater meiofauna differ much from those of its counterpart in the marine domain, although nematodes frequently dominate in abundance and biomass, just as in marine habitats. However, in many freshwater biotopes, rotifers are of equal importance, followed by copepods (cyclopoids and harpacticoids), tardigrades, “cladocerans” (Chydoridae), hydrachnid mites and oligochaetes. Insect larvae (mainly chironomids) are a specific temporary component of freshwater meiofauna without a counterpart in the marine world, but of considerable relevance in the limnetic food web (Pennak, 1940; Williams 1984; Schmid and Schmid-Araya 1997). However, variations from this generalized picture of freshwater meiofauna are often found when studying the different biotopes (Schwoerbel 1961a, 1967; Pennak 1988; Hakenkamp and Palmer 2000; Hakenkamp et al. 2002), and so they each require separate discussion. An artist’s view of the biotopes may help to illustrate the low concordance between marine and freshwater meiobenthic taxa (Figs. 8.11 and 8.12).
8.7.1
Running Waters: Stream and River Beds
The habitat. Although much of the biodiversity of the fauna populating stream beds is due to meiofauna, this faunal compartment is often neglected by freshwater ecologists (Stead et al. 2003). The first ecological studies on freshwater meiobenthos were performed in river beds, but this “hyporheic” and “phreatic” (in deeper sediment) meiobenthos is less well known than the meiobenthos in lakes (see Hakenkamp and Palmer 2000; Robertson et al. 2000b; Sect. 8.7.3). In many headwater habitats with coarse sand or gravel bottoms, this is probably due to problems with accessibility and sampling, making quantitative data scarce. The hyporheic habitat of the riverbed is mainly characterized by water flow, while the phreatic sediments continuously merge to the groundwater horizon (Fig. 8.13; Pennak and Ward 1986). In the groundwater aquifers underlying phreatic habitats, the fauna becomes more homogeneous with little or no altitudinal differentiation. Regarding the intense interactions between the epigean, hyporheic and groundwater zones of
330
8 Meiofauna from Selected Biotopes and Regions
Fig. 8.11 Artist’s impression of the freshwater interstitial environment and its fauna
Fig. 8.12 Artist’s impression of the marine interstitial environment and its fauna
8.7 Freshwater Biotopes
331
zone of aeration
zone of aeration
hyporheic zone phreatic or groundwater zone (stygal) rock
Fig. 8.13 The habitat zonation of a river bed. (After Pennak and Ward 1986)
a river, Brunke and Gonser (1997) termed the hyporheic zone a “connecting ecotone”—a corridor used and inhabited by rich, diverse and interacting meiofaunal communities. These physiographic connections, which result in exchange rates that were higher than previously purported, make any strict delineation of the biotopical subunits (Fig. 8.13) rather arbitrary (Danielopol 1989, 1991; Lafont and Durbec 1990; Lafont et al. 1992; Dole-Olivier and Marmonier 1992). The dominant ecological factor in freshwater habitats is water flow. Water flow separates the torrential headwater streams with their beds of pebbles and barren cobbles from the slower lentic lowland rivers, with their fine sediments and rich macrophytes. The other characteristic of all hyporheic habitats is astatic and seasonal fluctuations. Hence, hyporheic meiofaunal assemblages differ in the various parts of a running-water system. Seasonal changes in temperature and water level together with a variable chemical milieu add to this instability. Especially at low water levels, the hyporheic fauna can become threatened by river pollution. Characteristic abiotic features of the hyporheic pore water system compared to the overlying river water are an increased CO2 content, which is reflected by a decrease in pH by 1–2 units. Strong adsorptive forces in the sediment often also cause an increase in the concentrations of silica, iron and manganese. In the coarser sediments of the riverine headwater region, the strong water currents maintain an open interstitial system with a good permeability, which compensates for the oxygen depletion caused by the supply of organic matter. In particular, crystalline sediments on primary rocks maintain an open pore system, acting as a filter bed for the few detritus particles. Detritus will accumulate in the more lentic lower river bed, clogging the interstices of the sand. Silt and mud particles compact the river bed in sheltered riverine side-arms and coves. Combined with intense bacterial growth, this accounts for typical decreases in oxygen and pH (Fig. 8.14) as well as nitrate and sulfate concentrations, while reduced substances and free hydrogen sulfide may develop locally and can act as major controls over hyporheic communities (Strayer et al. 1997).
332
8 Meiofauna from Selected Biotopes and Regions 20
16 CO2
O2
pH
12
8
4
0 overlying water
sediment surface
sediment sediment 50 cm 100 cm
Fig. 8.14 Gradients of chemical characteristics in a vertical profile of a typical lakeshore. Note the decrease in oxygen values (in ppm) as opposed to the increase in carbon dioxide values (in ppm). (After Pennak 1939)
Generally, the hydrological and geomorphological regimes seem to override all other factors in running water ecosystems and determine the interactions and roles of meiofauna (De Bovée et al. 1995; Strayer et al. 1997; Hakenkamp and Palmer 2000): at high flow rates the macrobenthos dominates and the meiobenthos attains a subordinate role despite its often considerable diversity. At intermediate flow rates the meiofauna gains importance, filtering detrital particles and breaking down organic matter. This intensifies bacterial processes and contributes greatly to total secondary production. At low flow rates patches of accumulating detritus develop, where the depletion of oxygen can cause a vertical upward migration of the hyporheic meiofauna and can increase losses by drift. These discontinuities in the interstitial water flow and local biotic factors may modify the general hydrological and geomorphological scenario and thus the hyporheic fauna (Dole-Olivier and Marmonier 1992, Ward and Palmer 1994). Overall, and in contrast to the aquatic epibenthos, the hyporheic fauna of running waters is structured by the site-specific physiography rather than the elevation (mountain river vs. lowland river). Among the biogenic factors, organic matter is the key component that influences attracts hyporheic meiofauna and controls their patchy distribution. The load of detritus, with its rich epigrowth of bacteria and microalgae, serves as the main food source. Sites in headwater streams with open access to both detritus and oxygen will support a diverse and quantitatively important meiofauna (Ward and Voelz 1990). In low-flow river beds suboxic reaches often develop. Here, the composition of the meiofauna changes; the diversity and abundance will decrease. As an
8.7 Freshwater Biotopes
333
adaptation, many of the hyporheic animals seem fairly resistant to a reduced oxygen content in the pore water. However, water velocity, sediment composition, oxygen and detritus supply may vary on small scales—around one boulder, within a stand of vegetation, or in a sand riffle. Depending on the bottom microstructure, downwelling and upwelling sites can form. For meiofauna, these local properties are often the real controls over the distribution pattern and community composition. For instance, cyclopoid copepods and cladocerans prefer areas with reduced flow, while harpacticoids are more common in coarser sediments with a high water flow and shear stress (Robertson 2002). Water flow also enhances the rate of faunal suspension. Although surface films of bacteria and diatoms tend to stabilize the sediment, sudden floods with subsequent heavy erosion and faunal suspension are common phenomena, and not only in montane rivers. Passive drift accounts for a repeated redistribution of meiofauna and is important for the dispersal and regeneration of communities after severe disturbances by flash floods (Palmer 1992; Palmer et al. 1996). The degree of community destruction apparently depends on the duration of the flood, but losses are usually compensated within a few weeks and only occasionally within several months (Hancock 2006). This swift recovery is largely made possible through transport from regions higher up the river. Transport in turbulent waters will increase the probability of hitting the bottom and resettling (McNair et al. 1997). Although only a small proportion of the fauna has been found to immigrate from the hyporheic strata (Palmer et al. 1992), the (deeper) hyporheic biotope can be considered a refuge area for fauna, allowing them to avoid currents and drift. With its transitional position between the exposed riverbed and the stygobiotic groundwater sediments, the hyporheos attains considerable importance as a sheltered recruitment zone, especially for insect larvae. Nematodes were encountered down to 50 cm depth in running water sediments. While the deep hyporheic zone as well as lentic “pockets” along the shore may be important as havens that are well supplied with accumulating debris, their role as centers of recolonization, supporting the “hyporheic refuge hypothesis,” is limited (Robertson 2000; Palmer et al. 1992; Olsen and Townsend 2005). Thus, dispersal by flooding, regeneration by incoming drift, a high potential for rapid colonization, as well as frequent redistribution of communities are regular features of running water biotopes. Combined with a high degree of resilience, the consequences of flood events are limited (Palmer et al. 1995, 1996; Robertson 2000). At reduced interstitial flow rates micropatches with favorable or unfavorable combinations of factors can develop and create a high spatial heterogeneity that changes temporally whenever flow conditions change. The previously fairly homogeneous spatial distribution of meiofauna in the hyporheic zone starts to become patchy. Hydrodynamics apparently also structure the temporal development and composition of riverine meiofauna. Seasonality in hyporheic abundance patterns is well developed, with minima in the spring following high waters in springtime. Under low-water conditions (which often occur during the summer months), the epibenthic fauna is confined to a thin surface horizon while the endobenthic fauna dominate throughout all the deeper layers. This situation reverses at times of high water
334
8 Meiofauna from Selected Biotopes and Regions
discharge. During phases of low stream velocities, the fauna of deeper horizons often suffer from stagnant pore water conditions that favor, expecially in the summer, the development of hydrogen sulfide. In cold winters, the surface layers are easily exposed to frost and ice scraping, with its concomitant noxious impact on the inhabitants of these layers. They may survive frost by evading deeper down into the hyporheic interstitial (Schwoerbel 1967). In the cold season, nematodes, copepods and oligochaetes were found deeper down in the sediment (Palmer 1990a). Species composition. In the hyporheos, in great contrast to the marine meiofauna, rotifers, microcrustaceans (cyclopoids, harpacticoids, cladocerans, ostracods), microdriline oligochaetes and insect larvae often attain numerical dominance over nematodes. Usually, nematodes are favored in physically rigid sites, while other groups may prevail under less severe conditions. The headwaters of a river, with its bed of pebbles and gravel, its huge internal surface and its high structural complexity, harbors a more diverse hyporheic meiofauna than the sandy-to-muddy lower reaches. Cyclopoid and harpacticoid copepods, small isopods, tardigrades and smaller insect larvae prevail in the headwaters. With finer sediment and often increasing organic load, the riverine meiobenthos becomes dominated by oligochaetes and chironomid larvae, while the soft sediments of lowland streams are the domain of nematodes, chironomids and rotifers (Hakenkamp and Palmer 2000). In the bed of a low-gradient stream, Palmer (1990a) found 35–85% rotifers, followed by 20% juvenile oligochaetes, then chironomid larvae, and, to a lesser degree, nematodes and copepods. In field experiments with test tubes (Schwoerbel 1967), nematodes preferred the tubes filled with the finest sand (0.25 mm), while harpacticoids were most frequently found in tubes filled with gravel of size 4–6 mm; the third group of importance, chironomid larvae, dominated in sand of 1–4 mm grain size. Some rotifer genera (Notholca, Lecane) were found to be numerically dominant in alpine streams (Schmid-Araya 1995, 1998). In another stream, Ward and Voelz (1990) found that small ostracods and chironomid larvae accounted for 65% of all meiofauna. Another group characteristic of the hyporheos are chydorid cladocerans. Small isopods, which are rather frequent in European river beds, seem to be absent in North America (Williams 1989). In Scandinavian springs, oligochaetes and ostracods dominated the hyporheos (Särkkä et al. 1997). It is mostly at intermediate water velocities (~ 30 cm x s−1) that a relatively rich taxonomic diversity (nematodes, oligochaetes, microcrustaceans, flatworms, tardigrades and midge larvae) combined with high abundances can develop (Whitman and Clark 1984). In many lotic habitats the meiobenthos contribute far more than half of the total species richness (Robertson et al. 2000; Stead et al. 2003). Complete inventories often yield 150 species and can even exceed 300 species, of which nematodes contribute 30–50 spp (Hodda 2006). The sheer variety of microhabitats in lowland streams supports species richness. This is especially true for microcrustaceans, some of which are restricted to running waters (Rundle and Ramsay 1997). Still, our knowledge of the quantitative composition of hyporheic meiofauna is limited. Problems with quantitative sampling and detailed determination make calculations of species richness and population density per area or volume difficult (see below).
8.7 Freshwater Biotopes
335
Abundance, biomass, and production. In general, sediments of streams are not particularly densely populated by fauna. Coarse sand bottoms in the riverbed (less so in the shore sediments) harbor meiofauna with abundances of 20–30 ind. 10 cm−2. In an acidic stream the number of meiobenthic individuals was even less, although at >10 ind. 10 cm−2 (maximum 84 ind × 10 cm−2) it exceeded that of the macrofauna (Stead et al. 2003). Schwoerbel (1967) found a relatively high abundance of hyporheic meiofauna, 60–100 ind. 10 cm−2, in the sand of an alpine mountain stream; a range that is often similar to that of nematode communities (see compilation by Hodda 2006). Extremes of several thousands per 10 cm2 may represent patches, e.g., up to 6,000 meiobenthic organisms per 10 cm2 in the bed of a North American low-gradient stream with mostly rotifers (Palmer 1990a) or 4,100 nematodes 10 cm−2 in a small sandy river in Germany (see Hodda 2006). However, these figures do underline the taxonomic and numerical relevance of the hyporheic community of streams. Interestingly, Pennak and Ward (1986) found substantially more interstitial crustaceans in a mountain river than plankton in a nearby lake. Estimates of biomass and production rates for lotic meiofauna are rare. Depending on the substratum, between almost zero and 22% of the biomass (in sand) has been attributed to hyporheic meiofauna (Hakenkamp et al. 2002), with a higher contribution to the overall biomass in lentic than in lotic environments. Stead et al. (2003, 2005) published even higher meiobenthic biomass values: between 10 and 100 mg m−2 (dry wt), values close to the macrofaunal biomass. According to Kowarc (1990), the meiofaunal production in the gravel of mountain streams is rather low, with a P/B ratio of 3–6, which is probably caused by the low temperatures and generally oligotrophic conditions. A low-to-moderate (often less than 5%) contribution of meiofauna to the metazoan production has been adopted for lotic ecosystem productivity (Robertson et al. 2000; Hakenkamp and Morin 2000, Bergtold and Traunspurger 2006), but Hakenkamp et al. (2002) also report values of up to 50% in streams. Only exceptionally, especially in creeks with sandy bottoms, does this value increase considerably (Poff et al. 1993). However, regarding the rapid meiobenthic turnover, low biomass values can sustain a sizeable production (see Sects. 9.3.2 and 9.4). Stead et al. (2005) assessed the complete faunal spectrum of the benthos (retained in sieves down to 42 µm mesh size), evaluated over a period of more than a year, in an acidic English stream. They found that 15% of the total production (5.2 g dry wt. m−2 y−1) was due to the activity of permanent meiofauna, and together with the temporary meiofaunal taxa (oligochaetes, chironomids, plecopterans) more than half the production was contributed by meiobenthos. The authors argue that this result points to a “substantial underestimation” of meiobenthic productivity in many studies, arising from problems with appropriate assessment. Food relations. The predominance of bacterivores and detritivores among the meiofauna highlights the role these trophic components play, even in lotic freshwater habitats. Many chironomids, rotifers, ciliates and (to a lesser degree) nematodes graze extensively on bacteria. Particularly at times of high meiofaunal densities, the grazing pressure on bacteria and diatoms is significantly high (Borchardt and Bott 1955). Thus, meiofauna gain importance as a link between microbial grazing, reworking of the sediment and food for macrofauna. Numerous studies of various
336
8 Meiofauna from Selected Biotopes and Regions
running-water habitats and experimental data conclude that the balance between macro- and meiofauna in terms of biomass and production as well as functional relevance will favor meiofauna in fine-grained sediments and a low interstitial flow regime (see also Hakenkamp and Morin 2000). At the predatory level of the food web, considerable activity from mites and the larvae of tanypodid chironomids and plecopterans has been reported (Schmid and Schmid-Araya 1997). Caging experiments demonstrated that predation on meiofauna is significant, particularly among young fish. Also, the biomass size spectra of meiofauna can be used to gauge the importance of differently sized organisms, indicating predator–prey interactions. However, for lotic freshwater habitats this approach is controversial: the results of some studies indicate well-separated biomass peaks for meio- and macrofauna (Poff et al. 1993), while in others the biomass distribution is more even and macrofauna only slightly prevail (see Sect. 9.2; Robertson et al. 2000c). When generalizing the data on the meiobenthos in running-water systems, the following aspects should be emphasized: - Much more information beyond local studies is required to provide a general, zoogeographically reliable overview. - While streams are relatively better understood ecologically, the meiofauna ecology of large rivers requires urgent investigation, especially since the meiofauna here are purportedly of higher ecological importance, but are being placed under much anthropogenic stress. - Life history data, estimates of turnover and the productions of key species are urgently needed. - The impacts of ecological parameters such as bioturbation or water drift still need to be assessed. - An appropriate methodology that catches epibenthic as well as interstitial fauna is a prerequisite for a total-component analysis. Methods. The scarcity of generalizable data on the hyporheic meiobenthos is often due to inherent technical and methodological problems. Only in the sandy bottoms is quantitative work with regular station profiles possible. Here, the standpipe corer (Williams and Hynes 1974) allows quantitative sampling into deeper strata too, enabling intact samples to be collected from defined depths. Battery-powered suction corers, constructed for quantitative sampling in marine bottoms (Taylor et al. 1995), might be applicable, as might the use of Scuba divers. In the pebbles and gravel of most riverbeds, sample holes can only be driven with massive corers, and the pore water that enters is pumped up for faunal analysis. The typical method used for lotic freshwater habitats as well as for groundwater research (see below) is pumping with the Bou–Rouch pump (Bou 1974; Fig. 8.15). A small pump is mounted on a perforated metal tube that has been hammered into the riverbed and remains in position for the repeated sampling of interstitial water. This pore water and its inhabitants can enter the tube through perforating holes. It is arguable whether this method will yield reliable reproducibility or area-related quantification, since most of the methods based on pumping do not allow exact reference to a distinct depth or area. These arguments also apply to trapping methods (Hahn 2003),
8.7 Freshwater Biotopes
337
Bou-Rouch pump
O2-analyzer air-pump
groundwater pipe with holes
O2-sensor groundwater
5 cm sampling chamber
filter
Fig. 8.15 A coring tube for fractionated sampling of groundwater fauna. For details see the text. (After Danielopol and Niederreiter 1987)
which have yielded comparatively satisfying results in test series. The corers designed by Danielopol and Niederreiter (1987) and Tabacchi (1990) provide more quantitative access. They allow for fractionated vertical subsampling of interstitial water and fauna by subdivision of the coring tube into small chambers, which are evaluated separately (Fig. 8.15). Quantitative results can be obtained with a more
338
8 Meiofauna from Selected Biotopes and Regions
sophisticated but expensive method. A sediment core obtained in a metal tube is shock-frozen in situ by liquid nitrogen and then retrieved undisturbed after previous anesthetization of the fauna by electro-positioning (Bretschko and Klemens 1986). Limitations on quantitative sampling could perhaps mitigated by increasing experimental work (Palmer 1993). More detailed reading: general and ecology, Schwoerbel (1961a); Pennak and Ward (1986); Palmer (1990b, 1992); Ward and Palmer (1992); Brunke and Gonser (1997); Hakenkamp and Palmer (2000); Robertson et al. (2000c); Rundle et al. (2002); methods, Bretschko and Klemens (1986); Palmer et al. (2006).
8.7.2
The Groundwater System
Groundwater represents a freshwater reservoir of eminent importance. About 40% of all freshwater, ice included, is stored in the continental groundwater system (Danielopol 1989); the water masses in all lakes and rivers account for only about 4% of the global groundwater reservoir. An account of groundwater ecology (not specifically meiobenthos) has been edited by Gibert et al. (1994). An even wider topic is covered by Subterranean Ecosystems (ed. Wilkens et al. 2000). The importance of groundwater aquifers is based on the fact that they supply us with drinking water, an indispensable human resource. The contact with the surface waters, mediated by the hydraulic conductivity and the texture of the sediment, determines the typical abiotic characteristics of this biotope. With increasing depth, the homogeneity of its sediments, the constancy of its physiographic conditions and homeostasis over long geological periods are the main biotopical features of the groundwater system. Its abiotic milieu is characterized by fairly constant, low temperatures, a somewhat lowered pH, a slightly undersaturated oxygen content, a high amount of free CO2, and oligotrophic conditions (Fig. 8.14). However, groundwater systems are not separated from surface waters. They are progressively more “open” the closer they are to the surface and the coarser the sediment pores. The “stygobiotic” meiofauna of the groundwater aquifers and the “troglobitic” cave meiofauna are specifically adapted to the above ecofactors. They exhibit slow locomotion and thus a minimal migratory ability, low metabolic activity and growth, long generation times and lifetimes, late maturity and low fecundity (a few, large eggs), and hardly any diurnal rhythms. Progenesis seems to be a relevant evolutionary factor (e.g., in syncarid crustaceans, in some amphipods and isopods). All of these attributes characterize many meiofauna as K-selected specialists. Adaptations of the interstitial fauna to groundwater life are detailed in Coineau (2000). Typical stygobiotic species are mostly cold stenothermal but they are well adapted to the low oxic conditions of many groundwater habitats (see Fig. 8.16). A relatively welldeveloped tolerance to (low) salinity points to the evolutionary origin of many stygobionts from the coastal groundwater (see below). Energetic sources in the groundwater realm are scarce; the food web is rather simple with few trophic links (Gibert et al. 1994). The absence of primary producers and the scarcity of predators (cyclopoids) are why Gibert and Deharveng (2002) considered subterranean ecosystems
8.7 Freshwater Biotopes
339
Stygobionts
Other Animals
n /100 l
> 100
10
5
2 20
40
60
a
100
% O2
Saturation Stygobionts
% 100
n /100 l > 100
O2
80 O2 Saturation
80
60 10 40 6 20 2 20
b
40
60
80 100 120 Depth in Substrate (cm)
Fig. 8.16a–b Occurrence of stygobiont specimens in groundwater habitats of Caribbean islands. a Abundance of stygobionts vs. “normal” marine specimens in relation to oxygen saturation. b Occurrence related to depth in the substratum and oxygen saturation. (After Stock 1994)
“functionally truncated.” Depending entirely on heterotrophy and living on detritus entrained from the surface, the groundwater community shows ecological and sometimes even taxonomic parallels to that of the deep-sea (see below and Sect. 8.3). Altogether, the underground biocoenoses are less densely populated than the hyporheic riverbeds. In the ecotone layer, where hyporheic and groundwater meiofauna mix, biodiversity and density are concentrated. At a global scale the groundwater meiofauna can be considered almost unknown; our knowledge is mainly limited to a few areas in North America and Europe. They are characterized by a wealth of crustacean species (in Europe about 40% of all crustacean fauna are stygobites!) and an absence of the insect larvae that usually dominate in the riverine and lacustrine meiofauna (Danielopol et al. 2000). Figure 8.17 attempts to depict some characteristic inhabitants, the stygobites, of this “unseen ocean beneath our
340
8 Meiofauna from Selected Biotopes and Regions
feet,” which extends worldwide and to a depth of several hundred meters. Stygobites are often continentally distributed and their compositions are consistent enough to form characteristic biocoenoses (e.g., the Bathynella–Parastenocaris community). In many ways these biocoenoses are related to the fauna of caves and springs, as exemplified by the freshwater polychaete Troglochaetus beranecki or the amphipod genera Niphargus and Bogidiella. Typical representatives of the groundwater fauna are minute malacostracan crustaceans: amphipods and isopods that are tiny enough to live in the interstices of the mostly karstic sediments (amphipods Niphargus spp, Salentinella spp., Stygobromus, the latter so far only found in North America; isopods: Microcerberidae, Cirolanidae, Microcharon, Proasellus, Microparasellus). The other characteristic group is the copepods; often cyclopoids (Diacyclops, Acanthocyclops) but also some harpacticoids (e.g., Nitocrella; Phyllognathopus spp.; Chappuisius spp.). In addition, stygobitic ostracods are well represented, especially members of the family Candonidae. A few representatives of the rare crustacean orders Thermosbaenacea (Pancarida) and Syncarida (Bathynellidae) supplement the suite of crustacean groundwater inhabitants. The microdriline oligochaetes frequently encountered in groundwater belong to the haplotaxids, rhyacodrilids, and tubificids. Some of them are descendants of marine lineages such as Phallodrilus sp. or Troglochaetus, while other annelids represent temporary meiofauna. Even minute gastropod molluscs are encountered in the subterranean aquifers (e.g., Paladilhia, Arganiella). Groundwater aquifers and caves have attracted the attention of zoologists, initially for evolutionary reasons, and later (and especially in the karstic regions of Europe
Fig. 8.17 The interstitial habitat of subterranean meiofauna in an Austrian aquifer. (After Danielopol et al. 1994a)
8.7 Freshwater Biotopes
341
and North America) as reservoirs of new species, “hot spots” of biodiversity. The abundance of species decreases sharply with increasing depth and isolation from the surface waters. Here, the number of endemics and relicts without a connection to surface fauna increases. This is especially evident among small crustaceans, such as thermosbanaceans, which are found in groundwater biotopes in seemingly complete isolation. There are both geological and ecological reasons for this interesting evolutionary diversity. A scarcity of predators and reduced competition have permitted the persistence of taxa with low mobility and little competitive strength in a biotope with numerous microhabitats characterized by ecological isolation. What were the pathways along which the groundwater was colonized by meiofauna? It is generally accepted that seashores were the most probable regions from which the colonization of the continental subterranean sediments started. Along this marine–groundwater route of meiobenthic dispersal and evolutionary change, river mouths, brackish lagoons and anchihaline caves may have served as “ports” and interconnecting pathways (see Chap. 7). Thus, some of the meiofauna populating wide areas of ancient seashores (e.g., the Tethys Sea) could have avoided the stressful dynamics of continuous geological change (marine transgressions or regressions, desiccation events, but also competition and predation by macrobenthos) by entering the more protected deeper sediment layers, especially the adjacent groundwater horizons, by gradually becoming “stygobites”. It was not just small size that adapted some meiobenthos to become colonizers of the groundwater system and caves. Pre-adapted by their frequent former exposure to brackish water conditions (see Sect. 8.6), the species could slowly adjust to freshwater conditions provided that they were exempted in their new localities from the stress factors characterizing surficial habitats. Their often euryoxic nature also supported their evolution in caves and groundwater systems, since these habitats are frequently dysoxic (Fig. 8.16; Stock 1986, 1994). In the fairly static refuges, relicts of primitive marine taxa with little adaptive potential could survive (“relict refuge model”; Botosaneanu and Holsinger 1991). They reflect the fauna in their old “plesiotopes” and cannot cope with the adaptive demands of the open marine littoral. A karstic surrounding supported the “stygobization” of meiofauna and their further subterranean distribution. These processes resulted in the separation of the originally large, cohesive shore stocks into small, isolated hypogean populations that became “founders” of radiating lines by allopatric speciation. Under the low selective pressure of the static groundwater world, speciation is slow. As a result, many stygobiotic animals from the continental groundwater are endemics and considered isolated relicts of a primitive marine fauna with formerly wider marine distribution. Morphologically primitive and often neotenic crustacean groups like Bathynellacea and Pancarida (see Fig. 5.36) are good examples of this process. Since the vagility of this evolutionary static fauna was limited, their present-day distributions often reflect the geological fate of their former habitats, they persist along the ancient sea shores: vicariant, e.g., amphi-Tethyan or amphi-Atlantic distributions. This “geological scenario,” mainly developed by Stock (1994; see also Strayer 1994) would also explain the rich and isolated groundwater fauna of the
342
8 Meiofauna from Selected Biotopes and Regions
karstic Balkan, but also the astonishing insular groundwater fauna of the Caribbean and Middle American shores, where anchihaline caves are particularly frequent. Numerous fascinating meio- and macrobenthic animals have been retrieved from their shelter. Present-day islands, often ancient seamounts, also play a dominant role as stepping stones and centers of speciation (see Chap. 7). The “dispersionist school” of groundwater researchers is inclined to consider hypogean fauna as being mainly represented by animals that keep actively invading the stygobiotic freshwater milieu from other, often marine biotopes (Danielopol and Rouch 1991). Studying subterranean ostracods, Danielopol et al. (1994) state that dispersal by drift and active migration enable hypogean meiofauna to constantly proliferate into the surrounding habitats. The ostracod distribution is better explained by an ecological, dispersive scenario than a scenario of geological events. It is again mainly the degree of ecological flexibility and the capacity for tolerance that control the distribution of the species into the subterranean and epigean aquatic system. Another source of groundwater colonization may be the riverine meiofauna. There are, indeed, numerous connections between the groundwater and the hyporheic and epigean fauna of running waters. In downwelling areas of gravel bars, the epigean fauna from the surface becomes infiltrated into the interstitial of hypogean sediments. Contrastingly, in hyporheic areas with prevailing upwelling groundwater, the contribution of stygofauna from the deeper layers progressively increases in the interstitial system (from 8 to 47%, see Ward and Palmer 1994). A correspondingly varying influx of stygofauna into the riverbed was observed in studies of the floodplain of the River Rhone at sites with differing hydrological regimes (Dole-Olivier and Marmonier 1992). These studies emphasize the close connections of the groundwater assemblages to those of the surface waters. Boutin and Coineau (1991) point out that the colonization process represents a combination of dispersion and persistence. Their “two-phase model” is characterized by several successive steps. Initially there is active dispersion and vertical transition from surface waters to the interstitial. This is followed by passive persistence combined with gradual adaptations that further separate the interstitial fauna of the continental groundwater from that of the original marine habitats. Whichever mechanism, per se or in combination, was the driving evolutionary force, a confusing mosaic of archaic and derived characters is typical of the stygobios. Ecological stability and reduced competition in the simple subterranean ecosystem would explain the high biodiversity combined with the low abundance of the stygofauna. As examples, the karst regions in Europe and North America, and to a lesser extent the interstitial of alluvial aquifers, represent rich species reservoirs for some crustacean groups such as ostracods, harpacticoids, amphipods, and isopods. For these taxa and biotopes, the species richness can be equal or can even exceed those of surface freshwater habitats (Rouch and Danielopol 1997; Danielopol et al. 2000). In North America, the glacial boundary of the last ice age seems to represent an important distributional barrier for the hypogean fauna. A typical groundwater fauna that is speciose and rich in K-selected endemics, e.g., Bathynellacea, has only been found in formerly unglaciated areas. Both the upland plains (influenced by the ice cover) and the coastal plains (under the impact of the marine regime) are poorer
8.7 Freshwater Biotopes
343
in groundwater fauna (Strayer et al. 1995). Characteristic groundwater cyclopoids on the North American continent seem to follow the same distributional constraints. Once adapted to the physiological constraints of freshwater, stygobiotic species do not seem able to bridge the oceans (Schwoerbel 1967). Their continental distribution is linked to the large aquifers, for instance those of the Rhine and the Danube in Central Europe. Considering the limitations described above, the surprisingly frequent circummundane distributions of stygobiotic species cannot be conclusively explained. With the ongoing discovery of huge subterranean systems, it is likely that many more zoological “preciosa” will be discovered and that the distribution patterns of stygobiotic animals will become better understood. The ecological knowledge of groundwater meiofauna clearly lags behind that of rivers and lakes. This is especially true if we consider the scarcity of quantitative data on biomass (allowing for calculations of productivity), which is mostly due to methodological problems. Direct sampling of defined volumes in the aquifers deep beneath the surface is restricted. Even elaborate fractionated samplers (see Fig. 8.15), traps (Hahn 2003), and pumping systems that render access through narrow drill holes (Malard et al. 1994) do not allow us to draw quantitative conclusions about the sample volume collected. Notable exceptions are water accumulations in caves. If sufficiently isolated from the outside, the ponds and streams of caves harbor a “cavernicolous” (troglobotic or troglobiotic) meiofauna, which usually represents a typical groundwater fauna that shares the restrictive characters of stygobitic life but allows for quantitative evaluation. Ecological studies performed under these cave conditions, preferably experimental work, might allow conclusions on the biomass and production also of the ambient groundwater fauna. Usually, aquatic cave habitats are oligotrophic and carry little meiofauna. Exceptions include some thermomineral caves with effluents of methanic and sulfidic water, e.g., Movile Cave in a karstic area of Romania. Here the ponds are populated by a fauna completely based on a rich bacterial production which, in turn, supports a meiobenthic community dominated by bacterivores (mainly nematodes) and predators (cyclopoids). The rich supply of bacterial organic matter (and the high temperatures) enables high population growth and consumption rates (Muschiol and Traunspurger 2007; Muschiol et al. 2008a). The groundwater fauna inhabits a very delicate biotope that is of great importance to mankind’s water resources. Their high sensitivity to pollutants makes stygobiotic species valuable indicators of declining water quality. Changes in their populations should be used as early signals of the potential contamination of drinking water and the need for bioremediation. With regard to the slow speed of hypogean regenerative processes, the risk of pollution by anthropogenic chemicals requires a sustainable water management concept for groundwater habitats. Implementation of this concept is imperative in order to secure our future, considering the decrease and long-term deterioration of our groundwater reservoirs globally. Areas with a diverse or unique meiofauna possess a particularly high value beyond just the scope of scientific exploration: they demand protection (Hancock et al. 2005; Marmonier et al. 1993; Danielopol 1989, 2000b).
344
8 Meiofauna from Selected Biotopes and Regions
More detailed reading: Delamare Deboutteville (1960); Danielopol (1990b, 2000a, b), Botosaneanu and Holsinger (1991); Gibert et al. (1994); Stock (1994); Brunke and Gonser (1997).
8.7.3
Standing Waters, Lakes
Lake sediments are more easily sampled quantitatively and thus have been more thoroughly investigated than other freshwater biotopes. However, only the meiofauna from a few lakes have been reported, in contrast to the numerous papers on the littoral parts of the sea. In particular, quantitative data on lake meiofauna (abundance, turnover and production) acquired over longer periods of time are scarce, although the ecological role of such meiobenthos is becoming increasingly apparent (see below). Modern lacustrine meiobenthic research is based on the classic book by Strayer (1985) on Lake Mirror, USA. In Europe, thorough investigations of northern lakes have increased our knowledge of the contribution of the meiobenthos to benthic biomass and productivity (Sarvala 1998, focusing on copepods; Kurashov 2002). Some oligotrophic lakes in Germany have been studied in quantitative detail by Traunspurger (2000) and his team (Bergtold and Traunspurger 2005). A chain of volcanic lakes in Ethiopia was the focus of an international project coordinated by Tudorancea and Taylor (2002). The latter study connects calculations on the production of bacteria and protozoa to meiobenthic production (mainly nematodes) and discusses their role in the total benthos, underlining the need for a holistic ecosystem approach. The need for comprehensive, interdisciplinary projects in lacustrine (meio)benthology was also the basis for the Cytherissa Project in the Austrian Mondsee (Danielopol 1990a). The permanently submerged sediments of most lowland lakes and ponds (the “hydropsammal” according to Wiszniewski 1934) usually consist of fine sand rich in organic particles and silt, often covered with plants. The concentration of dissolved inorganic and organic substances in this sediment is often 40–50% higher than in the overlying lake water. Here, burrowing macrofauna and epifauna largely prevail; the meiofauna is restricted to the uppermost centimeters. Nematodes usually dominate, followed by crustaceans (cyclopoid copepods, ostracods), often rotifers, (juvenile) oligochaetes, and chironomid larvae. Tardigrades occur irregularly, but can locally reach a high abundance (Neel 1948; Holopainen and Paasivirta 1977). All other taxa are considered insignificant. During aestival warm-water conditions, the fauna in the subsurface layers is frequently exposed to oxygen deficiency and the formation of hydrogen sulfide. In deeper oligotrophic lakes in both the mountains and lowlands the ecological situation is different. The amounts of organic matter in the profundal bottom sediments of these lakes are much less; oxygen is present even in the warm season while temperatures remain low. Further comparisons indicate that the ecosystems of very large and deep lakes (sensu Beeton 1984) often differ from those of smaller ones (Särkkä 1996b; Kurashov 2002). While in ponds and small lakes adverse
8.7 Freshwater Biotopes
345
factors such as pollution and/or oxygen deficiency can easily and rapidly affect the whole benthic ecosystem, in large lakes these remain local problems that are often compensated for due to mixing with other water bodies. The moist shore sediments above the water level (“hygropsammal,” Wiszniewski 1934) usually comprise a sandy belt of 1–3 m width with a grain size composition that depends on the exposure and the slope of the shore, with the more exposed sites having medium sand (Md > 250 µm). This zone, which is well supplied with oxygen and rich in detrital food because of the debris washed ashore, is richly populated by meiofauna. Pennak (1940) found that rotifers sometimes dominated the psammofauna with extreme densities (>10,000 ind. 10 cm−3). Next in abundance were a few species of harpacticoids (e.g., Parastenocaris; Phyllognathopus ), represented by numerous individuals. In addition, nematodes, oligochaetes (including aeolosomatid annelids) and tardigrades belonged to the ecologically dominant groups of these lacustrine shores; all others were considered less important. Meiofaunal compositions in lakes vary considerably, mostly depending on the trophic status, the profundal oxygen content, and the size and depth of the lake. Water depth and eutrophic conditions are negatively correlated with species richness and trophic group diversity. About 50 nematode species (but sometimes >100) can be expected in an oligo- or mesotrophic lake. Strayer (1985) reported from the shore sediments of Lake Mirror (USA) that 70% of all meiobenthic animals were nematodes, accompanied by turbellarians (often found only occasionally), gastrotrichs, cladocerans, copepods, just a few rotifers and tardigrades, and hardly any harpacticoids. A similar composition with a dominance of nematodes (between 70 and 189 ind. 10 cm−2 = 77%) but a relatively high share of rotifers (up to 21 ind. 10 cm−2) was recorded from oligotrophic lakes in Southern Germany (Bergtold and Traunspurger 2004, 2005; Peters et al. 2007). In a seasonal study of a lake receiving thermal effluents from a nuclear plant in South Carolina, USA, Oden (1979) recorded 400–3,047 nematodes and 88–740 rotifers 10 cm−2 in the control (cold) section; these densities were reduced in the effluent-affected (warm) areas of the lake. Oden found that rotifers, with 372 species from 19 families, were second in abundance to nematodes. In a eutrophic Chinese lake (total meiofaunal abundance between 15 and 400 ind. 10 cm−2) nematodes prevailed, followed by crustaceans and occasionally also rotifers or oligochaetes (Wu et al. 2004). Other eutrophic lakes contained 40–6,000 nematodes 10 cm−2. Upon compiling the various data on the predominance of nematodes in different lakes it would appear that, in sediments with a high organic load, nematodes can reach densities comparable to marine sites and dominate as long as the sediment is oxic. Copepoda, mainly Harpacticoida, are numerically and ecologically second in abundance in many lakes (around 50 ind. 10 cm−2 in Lake Brunnsee). In oligotrophic lakes at northern latitudes with a good oxygen supply and moderate organic input, their share was relatively high compared to nematodes (max. 250 ind. 10 cm−2; Sarvala 1998). Together with cyclopoids (20 ind. 10 cm−2) they were especially common in the periphyton and in shallow sediments. Towards the deeper profundal, harpacticoids decreased in abundance, while nematodes gained in importance. The total abundance of the meiobenthos in the meso-/eutrophic Upper
346
8 Meiofauna from Selected Biotopes and Regions
Lake Constance (Germany/Switzerland), which has a typical, nematode-dominated meiofauna composition, was 550 ind. 10 cm−2 as a mean value, while the Lower Lake Constance contained 800 ind. 10 cm−2. Maximal meiofauna density in this large lake was almost 2,000 ind. 10 cm−2 (Kurashov, unpubl.). The rich ciliate fauna is not usually taken into account in assessments of meiofaunal abundance. Pennak (1939) counted about 10,000 protozoans compared to only about 500 metazoans per 10 cm2! Strayer (1985) quantified the whole spectrum of lacustrine meiofauna and related it to macrofauna. He found the lake bottom was on average inhabited by 1,200 ind. 10 cm−2, 60 times the number and one-third the biomass of the corresponding macrobenthos. Nutritionally, most of this fauna fed on diatoms, including the food-selective species like some rotifers. Diapausing resting eggs of planktonic copepods (“zooplankton egg banks”) have also been assumed to be a food source for meiobenthos, an interesting aspect of bentho-pelagic coupling (Hairston and Kearns 2002). The large majority (80%) of the meiobenthos was grazed down by the predaceous larvae of Tanypodida (Diptera) and other insects. In the oligotrophic Lake Brunnsee (Bergtold and Traunspurger 2005), meiofauna contributed only 0.1 g C m−2 or about 4% of the total biomass (for comparison, the proportion of macrobenthos was 40%). In Finnish oligotrophic lakes this low biomass of meiofauna was surpassed by the biomass of benthic littoral copepods alone (15–30%), and with increasing depth and diminishing macrofauna, the copepod biomass increased to up to 40% of the total meiofauna biomass (Sarvala 2002). This shift is in line with the general trend that meiofaunal biomass in food-limited freshwater environments may exceed that of macrobenthos (Strayer 1991; Kurashov 2002). On a general scale, this wide range is reflected in the biomass contributions of meiofauna to the total zoobenthos biomass vary greatly (7–6%, Morgan et al. 1980). Compared to streams, the meiofauna biomass in lakes is on average 25% higher, and in certain favorable lacustrine sites it is >50% (Hakenkamp et al. 2002). Since calculations of annual production and biomass/production (B/P) ratio are strongly dependent on fluctuating factors (individual body mass, temperature, nutrient supply; see Plante and Downing 1989), the production values for lake meiobenthos in relation to macrobenthos and total zoobenthos vary widely in the literature. While production figures will be given in Sect. 9.3.2, the contribution of the meiobenthos to the overall production of the lake and its comparison with the value for the macrobenthos are of interest here. In many oligotrophic lakes the meiobenthic production is 2–4× lower than the production of macrobenthos. If microorganisms are included in this calculation, the total benthic production exceeds 12× the production of meiofauna (Bergtold and Traunspurger 2005). Again, under food limitation, the relative importance of meiobenthos may increase considerably (Sarvala 1998). According to Kurashov (2002), in large lakes the contribution of meiobenthic to macrobenthic production is on average 50–60%, while in the profundal of smaller lakes with scarce macrobenthos the meiobenthic production increases considerably and can surpass that of macrobenthos by a factor of >10 (small lakes in Latvia). However, under the impact of eutrophication, the macrobenthos gained in abundance while the meiofauna lost
8.7 Freshwater Biotopes
347
its productive relevance and became negligible. Compiling data from various lakes, Hakenkamp et al. (2002) concluded that the biomass and production of lake meiofauna will be highest in fine sand containing a limited amount of detritus where the meiobenthos is exposed to a limited predation pressure and relatively frequent disturbances. Considering the divergence in generation time and metabolic needs between meio- and macrobenthos, the productive turnover, measured as the mean P/B ratio, perhaps better underlines the ecological role of lake meiofauna. P/B ratios > 10 are reported in Bergtold and Traunspurger (2005) for the oligotrophic Lake Brunnsee; in other lakes P/B ratios of up to 37 (Bergtold and Traunspurger 2006) suggest a high importance of meiobenthos in the ecosystem. However, in numerous less productive northern lakes with slower growth, the P/B ratios did not exceed five regardless of their trophic classifications (Kurashov 2002). Sarvala (1998) computed a ratio of only 1–7 for harpacticoids. These values are in the same range as those for marine meiobenthos (see Sect. 9.3.2). Using biomass as a basis, Strayer (1986) analyzed the size spectra of benthos from various limnetic biotopes, comparing it with corresponding figures by Schwinghamer (1981a) for the marine benthos. Regardless of biotope (lakes and streams) or sediment type, he could not group the limnobenthos into separate units of micro-, meio- and macrobenthos. This remarkable difference between marine and freshwater benthos was also confirmed by lotic freshwater studies (Traunspurger and Bergtold 2006; see Sect. 9.2 for explanations). Calculation of size spectra is now an accepted way of describing the holistic functioning of ecosystems and assessing the ecological relations of benthic groups and their metabolic roles (Hakenkamp et al. 2002). But, as Strayer (1991) cautioned, many factors other than grain size also have a potential influence on the size structure of the benthos, such as the degree of physical refuge or the predation pressure. These multifactor relations will cause considerable ambiguities in interpretation (see Sect. 9.2). Regarding nematode feeding groups, it seems that deposit feeders (mainly bacteria) dominate the littoral in all lake types. In oligo-/mesotrophic lakes, the range of chewers (predators and omnivores) increases considerably, especially at profundal depths. Suction feeders living mainly on plants and fungi are of (limited) importance and occur mostly in the littoral region (Traunspurger 2002; Moens et al. 2006). This corroborates the findings of Wu et al. (2002) that the greatest variety of feeding groups and nematode species (20 spp.) were found in the least nutrient-rich area of an otherwise eutrophic Chinese lake. Regarding copepods, oligotrophic northern lakes were inhabited by some 20 harpacticoid species and 30 cyclopoid species (Sarvala 1998). Data on lacustrine meiofauna are difficult to generalize, because each lake with its environment is a restricted biotope of its own. Its meiofaunal composition and ecological role in relation to other faunal compartments will differ, being subject to a variety of controlling influences. The conditions, and therefore meiofaunal functioning itself, depend much on seasonal fluctuations, which are influenced by the geographical location and the climatic situation of the body of water
348
8 Meiofauna from Selected Biotopes and Regions
studied. In contrast to many marine habitats, the relatively small water bodies of lakes are directly subject to these factors. Thus, performing quantitative comparisons of different lakes and sampling periods, beyond considering different sampling methodologies, is a problem. More detailed reading: Wiszniewski (1934), Pennak (1940), Strayer (1985, 1991), Sarvala (1998); Traunspurger (2000); Kurashov (2002); Eyualem-Abebe et al. (2006).
Box 8.8 Freshwater Meiofauna—A Parallel World: Similarities, Dissimilarities and Transitions to the Marine Realm Hydrodynamics, hydrochemistry (particularly oxygen), disturbances, food supply, and competition: the same key factors control both freshwater and marine meiofauna. And yet there are far-reaching dissimilarities—differing methods, research histories and distributional pathways. Studies of community dynamics and functional structure suffer from impediments to adequate sampling, especially in streams and groundwater aquifers. The unique and barely accessible world of groundwater animals is sometimes visible in cave waters, where “troglobites” closely correspond to the “stygobites,” the groundwater fauna. Groundwater studies offer unique opportunities to reconstruct zoogeographical connections and evolutionary pathways. How can we explain the meiofaunal paradox: a wide distribution of animals with almost no dispersive capacity? For an adaptive fauna (oxygen, salinity), the coastal groundwater, river mouths or marine caves are believed to be ports to the continental groundwater aquifers. The reduced competitive pressure in the groundwater or cave habitats enables the survival of relict and often primitive taxa. Thus, groundwater studies can help us to understand evolutionary trends linking marine and freshwater meiofauna. The composition of the freshwater meiofauna is quite different from the marine world. Nematodes often lose their dominance while harpacticoids take the lead; insect larvae are important temporary meiofauna, both as detritivores and predators; also the predacious cyclopoid copepods have no equivalent in marine environments. Running freshwaters (streams, creeks, rivers) are favorable sites for (experimental) studies on meiofaunal drift, emergence, and recolonization. Parallels can be drawn from their unidirectional flow regimes to the more complex multidirectional hydrodynamic patterns in the sea. In lakes, the composition and ecology of the meiobenthos vary strongly with the physiography and chemistry. Aside from the nematodes that usually dominate, other groups such as rotifers, chironomid larvae or oligochaetes can locally and temporally become abundant. Copepods may prevail in the oligotrophic profundal of deep lakes. The lacustrine meiobenthos, despite its low biomass can be a highly important producer compared to the macrobenthos.
8.8 Polluted Habitats
8.8 8.8.1
349
Polluted Habitats General Aspects and Method Survey
The assessment of ecosystem health using meiofauna was an innovative approach when the first edition of this book was written. This changed with the review by Coull and Chandler (1992). This major compilation of the field still provides an important foundation and will most likely serve this function in the future. Success in pollution research means convincing laymen and critical opponents, as well as obtaining research investment, for the sake of the health of the environment’s health. Impact studies using meiofauna have now been accepted by international governmental agencies because the advantage of using meiofauna is obvious. Today, the use of using meiofauna is a widely acknowledged method of assessing the environmental status of a biotope. What is the basis for this paradigm change? Monitoring programmes and case studies performed after both natural and manmade disturbances have shown that the resolution can be improved by studying meiobenthos because of its commonly higher sensitivity and turnover compared to those of macrobenthos. Meiofauna are speciose, abundant and ubiquitous; meiofauna are even represented in extreme areas where macrofauna become scarce; and meiofauna are relatively cost-effective to handle. The multitude of anthropogenic impacts and ecotoxicological agents and the diversity of meiofaunal investigations from polluted habitats is overwhelming and, in a treatise like this, forces reductive concentration. Hence, we will first discuss general aspects and problems. This will be followed by sections on the impact of petroleum hydrocarbons, metals and pesticides, three priority categories, as the almost 300 marine meiofaunal studies on the impact of pollution demonstrate (see Coull and Chandler 1992, plus more recent compilations). Oil obviously has a strong impact, but recovery is often surprisingly fast. Metals are persistent and recovery by metabolic degradation is limited, so long-term effects and subtle changes to community composition are common. Pesticides share both features: some are highly toxic but biodegradable and the meiofauna exposed have a substantial recovery potential. The more persistent and/or adsorptive pesticides cause long-lasting community effects that require special attention. Impact studies on freshwater meiofauna have been compiled by Traunspurger and Drews (1996), Särkkä (1996a) and Höss et al. (2006). The comprehensive body of information in these papers and their bibliographies allow one to focus here mainly on the marine realm. A well-structured survey of the advantages and problems associated with using meiobenthos in pollution studies is presented by Kennedy and Jacoby (1999). The underlying mathematical principles and details of calculation will not be covered here. The reader is referred to the competent accounts by Underwood and Chapman (2005) and to references cited below. The effects of pollution by anthropogenic organic enrichment on meiobenthos will not be detailed here, although they do occur universally, especially in coastal sites and freshwater bodies. The compilations mentioned above should be consulted
350
8 Meiofauna from Selected Biotopes and Regions
here. The general effects in this field are rather uniform and well investigated: an initial enrichment in meiofaunal numbers in the eutrophicated area is linked to a decrease in diversity and an increase in the dominance of a few species, mainly rstrategists. Most harpacticoids disappear before nematodes. As oxygen concentrations decrease, all of the physiologically demanding species will drop out. Development of hydrogen sulfide will exterminate all but the thiobiotic species, which can persist by feeding on the rich supply of bacteria and decaying animals and plants (see Sect. 8.4). While pollution by organic anthropogenic wastes is today of minor relevance for most marine sites, it still devastates many freshwater habitats. An increasing threat to the marine coastal benthos, however, are side effects of the rapidly growing aquaculture industry. Especially in the early phase, aquaculture plants in many sheltered bights caused heavy depletions in the (meio-) benthic assemblages. Today, with better knowledge and sophisticated, controlled feeding methods, it is shown that the deterioration can be largely reduced and localized to small areas that will become recolonized relatively quickly (see Sutherland et al. 2007). As Heip (1980b), Hicks (1991) and Warwick (1993) point out, there are a suite of indisputable advantages to using meiofauna rather than macrofauna in pollution studies: -
Widespread occurrence, high availability Permanent and intimate contact with contaminated sediment Superior sensitivity and rapid reactions allow for short research periods High abundance, even in small sites or usually macrofauna-impoverished biotopes (estuaries, exposed beaches, high organic loads), allows for reliable statistical evaluation (e.g., Josefson and Widbom 1988, Austen and Widbom 1991) High species richness allows evaluations of changes in community structure Indicator species are widespread and present in various taxa Short generation cycles allow for tests of sensitive reproductive stages Cost-effective experimental and field work Low sensitivity to mechanical disturbance of the sediment enhances the possibility separating mechanically induced and pollutant-induced impairment (Austen et al. 1989, Warwick et al. 1990a).
These advantages of using meiobenthos for impact studies stand against some inherently “weak” arguments that seem to suggest the superiority of the macrobenthos as a pollution indicator: - Little previous information: Baseline studies and time-series observations on meiofauna prior to the pollution event are rare (e.g., Herman et al. 1985; Bodin 1988). Extrapolating from “similar” biotopes is a deceptive approach because of the inherent patchiness and variability of meiofauna. Long-term sampling and monitoring is needed. - Unimpressive size of the affected fauna: Pollution studies with inconspicuous microscopic animals pose a problem to the public (see Box 9.3)—they cannot see them! The choice of the relevant indicator species is crucial.
8.8 Polluted Habitats
351
- High sampling frequency: The high temporal and spatial heterogeneity of meiofauna requires detailed spatial and temporal coverage. - Difficult identification: Only instructive, easy-to-use and computer-based pictorial keys can help in the identification of fauna as diverse and small as meiofauna. Some additional achievements have recently promoted the use of meiofauna in impact studies: - Standardized bioassays acknowledged by national and international agencies for general use have been designed (Chandler 2004; ASTM 2004; Bejarano et al. 2006a). - Electronic identification keys exist for major meiobenthic groups (Diederich et al. 2000; Wells 2007). - Pollution-specific manuals facilitate evaluation (Somerfield and Warwick, 1996). - Computation is supported by reviews (e.g., Neher and Darby 2006) and software programs. - New indices using meiofauna have refined pollution analyses in both freshwater and marine habitats (Särkkä 1996a; Neher and Darby 2006; Vassalo et al. 2006). - Genetic-based ecotoxicological studies of meiofauna material have become a priority research field (Staton et al. 2001; Kammenga et al. 2007). Thorough research on the impact of pollutants should generally evaluate the intrinsic toxicity, as evidenced by the immediate responses, and the recovery potential of the community. This requires a combination of (1) chemical analyses, (2) field studies, and (3) laboratory experiments. This “triad” in general pollution research (Long and Chapman 1985; Chapman 1986) obviously also applies to impact studies performed with meiofauna. The main strategies (all of which require evaluation by careful statistical analyses) are: 1. Chemical analyses of the pollutant and its concentration 2. Field studies of the meiofaunal assemblage with time-series observations (case studies after incidents, large field experiments) 3. In-vitro laboratory assays of toxicity thresholds in the aqueous phase (unifactorial, multifactorial) using test animals (acute toxicity with representative species, long-term impact through generations with culturable test organisms) 4. Mesocosm studies in containers with sediment, agar (nematodes) or artificial substrates, either applied with a toxicant or transplanted into the polluted environment In addition to the traditional surveys performed after pollution incidents, largescale field experiments have recently been performed (Schratzberger and Warwick (1998 a,b). Furthermore, standardized experimental field approaches have been designed (Mirto and Danovaro 2004). Despite their complex interpretation, the results from these studies in the natural marine environment are probably more generalizable than single-factor tests performed in the laboratory.
352
8 Meiofauna from Selected Biotopes and Regions
Many aquatic pollutants are sorbed to the sediment particles and so there are typically higher concentrations in the sediment than in the overlying water. This would suggest that sediment-associated tests yield more realistic information for meiofauna, considering their intimate contact with the sediment (Kovatch et al. 1999; Schratzberger et al. 2000; Chandler and Green 2001). However, toxicants in sediments have typically proven to be less toxic than their counterparts under “inwater conditions” (e.g., Austen and McEvoy 1997b; Green et al. 1993 for metals). In the “sediment situation,” many toxicants may not be bioavailable to meiofauna, their toxicities having been reduced by chelation or complexation, or by binding to organic ligands and colloidal aggregations on the sediment surface. Hence it is the bioavailability, not the absolute concentration of pollutants measured in water, that determines the noxious effects on the benthic environment. In nature, where single factors often act synergistically, the multifactorial impact of pollutants interacting with natural stressors (salinity, temperature, perturbation, high organic load) is usually more detrimental than the effect of adding the impacts of each individual toxicant (but there have been exceptions where the combined effect was less severe than the additive one). Hence, multiple stress experiments (e.g., low salinity plus elevated metal concentrations) will achieve more representative results. Another facet that illustrates the complexity of toxic interactions is that the pollutant toxicity can be modified by the addition of dissolved organic matter (DOM) (Höss et al. 2001; Bejanaro et al. 2005). One specific advantage of using meiofauna in pollution studies is their short generation times, since multigeneration tests are often needed to obtain reliable results on long-term community effects, which are cryptic at first sight. Sometimes, shortly after a pollution event, the overall abundance of meiofauna may increase (depending on the species, the nature and concentration of the pollutant, etc.), superficially suggesting a negligible impact. This could easily lead to erroneous rash conclusions about a non-detrimental effect. Often it is long-term parameters, such as lowered reproductive rates, decreased taxonomic diversity and a reduced ecological diversity, that provide the most severe consequences for the community. It is generally more informative to concentrate on the highly sensitive reproductive output parameters (clutch size, larval stages) than to focus on adult abundance and biomass (Giere and Hauschildt 1979; Bejarano et al. 2006a,b, for oil pollution; Chandler 1990; Bejarano et al. 2004, for pesticide contamination). Toxicity tests with developmental stages for pollution assessment require laboratory cultures, experiments and standardized bioassays with sensitive and preferably widespread and easily achievable common “indicator species.” But are there any easy-to-recognize species? And can we perform life cycle bioassays with them? So far, experiments with about 50 continuously lab-cultured meiobenthic species (among them about 15 harpacticoid copepods and some 30 nematodes) have been performed. The harpacticoids in particular combine sensitivity with frequency (mainly Amphiascus tenuiremis, Microarthridion littorale, Nitokra sp. and Robertsonia sp.). The culture of marine harpacticoid species as pollution indicators is most advanced in the USA (Chandler 1986; Chandler and Green 2001; Bejarano et al. 2006a,b). Tisbe sp. and Tigriopus japonicus are often used in Europe or in the Pacific area and new representatives are added regularly. Chandler’s working
8.8 Polluted Habitats
353
group succeeded in getting standardized bioassays officially licensed by the American Society for Testing and Materials (ASTM 2004), and (more recently) globally licensed by the OECD. Among the marine nematodes, Chromadora spp., Chromadorita tenuis, Pellioditis marina and Diplolaimella spp. are classical targets for culture (Tietjen and Lee 1984; Jensen 1983; see compilation by Moens and Vincx 1998); nematodes cultured for bioassays often belong to Monhysteridae. Meiobenthologists have also elaborated culture methods for freshwater nematode species (Panagrellus redivivus, Caenorhabditis elegans; see Traunspurger and Drews 1996; Heininger et al. 2006; Höss et al. 2006). Bioassays allowing to monitor the environmental impacts of noxious events on communities and their reestablishment integrated into the natural multifactorial network require standardized substrate conditions (e.g., bottle brushes, plastic sheets) (Mirto and Danovaro 2004). With the progress in technology, modern analytical procedures do not require large amounts of biomass, which was an obstacle that often disadvantaged meiofauna with respect to macrofauna. Bioaccumulation studies can be performed by analyzing tissue extracts in the microgram range (Wirth et al. 1994; Haitzer et al. 1998); changes in the genetic backgrounds of populations exposed to chemicals can be detected by “ecotoxicogenomic” screening (Watanabe and Iguchi 2006); the DNA barcoding of whole populations on a few slides (Bhadury et al. 2006a) offers the chance to exploit their biodiversity, which is often reduced as a response to pollution and environmental depletion (see below). Rapid in vivo effects of toxicants on meiofauna and their offspring can be documented and evaluated semiquantitatively under a confocal laser scanning microscope, thus linking the huge potential of fluorescence labeling to bioassay techniques (Chandler and Volz 2004). Ecophysiological parameters such as micro-respiration measurements can be used to derive the status of physiological resistance (Moens et al. 1996). Studies on meiofauna from many regions suggest that stress caused by disturbances, organic wastes or toxicant contamination causes a decrease in the various types of diversity (Warwick 1988b; see Sect. 9.1) and an increase in dissimilarity compared to reference communities. This deviation from “normal” is indicated by similarity indices that compare differences in the species compositions of two sites (Sørensen Index, Jaccard Index, Hill’s Index). Dominance indices (e.g., Simpson’s Index) relate the total abundance of species in samples (communities) to single species abundances. The Shannon diversity relates the observed species number to the (expected) species number, while the evenness considers aspects of how the individuals in the sample are apportitioned among species (distribution/aggregation). Biodiversity (taxonomic diversity in its various forms, such as weighted diversity) measures the number of taxa (species) per habitat/sample. All of these “classical” indices have a distinct information value and are used in pollution studies (see the good overview in Neher and Darby 2006). Their calculation is not detailed here, since such explanations can be found in numerous ecology textbooks and software programs; for comments and restrictions see Heip et al. (1988). Recent improvements of these indices also consider, in a probabilistic approach, those species present at the sites, but unseen in samples (“expected number of species-index” or “improved Shannon index”). Thus, they reduce a considerable bias in many indices.
8 Meiofauna from Selected Biotopes and Regions
Shannon-Wiener function (bits ind.−1)
354
5
Copepod species Copepod genera
4
3
2
1
0 6 north
4
2 2 0 Distance from dump centre (km)
4 south
Fig. 8.18 The decrease in harpacticoid copepod diversity related to the distance from a polluted site in the Firth of Clyde, Scotland. Note the corresponding curves from evaluating species and genera. (Moore and Bett 1989)
Thus, they allow for more refined results than do the routine procedures (Chao and Shen 2003; Chao et al. 2006). Most meiofaunal studies have demonstrated that pollution results in causes marked changes in diversity (Fig. 8.18 ). Lambshead et al. (1983) introduced into ecology a method of assessing stress through pollution, the “k-dominance method.” An updated method, the “ABC method” (Warwick, 1986; Warwick et al. 1990a) relates abundance to biomass in two comparative curves. Both calculations are based on the assumption that in undisturbed biotopes the more K-selected specialist species (persisters) account for high individual biomass, even though they have low population densities and thus only a low numerical rank. In contrast, communities of r-selected generalists (colonizers), which dominate in disturbed areas, are typically characterized by just a few species that exhibit low individual biomass but large population density, thus attaining a high numerical rank (Fig. 8.19 ). (The general meanings of the character complexes r- and K-strategists are reviewed in Parry 1981.) Now, if the species are plotted on a log-scale-abscissa according to their rank, and on the ordinate according to their cumulative percentage dominance, the result can be highly indicative when comparing the curves for abundance and biomass. A disturbance (e.g., due to pollution) can be postulated if the abundance curve is elevated above that the biomass curve. If the biomass curve exceeds that of the abundance, the site can be considered undisturbed (Fig. 8.19). Essentially, this method is a graphical comparison of the two components of diversity, species richness and evenness (Platt et al. 1984). Although this convenient graphical method only allows a rather coarse discrimination of a few levels of pollution, it does not require control samples: the two curves represent an “internal control.” Warwick et al. (1990b) demonstrated that
355
Cumulative %
ss ma
Bio
ce
n nda Abu
ce an nd u ss Ab ma Bio Moderately polluted
Unpolluted
Cumulative %
Cumulative %
8.8 Polluted Habitats
e nc da n bu
A
s
as
m Bio
Grossly polluted
Species rank
Fig. 8.19 Hypothetical k-dominance curves for areas polluted to varying degrees. (Warwick 1993)
the k-dominance method can also be applied to meiofauna as a valuable indicator of disturbances despite the minute divergences in biomass between meiobenthic rand K-strategists. On the other hand, this study also revealed the inherent problem of interpretation: depending on the taxon, each kind of disturbance has a different impact and causes a different curve shape; e.g., nematodes are clearly affected by sediment disturbance (bioturbation), harpacticoids less so. Many of the characters listed above should make meiofauna an ideal tool for studying changes in benthic ecosystems. However, there is one major impediment: for optimal information value, many of the above indices require identification to a low and uniform taxonomic rank (species, genus). Even with the necessary instrumentation and literature, this thorough identification requires manpower, time and experience. To circumvent this problem, approaches based on “taxonomic minimalism” have been developed, which require bulk recognition to higher taxa only, and can be performed (with some guidance) by untrained personnel too. However, is too much informational value lost if we simply compare characteristic groups? Does the high ecological differentiation in meiofauna allow for a summative identification of higher meiobenthic taxa? Large-scale and comparative meiofaunal studies have shown that data based on taxonomic ranks as high as families still allowed a fairly clear separation of stations according to their degree of disturbance/pollution (Heip et al. 1988; Herman and Heip 1988; Warwick 1988a; Heininger et al. 2007). Especially in nematodes, data - aggregated to higher taxonomic ranks - yielded a fairly consistent discrimination between sites. The frequently used “maturity indices” (Bongers 1990; Bongers et al. 1991) characterize the relation between the “quality” and the stress situation of an environment. Based on species, genera or even families, they rank the taxa according
356
8 Meiofauna from Selected Biotopes and Regions
to their life history characteristics into various degrees of “persisters” or “colonizers.” Members of these categories are known to be differently sensitive to stress or pollution. In disturbed sites the index is low; in areas with little or no disturbance the values are high. Upon applying this maturity index to various areas stressed by an oil spill, an overload of organic matter, heavy metal contamination, etc., it was shown that the index is widely applicable for the discrimination of stressed from unstressed biotopes in terrestrial, freshwater, marine, deep-sea and tidal flat habitats (see Essink and Romeyn 1994). With the increasing reliability of the allocation of taxa to stress classes, this method could provide a relatively simple means of assessing biotope disturbance. However, Heininger et al. (2007) showed that it can give results that conflict with the conclusions drawn from the chemical analysis of the same habitat. Modifications of the classical maturity index incorporate feeding groups, which, again, reflect environmental variables (see Neher and Darby 2006). Another approach that attempts to simplify working with meiofauna when investigating the effects of pollution is to focus on characteristics of the two most abundant taxa, nematodes and harpacticoids. In cases of organic enrichment, and often also chemical pollution, nematodes are on average more stress-enduring than harpacticoid copepods. This often observed ecological discrepancy is the basis for the “nematode/copepod ratio” (N/C index; Raffaeli and Mason 1981), proposed as easily applicable means of measuring the impact of pollution. The authors contended that this trend would be generally valid for sandy eulittoral coasts. However, the N/C ratio also turned out to be sensitive to natural environmental variables (grain size, water content), which brought its general reliability into question (Coull et al. 1981; Lee et al. 2001). Nevertheless, the ease of calculation made this ratio one of the most applied but also debated pollution indices in the field of marine meiobenthology. Raffaeli (1987) partly corrected his earlier oversimplified contention. What is the basis for the different ecological reactions of the two dominant meiofaunal taxa, and in which cases did the N/C ratio yield questionable results and thus require refinement? Harpacticoids comprise two major groups with different substrate-related life histories. The mesobenthic (interstitial) species live in sand, while their epibenthic or endobenthic counterparts instead prefer water-saturated fine sand and mud. Nematodes seem less dependent on sediment structure. This difference in substrate-relations is enhanced by the differences in the trophic niches of the two groups. Most nematodes are linked to a short, detritus/bacteria-based food chain. Hence, with organic enrichment of the site (generally typical of finer sediments), their abundance will usually increase even if the oxygen levels sometimes decrease (Fig. 8.20, compare Essink and Romeyn 1994). In contrast, harpacticoids are mainly microalgae-based and oxygen-sensitive members of the food web. The interstitial subgroup of copepods will react negatively to an increase in the organic load (compare Rudnick 1989; Vincx et al. 1990) with concomitant depletion in oxygen and clogging of the void system by debris. This scenario would explain the divergent N/C reactions of the taxa in many cases of organic pollution. Various studies (e.g., Sandulli and De Nicola 1991) that performed experimental work alluded to confirmed the different reactions of the two meiofaunal groups.
8.8 Polluted Habitats
357 Nematodes
org
ani c
pol
luti
on
ution
al poll
Abundance
chemic
epi- and endobenthic Copepods
mesobenthic (interstitial) Copepods
polluted
clean
Fig. 8.20 The divergent response of nematodes and copepods (harpacticoids) to a pollution gradient. For further explanation see text. (After Raffaeli 1987)
However, there are still interpretations that are not easily explained and that reduce the general applicability of the N/C ratio: the endo/epibenthic subgroup of harpacticoids is not greatly impaired by enrichment of organic matter. It may even initially increase in abundance with the increased food supply. This questions the indicative value of the N/C index if one generalized taxon “benthic copepoda” is used. Therefore, Shiells and Anderson (1985) suggested restricting the calculation of a refined N/C ratio to just the interstitial (mesobenthic) harpacticoids. Moreover, in cases of chemical pollution, the nematode populations can also decrease, so that the curves for both groups are similar and are not indicative. It appears that a refined version of N/C ratio is useful in cases where the effects of a pollution incident in a restricted area and the recovery phase are monitored over time without detailed identification (Raffaeli 1987).
358
8 Meiofauna from Selected Biotopes and Regions
When not over-interpreted and not universally applied, the community indices outlined above, regardless of some criticism related to their insufficient consideration of ecosystem interactions, still appear to be useful and indicative tools for investigating many man-made changes in (meiofauna) community structure. The taxonomic distinctness of a community—the degree of relatedness of the taxa within a sample—has been transformed into an index, the TDI index (Warwick and Clarke 1995, 2001; Clarke and Warwick 1998). This index incorporates biological and ecological aspects as well as the distribution pattern. With increasing stress and decreasing habitat heterogeneity, the community will become more monotonous and the taxonomic distinctness will tend to decrease (Barbuto and Zullini 2005). For practical work it is important that the TDI is independent of the sampling success during data collection, while the other indices are strongly influenced by the number of sampled taxa. While many of the procedures outlined above also are applicable to freshwater field studies to evaluate anthropogenic impact, some specific and simple indices may help to avoid the need to evaluate all fauna. For northern lakes, Särkkä (1996a) contends that characteristics of the meiofauna such as easy access, large diversity and abundance, as well as its reduced seasonality, would support the use of meiofauna compared to macrofauna in assessing pollution. As easily recognizable indices, he proposed the numerical relation of permanent to total meiofauna, the ratio of aeolosomatid annelids to oligochaetes, and the percentage of naidid oligochaetes to total oligochaetes. For river-beds Zullini (1976) focused on nematodes and showed that the relation of the more tolerant Secernentea to Adenophorea might be a good pollution indicator. A valuable overview of the methods used in freshwater environmental science is given by Höss et al. (2006). Perhaps the biologically most meaningful method of linking community data to an environmental impact such as pollution is application of multivariate statistics. In many independent papers on meiofauna it has been shown that multivariate analyses (e.g., classification using the Bray–Curtis dissimilarity and multidimensional scaling ordination, MDS) are more sensitive approaches to reveal disturbances and impact events than the univariate indices (Gray et al. 1990; Austen and Somerfield 1997; Schratzberger and Warwick 1999b; Wetzel et al. 2002; Saunders and Moore 2004; Heiniger et al. 2007 for freshwater). Although the computational basis of multivariate statistics is fairly complex, especially in the case of the multidimensional scaling method (Field et al. 1982), the yields are an easy-to-conceive graphical document: a “map” of sample similarities influenced by environmental variables, e.g., pollution (Fig. 8.21). The superior sensitivity and the general applicability of MDS-method have proven valuable in numerous examples related to the macro- and meiobenthos, and for both abundance and biomass values (Warwick et al. 1993a; Somerfield et al. 1994). Moreover, MDS ordinations allow for a “taxonomic minimalism.” Calculations based on ranks above the species level (genera: Somerfield and Clarke 1995; families: Herman and Heip 1988; Warwick 1988a; Christie and Berge 1995) result in essentially similar patterns (Fig. 8.22), although a loss of discrimination/ information with coarser resolution is evident (Quijon and Snelgrove 2006; but see
Single indices (e.g. diversity, species richness, evenness
univariate methods
Species
Samples distributional techniques
Cumulative %
8.8 Polluted Habitats
359
k-dominance plots
Species rank
multivariate methods
Dendrogram Similarity matrix
1
28 4 3 7 6 5 5
6 8 7
MDS plot
24 13
Fig. 8.21 Schematic procedure for various statistic calculations including MDS scaling. (Courtesy M. Schratzberger)
Species
Genera 4 44
4
5
4 4
666 5 5
5 6 66 55
Families 4 44
6 6 6 55
5
Fig. 8.22 Multidimensional scaling ordination (root-transformed abundance) for copepod data from the Firth of Clyde, Scotland. Samples from unpolluted sites are stippled, those from polluted sites are in bold black. The use of different taxonomic ranks results in largely the same sample configurations. (Warwick 1988)
360
8 Meiofauna from Selected Biotopes and Regions
Carman and Todaro 1996). Nematodes seem especially robust to the grouping in higher taxa. Hence the MDS- method, often combined with additional analyses, represents a valuable tool for revealing the impact of an environmental variable (e.g., pollution), even in cases where mere diversity assessment could not give a significant answer (Warwick 1988a). Today the statistics in meiofaunal studies on disturbance and pollution are analyzed using convenient software programs (e.g., PRIMER in its various versions; Plymouth Marine Laboratory; http://www.pml.ac.uk/primer/index.htm). This classic in the field of ecological impact studies is available as a comprehensive package containing, among other methods, similarity profiles, Bray–Curtis dissimilarity calculations, cluster analyses in illustrative graphics, and various univariate indices. Training courses for the application of PRIMER are being offered. By using factorial correspondence analyses, interconnections between environmental parameters and meiofaunal community structure can be revealed (e.g., Villiers and Bodiou 1966). Their interpretation, though, requires a good biological understanding and is less easy than the computer-assisted calculation. An easier graphical interpretation is offered by the principal response calculation (PRC), a multivariate method that illustrates the relation between the sampling period and the first principal component responsible for the variance. In canonical correspondence analyses (CCA), the taxon data are compared with various environmental variables depicted as vectors on a biplot. The relative importance of the vector (variable) is then indicated by its length, and its correlation to other, neighboring variables by its angle. Canonical correspondence methods are embodied in the PRIMER package mentioned above, computer assistance facilitates their use considerably. Recent methods emphasizing the functional diversity and complexity of communities sometimes offer a useful option for indicating disturbance/pollutioninduced changes in ecosystem structure. In nematode studies especially, alterations of feeding group abundances are examples of such changes. However, the interdependence of trophic structure and substrate granulometry, which is especially well known for nematodes (see Figs. 5.21, 5.22; Kennedy et al. 1994; Schratzberger et al. 2007b), may hamper a simple interpretation (Höss et al. 2004). Regarding the rapid advance of simple-to-use and graphically attractive mathematical/statistical software, one caveat is appropriate: as easy as it may become to apply statistical programs, the user’s understanding of the underlying principles is also required. Presentations of sophisticated statistics/factor analyses/correlations often give the impression of an erratic trial-and-error-procedure without too much comprehension of the underlying biological processes. Easy application of statistical tests cannot substitute for a lack of understanding, an unclear conception, insufficient background knowledge and a thorough literature inquiry. The various methods used to assess changes in community structure (and community stress) are summarized by Warwick and Clarke (1991), Clarke (1993) and Warwick (1993): 1. Univariate methods, where the relative abundances of the various species are reduced to a single index. The appropriate statistical test is the classical ANOVA; the most frequently used univariate method is the Shannon Wiener
8.8 Polluted Habitats
361
diversity index H’ (Shannon and Weaver 1949), usually combined with calculations of evenness (equality of distribution). 2. Graphical/distributional methods, where the relative abundance (or biomass) of a species is plotted as a curve. The typical example is the k-dominance curve. This group of methods provides more information about the nature and distribution of the fauna than a diversity index. 3. Multivariate methods of classification and ordination, where faunal communities are compared with respect to their specific identities and (relative) quantitative importance. These methods are exemplified by techniques such as multidimensional scaling ordination (MDS). The two first types of methods have a disadvantage: they can yield identical results from communities with completely different taxonomic compositions. They are also less sensitive to detecting community changes than the multivariate methods. However, they can give a fairly clear assessment of the presence of detrimental (e.g., pollution) effects. On the other hand, the third group of analyses, multivariate statistics, document faunal changes with precision and are widely applicable, but they give few indications of the possible reasons. The option to compensate for the shortcomings of the different types of methods and to optimize their specific advantages is the combined use of several methods (Clarke 1993; Dauer et al. 1993).
8.8.2
Selected Cases of Pollution and Meiofauna
8.8.2.1 Oil Pollution Oil spills frequently devastate the environment, especially in coastal regions. Since environmental spilled oil in the environment is dreadfully spectacular, the effects of petroleum hydrocarbons on marine sites are frequently studied. Surprisingly, benthic fauna, including meiofauna, often appear only mildly affected. A long-lasting depletion of the fauna is only rarely reported, even after major oil spills. However, rash conclusions based on the behavior of some robust species only embody a potentially high political and societal risk that there are few effects of oil spills and recoveries from them occur quickly. But have the fauna really recovered? After a dramatic decline in abundance and diversity immediately after the impact, the return of the fauna is quite rapid (weeks, some months). Pioneered by some robust species, this return may misleadingly indicate an environmental recovery. The complex and sometimes controversial impact of hydrocarbon compounds on the environment can only be comprehended by detailed investigations involving field and experimental work, long-term monitoring and acute toxicity tests. Below I attempt to provide a clear picture of the multifactorial effects of oil pollution in the environment. Crude oil, consisting of thousands of chemical compounds with various toxicities and properties, is a natural product to which many meiofaunal species can partially adapt. During the “aging of oil,” the most toxic substances—the short-chained aromatics—evaporate and dissolve very quickly, so the oil remaining in the sediment
362
8 Meiofauna from Selected Biotopes and Regions
is more persistent. Natural oil seeps in the sea are equivalent to organic enrichment (Montagna et al. 1989, 1995; Lee and Page 1997), and the stocks of nematodes are particularly rich here. Even negative impacts of oil platforms appear to be restricted to the direct vicinity of the outlet (Carr et al. 1996). Some of the negative results recorded from areas contaminated with oil were probably caused by the oxygen depletion induced by the degradation of excess organic matter rather than the toxicity of the oil itself (Bodin 1988). For the benthos in subtidal sediments, dissolution of many toxic substances in the water column further reduces the toxicity of natural crude oil; only a small part of spilled oil reaches the bottom (1–13% according to Lee and Page 1997). Boucher (1980) could not find a significant reduction in subtidal meiofauna, even after the huge Amoco Cadiz oil spill (see below). For shallow subtidal diatoms, Suderman and Thistle (2004) also confirmed a lack of significantly noxious effects of fuel oil. In subtidal samples from the Ligurian Sea (Mediterranean), meiofaunal populations after oil contamination returned to normal after only one month (Danovaro et al. 1995). The effects of oil spill accidents on shore life, specifically on meiobenthic communities, are divergent and difficult to summarize because of the different nature of the oil and the local physiographic and climatic conditions, e.g., wave exposure, sediment structure, season and temperature. This complexity makes reliable general predictions about the impact of oil spills almost impossible. In the large oil spill off La Coruña (Northern Spain) in 1976, all of the eulittoral meiofauna on beaches adjacent to the oil outflow were exterminated, and only a few opportunistic species had survived one year after the spill (Fig. 8.23). A similar massive destruction of meiofauna was reported from oil spills off Hong Kong and in brackish water of
Fig. 8.23 The impact of an oil spill on the abundance of nematodes in beaches off La Coruña, Northern Spain. Comparison of two data sets obtained six weeks and one year after the spill. (After Giere 1979)
8.8 Polluted Habitats
363
the Baltic Sea (Wormald 1976; Elmgren et al. 1983). Since spilled fuel oil can also drastically destroy the meiofauna (Ansari and Ingole 2002), the risk of massive destruction does not apply solely to oil tanker wreckages. In less destructive cases, the immediate meiofaunal response to oil spills is usually a strong reduction in harpacticoids, ostracods and turbellarians, and a less severe impact on annelids. The decrease in nematode populations is often only subordinate and difficult to separate from natural fluctuations. Correspondingly, the begin of the recovery phase after a spill is first indicated by nematodes which have a greater ability to survive oil compared to copepods (Christie and Berge 1995). Correspondingly, effects resulting from oil contamination and subsequent bioremediation are often indicated by subtle sublethal reactions of the communities, e.g., by changes of the dominance pattern, altered diversity and evenness (Warwick et al. 1988; Schratzberger et al. 2003). In microcosm experiments with differentially oiled salt marsh sediments and impact periods, Carman et al. (1997) found also changes in the nutrition. Grazing rates of most copepod species were reduced with increased exposure to and concentration of oil (see also Christie and Berge 1995). The Amoco Cadiz spill in 1978 represented a rare case in which the meiofauna (nematodes) had been monitored for years prior to the oil spill. However, at least for nematodes, univariate statistics did not reveal a significant negative impact that was discernible from the disequilibria caused by natural environmental variables. The complexity of the field conditions did not allow for a straightforward interpretation (Bodin and Boucher, 1983; Bodin 1988). Only the sensitive harpacticoids reacted with a decline in abundance and diversity right after the spill. Specifically, the more susceptible juvenile stages of harpacticoids were severely reduced and reproduction was delayed, which caused changes in the population dynamics due to depleted copepodite stages. Only MDS- and ABC-methods were sensitive enough to demonstrate the impact of the oil spill (Warwick and Clarke 1993). After two or three years, the “degradation phase” ended, but according to long-term studies it took almost six years for the meiofauna to recover and re-establish their status prior to the spill (Boucher 1985; Bodin 1988, 1991). Recovery after a spillage starts often with an unbalanced blooming of microalgae within a few months. This first sign is followed by rapid population outbursts of some robust nematodes (e.g., Sabatieria pulchra) and harpacticoids (e.g., Cletocamptus deitersi) accompanied by erratic fluctuations in the dominance pattern. Strong population growth in some species/groups is accompanied by the destruction decrease of more sensitive competitors and predators and supported by a rich supply of microalgae (Fleeger and Chandler 1983; Montagna et al. 1995). These outbursts are often followed by sudden population breakdowns. Especially in exposed, sandy shores, complete recovery of meiobenthic assemblages (measured in terms of diversity and evenness) may be achieved relatively rapidly, sometimes in less than one or two years (Rodriguez et al. 2007). The speed of recolonization depends much on the presence of neighboring donor assemblages (Gourbault 1987). However, in sheltered muddy bights and estuaries, depletion will last much longer due to the long persistence of undegraded toxic substances in
364
8 Meiofauna from Selected Biotopes and Regions
the absence of oxygen in deeper layers. In the Amoco Cadiz spill, the depression of the meiofauna in the ecologically delicate muddy Bay de Morlaix was so pronounced that (based on abundance-related species-rank calculations) it took six years before the meiofauna had recovered (Gourbault 1987). After a smaller oil spill in the Ligurian Sea, Danovaro et al. (1995) could not determine any impact on the nematode populations using the N/C ratio (see also Carman and Todaro 1996; Ansari and Ingole 2002). However, a decreasing diversity is apparently a better indicator, as shown after the oil spill at the Hebridean shores (Moore and Stephenson 1997). Refined multivariate statistics may reveal that these seemingly mild cases of oil pollution might also have a clearly negative effect on meiofaunal communities (Warwick and Clarke 1993). What about cleaning up oil-polluted sites with dispersants? Does bioremediation with fertilizers help? While earlier oil dispersants often acted as additional stressors that enhanced the toxicity of the oil/dispersant mixture (Giere and Hauschildt 1979), modern products are fairly neutral if appropriately applied. Bioremediative additives might stimulate the growth of oil-degrading bacteria but they do not seem to enhance recolonization rates of meiofauna (Schratzberger et al. 2003). Experimental work on the impact of oil on meiofauna can be problematic considering the labile chemical processes and the multifactorial situation in the field. Small-scale laboratory experiments often result in extremely rapid recoveries, in the range of weeks or a month (Alongi et al. 1983; Fleeger et al. 1996). On the other hand, larger mesocosm experiments have produced drastic declines in meiofaunal populations, especially of harpacticoids, and indicated recovery times of about two months (Grassle et al. 1980–81). In field experiments three months were needed until the depressed diversity and increased evenness of meiofauna in artificially oiled sediments returned to the values of the corresponding reference samples (Schratzberger et al. 2003). Also, from a beach site treated with different dosages of crude oil, McLachlan and Harty (1982) recorded a recovery period of a few months (at least for the more robust nematodes) after an initial general decline was recorded (McLachlan and Harty 1982). In order to discover sensitive reactions that are not concealed by the survival of robust species, laboratory work often focuses on life cycle-based assessments of oil toxicity. Standardized and normative experiments on the impact of oil bioassays have been developed by Chandler and his team (see Bejarano et al. 2006 a,b), with the harpacticoid Amphiascus tenuiremis used as test species. As already shown with oligochaete offspring (Giere and Hauschild 1979), reproduction and development of these copepods is a sensitive indicator of pollution damages at concentrations that are indifferent to adults: for example, maturation time gets delayed, fertility is reduced, and larval stages (especially nauplii) become impaired or halted in their development. Palmer et al. (1988) summarized three main reactions of meiofauna after oil spills. (1) A “dramatic decline” in the abundance and diversity of the meiobenthos occurring in direct contact with the oil. (2) “No change,” probably only found in subtidal sites not immediately exposed to the most toxic, highly volatile/soluble oil compounds. (3) “Enhanced abundances” after contamination, momentary phases of
8.8 Polluted Habitats
365
the unbalanced, erratic fluctuations of some robust species. Typical of stressed communities (Warwick and Clarke 1993b), they indicate a severe disturbance in the early contamination phase. Twenty years after the summary by Palmer et al. (1988), additional compiled experiences on the reaction of meiofauna to oil impact allow a differentiation of the variable and partly controversial meiofauna reactions: (a) Depending on the extension of the spill and the nature of the oil, the initial losses in meiofaunal abundance and diversity will (at least in “high energy” sites) only persist for a relatively short time (months to a year) compared to more sheltered areas. Here the recovery will be retarded and take much longer (on the order of several years). Since the statuses of the intact neighboring sites are crucial to recolonization, small-scale contaminations, especially in exposed areas, will return to normal in the range of several weeks. (b) Juveniles are more sensitive to oil pollution than eggs or adults. Hence, bioassays analyzing reproductive success (fertility rates) and survival of first larval instars will better reveal sublethal damages after light contaminations (Bejarano et al. 2006a,b). (c) Only the most recent formulations of oil dispersants or bacterial fertilizers do not enhance the toxicity of the oil. In most cases, natural chemical and biological degradation processes and the natural supply of oil-degrading bacteria will be appropriate for effective oil-cleanup. Of course, these “invisible” and timeconsuming natural clean-up activities are often not enough spectacular for the media.
8.8.2.2 Effects of Pollution by Metal Compounds Invisible, highly persistent and ubiquitous, metal compounds are probably a greater threat to the environment than the spectacular but transient and local oil spills. Reflecting this notion is the increasing number of meiofaunal studies on the impact of (heavy) metals and their derived compounds, such as many antifouling agents. However, the majority of these studies are laboratory in vitro tests (Coull and Chandler 1992). In situ and sediment-bound, metal compounds, like other pollutants, are less toxic than in the aqueous phase. This is especially true in organic-rich muds (Austen and McEvoy 1997b; Austen et al. 1994). Chemical, physical and biological processes modify the toxicity: chelation, binding preferences (redoxdependent), pH, bioturbation, adhesion to biofilms and mucous secretions, metabolic uptake or selective storage. The synergism or even antagonism of several metals acting simultaneously in polluted areas confounds dosage effects of single metals (Mahmoudi et al. 2007). Hence, the toxicities of metals depend greatly on their bioavailability and barely relate to “total concentrations.” Therefore, the specification of actual threshold concentrations in sediments is probably not helpful, since interactions with ecological factors and different community patterns render them of local significance only.
366
8 Meiofauna from Selected Biotopes and Regions
Metals vary in toxicity; according to the impact scale, mercury and copper seem to be particularly toxic, more so than zinc, cadmium or tin (see Van Damme et al. 1984; Austen et al. 1994, Austen and McEvoy 1997a,b; Austen and Somerfield 1997). Moreover, copper compounds are particularly relevant since they are present in numerous antifouling paints and are thus widespread in the aquatic world. This explains why many meiofaunal studies have been performed with this metal. Experiments by Alsterberg et al. (2007) showed that total meiofauna biomass decreased significantly with exposure to copper pyrithione. Freshwater harpacticoids with a high content of the food-derived carotinoid astaxanthin were better protected against toxic oxidants such as copper (Caramujo et al. 2008). Mixtures of several metals can have a different toxicity compared to that of the individual metals (Fig. 8.24). In solutions containing copper with mercury or with zinc, paired (synergistic) exposure was less toxic to the common marine nematode Monhystera disjuncta than exposure to each metal individually (Vranken et al. 1988a). However, in corresponding experiments with the harpacticoid Nitokra spinipes the combined effect of mercury and copper increased the mortality (synergistic effect) (Barnes and Stanbury 1948). Newly developed organo-metal compounds such as copper pyrithione are effective antifouling biocides that are added to paints. They seem to have little negative impact on meiofauna (nematodes), they tend to mainly affect prokaryotes and fungi more (Larson et al. 2007). On the other hand, exposure to metals in combination with the organic pesticide phenanthrene (see below) proved more toxic to meiofauna than any of the pollutants alone (Fleeger et al. 2007). The physiological mechanisms behind these different combined or 6
% Mortality (Logits)
4
2
0 Cu −2 Zn mixure
−4
−6 0
0,2
0,4
0,6
0,8
1
Log10 Toxic Unit Fig. 8.24 Impact of heavy metals on the nematode Monhystera disjuncta. Higher mortality results from exposure to single metal compounds than from paired exposure to two metals. (After Vranken et al. 1988a)
8.8 Polluted Habitats
367
singular effects differing in the various meiofauna groups are unclear. Toxicity even differs with the chemical form of the metal administered to the meiofauna; for example, methylmercury is more toxic than other mercury compounds. In contrast to the contention of Somerfield et al. (1994), the overall data suggest that harpacticoids are on average more sensitive than nematodes, just as they are to other pollutants. Females seem more severely affected than males, a conclusion derived from several studies and probably related to the better solubility of metal compounds in the richer lipid deposits of the (mature) female body. The overall flux of metals through the food web from meiofauna to macrofauna varies depending on the transfer rate, which is, apparently, greater in nematodes than in harpacticoids (Fichet et al. 1999). Ecological group parameters might suggest changes in community structure where single-species analyses fail. Parallel to increased levels of metal concentrations, the diversity, dominance pattern and evenness of meiofaunal nematodes decreased in the heavily polluted New York Bight (Tietjen 1980b). Conversely, an increasing species richness and abundance paralleled decreasing contamination in a North Sea estuary (Somerfield et al. 1994). In this study, the superiority of multivariate statistics again demonstrated concealed changes and suggested “that nematode community structure changes in a smooth and ordered fashion with increasing sediment metal concentration.” Among both harpacticoid copepods and nematodes, there are even adapted species (or local intraspecific strains?) with a high tolerance to heavy metal compounds (e.g., Tachidius discipes, Microarthridion fallax, Pseudobradya sp., Tisbe sp. among harpacticoids and Molgolaimus demani, Sabatieria pulchra, Axonolaimus, paraspinosus, Oncholaimus campylocercoides, Bathylaimus capacosus among nematodes, see Warwick et al. 1988; Somerfield et al. 1994; Hedfi et al. 2007). Species of the harpacticoid genera Cletodes, Laophonte and Stenhelia are also considered robust (Saunders and Moore 2004). Additionally, adaptive effects may expand the tolerance range within the same species: Enoplus brevis (Nematoda) from a polluted site was in experiments more tolerant than specimens from unpolluted sites (Somerfield et al. 1994). Only after longer periods of exposure to metals a certain selection for more tolerant species appears to influence the community composition: along a copperenriched estuary the number of Cu-tolerant nematode species increased when compared to uncontaminated reference sites (Millward and Grant 1995). The drastically depressed abundance of harpacticoids in the Westerschelde estuary (North Sea) was considered to be due to the increased levels of heavy metals, especially of copper (Van Damme et al. 1984). In Chilean beaches exposed for years to copper mine tailings, meiofauna was restricted to certain tolerant nematode species, while the number of harpacticoids, correlating with the copper concentration in the pore water, was negligible at many stations (Lee at al. 2001b; see also experiments by Lee and Correa 2006). These results suggest that harpacticoids, especially their larval stages, are sensitive indicator organisms for ecosystem deterioration due to exposure to metal pollution. A bizarre and hopefully unique experience regarding meiofauna and metal contamination is the report by Pogrebov et al. (1997) on the severe plutonium pollution in the bottom sediments of an Arctic inlet after several nuclear test explosions.
368
8 Meiofauna from Selected Biotopes and Regions
While changes in the macro- and meiobenthic communities were not reported to be severe, in some areas the ciliate fauna was found to be massively impoverished or “eliminated,” while flagellates seemed unaffected. Moreover, the genus Euplotes showed clear morphological, anatomical and behavioral aberrances. There are a few publications on freshwater meiobenthos exposed to metal pollution. Most results correspond to those outlined for the marine realm (Burton et al. 2001), with copper being most directly correlated to reductions in the species richness of meiofauna. Again harpacticoid copepods seemed notably sensitive (exception: Bryocamptus spp.). Cyclopoid copepods (especially Diacyclops spp.) and semibenthic cladocerans (Chydorus sp.) as well as water mites (especially the halacarid Porohalacarus) were more tolerant. The authors ascribe the small changes in total species richness between contaminated and uncontaminated sites not to a small impact of heavy metals, but to a replacement of susceptible species by robust ones. Hence, metal contamination in freshwater (streams) also seems to massively alter the community composition of meiofauna. In sediments from various polluted and unpolluted German rivers, nematode community structure was related to metal pollution, but also to the hydromorphology of the sites (Heininger et al. 2007). Interestingly, predatory and omnivorous genera, such as Mononchus and Tobrilus, appeared more abundantly at sites with high rather than low metal pollution (perhaps a result of reduced competition?). Insect larvae, which in streams are a dominant taxon at the boundary of meiofauna and macrofauna, display a graded and fairly predictable response to metal pollution. Since chironomids appear to be rather resistant to heavy metal pollution, they dominated (80% of all insects) at the most grossly polluted stations (Winner et al. 1980). Hence, the authors suggest that the percentage of midge larvae in samples is a useful index for assessing this type of pollution. Studies on physiological processes in meiobenthic organisms impaired by toxic metals compounds are, thus far, rare. Binding of metals in mucus excretions might be interpreted as a defence mechanism and could play a major role, but detailed studies are lacking. In the Enoplus spp. (Nematoda) the metabolically active cuticle and hypodermis are the main organs that uptake and sequester metals (Howell 1983). Considering the situation in macrobenthos with significant physiological and ecological impacts of metals (e.g., tributyltin = TBT), also in meiobenthos detailed research on metal impacts using suitable indicator species is urgently required (see Schratzberger et al. 2002b). This would help to better understand the patterns and pathways of metal pollution in meiofauna and their physiological reactions. In streams and rivers, lead, a ubiquitously present metal of environmental relevance, should also be included in meiofaunal impact studies.
8.8.2.3
Toxicity of Pesticides
Similar to heavy metals, pesticides (herbicides, fungicides and insecticides) are ubiquitous in the environment, since they are often slow degrading and are longlived. By their intensive use, not only in agriculture, they tend to accumulate in
8.8 Polluted Habitats
369
freshwater runoffs and coastal marine sediments. Their adsorption into the biofilms of sediment surfaces means that meiofauna provide relevant test organisms not only for acute toxicity tests but also for life-cycle toxicity, including severalgeneration bioassays. The continuous development of function-designed pesticides contrasts sharply with the few studies on their effects on marine meiofauna, as noted by Coull and Chandler (1992, see their Table IV). Many of these few studies were performed in the aqueous phase, which usually results in a lower endpoint. As with heavy metals, the tests organisms appear less sensitive in the presence of sediment, so that threshold concentrations become rather problematic. Some general features from selected marine studies will be presented here without claiming any completeness. For details of the situation in freshwater meiobenthos, the reader should consult the review by Höss et al. (2006) and references therein. Pesticide toxicity is rarely found to directly limit survival. In tests with modern pesticides it is the exception rather than the rule that exposure to environmentally relevant or recommended concentrations induces direct mortality. Atrazine, a common herbicide, has been introduced into mesocosms with estuarine sediments at concentrations near the threshold proposed by environmental authorities. Rather exceptionally, this caused a 70% population decrease in several harpacticoid species, while nematodes were on average barely affected (Bejarano et al. 2005). Usually, the damage symptoms are more concealed, as shown in the lifecycle bioassay with Amphiascus tenuiremis (Harpacticoida) using Fipronil (Cary et al. 2004). A deceptive indication of this rather obscure pollution impact is that eggproducing females often seem less sensitive to the short-term impact of PCBs and other lipophilic toxicants than males (Carman and Todaro 1996; Bejarano et al. 2005). This is probably because the noxious substances are sequestered and deposited into the lipid-rich yolk of the eggs. Under realistic concentrations these widely used insecticides caused no significant lethality, but did cause sex-specific reproductive dysfunctions. In certain mating combinations there was an 80% decrease in successful reproduction, while in other combinations the reproductive success was not impaired, but the developmental time of the eggs was delayed. When these subtle, adverse effects act over several generations the ecological consequences become disastrous and massively change the population structure. Similarly adverse effects, be it a reduction in survival or in reproductive success, have also been recorded in other studies for Fipronil (Chandler et al. 2004a), and for Chlorpyrifos (Green and Chandler 1996; Green et al. 1996). Only a few experiments with licensed pesticides lacked direct toxic effects, at least through one generation: the sedimentassociated insecticide Fenvalerate seems to bind so tightly to sediment surfaces that its impact on A. tenuiremis was negligible, at least under the conditions applied (Strawbridge et al. 1992). Species react differently to the same pollutant, an obvious fact evidenced by experiments of Bejarano and Chandler (2003), and one that should caution us about making any rash generalizations. For example, Amphiascus tenuiremis, the model harpacticoid for many bioassays, was exposed to similar concentra-
370
8 Meiofauna from Selected Biotopes and Regions
tions of Atrazine as were various other mesocosm harpacticoids tested by Bejarano et al. (2005). Through two generations neither survival nor developmental time of Amphiascus were massively affected. However, reproductive failures by the decreasing number of hatching nauplii were recorded. These life history effects, which often increase from one generation to the next, will, under natural conditions, obliterate the population as certainly as direct mortality. The incorporation of molecular genetics into pollution studies has revealed details of intrapopulation variability that might explain different adaptive capacities along pollution gradients. Toxicants have been shown to alter regular gene expression (Schizas et al. 2001; Staton et al. 2001). Genetically (mitochondrially) different lineages within an estuarine population of Microarthridion littorale (Harpacticoida) showed different survival in toxic concentrations of the organophosphate Chlorpyrifos (see above) in the laboratory (Schizas et al. 2001). This corresponded to their prevailing field occurrence in a pollution gradient in the field (Schizas et al. 2002). These interrelations between population genetics, distribution and ecotoxicology might be a field of extreme importance in the future. Since copepods are an important food source for juvenile fish (see Sect. 9.4.2), and many pesticides are not biodegradable, the problem of pesticide bioaccumulation through the food chain is of high relevance in polluted environments. Transfer of the insecticide Guthion, an organo-phosphate, to copepod-feeding juvenile spot (Leiostomus xanthurus) resulted in a twofold concentration in the fish over the amount in the sediments (DiPinto 1996). Also, from other studies it emerges that sediment-associated particulate organic carbon is a main transfer route of lipophilic pollutants to both sediment-screening meiofauna (e.g., the copepod A. tenuiremis) or sediment-ingesting macrobenthos and fish (Wirth et al. 1994). The question of whether the fish can modify the pesticide accumulation by their metabolism or excretion remains unanswered. Considering the metabolically highly active cuticles of nematodes, their notably sediment-associated biology, and their ubiquity and abundance, more representatives of this dominant taxon need to be included in future studies on pesticide toxicity. More detailed reading: Zullini (1976); Platt et al. (1984); Heip (1980b); Heip et al. (1988); Bouquegneau and Joiris (1988); Warwick et al. (1990a); Coull and Chandler (1992) Warwick (1993); Kennedy and Jacoby (1999); Chandler (2004); Austen and Widdicombe (2006); Neher and Darby (2006); for freshwater: Burton et al 2001; Höss et al. (2006); Heininger et al. (2007).
8.8 Polluted Habitats
371
Box 8.9 Tiny, But Powerful: Meiofauna as Pollution Indicators What are the main characteristics that make meiofauna superior to most macrofauna when assessing ecosystem health? - Ubiquitous occurrence - Rich populations and numerous species—even in small samples—allowing for reliable statistics - High turnover of generations permits control over several generations in a short time period - Saves handling time, space and money These inherent advantages of meiofauna can provide rapid and reproducible answers about the effects of pollution. Studies with meiofauna cover all typical procedures: (1) Field studies of the polluted sites, supported by computer-based multivariate analyses. Identification at a higher taxon level does not cause much information loss and can enable convincing and quick assessments of community changes after pollution incidents. (2) Toxicity bioassays with significant test organisms can explore and quantify uptake rates within short time periods, and determine sublethal reactions to deterioration or recovery over generations. (3) New analytical, histological and genetic methods can show pollution damage in single animals. (4) Mesocosm experiments combine the advantages of sediment-based field studies with those of laboratory assays in one system. They allow for a “realistic” testing of toxic effects including concomitant bacterial uptake and metabolism, accumulations in biofilms, and combined reactions to various chemicals.
Chapter 9
Synecological Perspectives in Meiobenthology
9.1
Community Structure and Diversity
Community structure. The habitats of benthic assemblages are structured, their species richness (biodiversity, alpha-diversity) and ecological diversity regulated by the interaction of ecological and physical processes. Therefore, some basic questions are: what are those structuring factors? Why is it that sandy bottoms tend to harbor more meiofauna species than muddy bottoms? There are no generally valid answers to these questions. Many scientists contend that biotic factors such as food supply, predation, competition, and reproductive strategies are decisive; others emphasize the impact of abiotic parameters such as exposure, temperature and salinity. Of course, there are good examples of both of these positions in the ecology of meiobenthos. The conclusions depend much on the area investigated (exposed vs. sheltered habitats), the taxonomic and ecological nature of the animals studied (opportunists vs. specialists), and the methods used (life vs. fixed; sieving vs. sorting). In stable environments such as sheltered flats, nontidal seas, groundwater systems and deep-sea bottoms, biotic factors will have the stronger structuring effect on meiofauna. In extremely stable ecosystems, competitive interactions may induce instabilities in conflicting populations and ultimately cause the displacement (“amensalism”) of the less competitive species. According to the time-stability hypothesis, this means that biotope stability would reduce diversity (Rhoads and Young 1970; Woodin and Jackson 1979a,b; Warwick et al. 1986b). However, in any natural ecosystem incessant small disturbances create subtle ecological disequilibria and interfere with complete stability. Minute habitat heterogeneities create mosaics of small patches with small temporal variations. Food web interactions (predation, succession, food supply) diversify the trophic situation. Strong trophic controls favor the dominance of highly discriminative feeding types, e.g., selective deposit feeders (Warwick et al. 1990b). These oscillations and differentiations reduce competition and create a scenario that sustains highly specialized species with small populations. Usually grouped as K-strategists, their representatives live in highly interactive, patchily distributed and hierarchically structured communities. A rich pattern of biotically constrained microniches is a characteristic of these species-rich assemblages (Gray 1978). O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
373
374
9 Synecological Perspectives in Meiobenthology
While small oscillations reduce the intensity of competitive displacement and tend to enhance diversity, severe disturbances negatively affect diversity (Schratzberger et al. 2002a). This is typical of physically stressed habitats such as exposed shores with rigid hydrodynamic factors e.g., tidal and oxygen regimes. These habitats harbor a meiofaunal assemblage controlled by harsh and irregular abiotic fluctuations. Also, strong seasonal and tidal variations favor meiofaunal assemblages with more homogeneous species compositions and distributions (Hulings and Gray 1976). Dominated by a few species with large populations, these low-diversity meiofauna communities contain a high percentage of opportunistic r-strategists which co-occur under reduced hierarchical and biological interactions. Among nematodes, Sabatieria pulchra is a good example of such a species (e.g., Modig and Ólafsson 1988). Naturally, there are various transitional phases and “compromises” between these rather extreme structural patterns in meiofaunal assemblages. For instance, opportunistic meiobenthic populations can become regulated by biological interactions such as facilitation, inhibition or depletion (compare with macrobenthos: Whitlatch and Zajak 1985; Connell and Slatyer 1977). Even in eulittoral areas where abiotic disturbances dominate, considerable meiofaunal diversity can be maintained provided intensity of the disturbances is not too drastic. This could explain the high turbellarian diversity in beaches on the Island of Sylt in the North Sea (Armonies 1986; Reise 1988). Diversity. Assessing biological diversity in its various aspects, from species richness to evenness or taxonomic distinctness, in an ecologically meaningful way depends much on the method of measurement, since each aspect has its own justification and limitation. A comprehensive account of the development and use of diversity estimations in benthic communities is given by Carney (2007). The characteristics and the strong and weak points of the various diversity indices are used are also discussed in Heip et al. (1988). This treatise will not detail or reiterate them. Provided the taxa have been carefully identified to a low and homogeneous taxon level (see Sect. 8.8.1), it is simple to calculate diversity indices, especially using computer programs, but their adequate interpretation remains difficult. If not correctly applied, “diversity indices hide more than they reveal” (Platt et al. 1984). Following this line of reasoning, Crisp and Mwaiseje (1989) stated that a diversity index “is most frequently used to describe examples of impoverished data collecting.” Considering the more general aspects of diversity, the effects of various interactive factors such as intensity of disturbance, pattern of distribution and life history of the species on the diversity are of prime interest here. To what extent does grazing, organic enrichment and high production affect diversity? Answers to these basic questions would allow a deeper understanding of diversity. They should be acquired preferably through experimental approaches. In an analysis of numerous data sets, Hillebrand et al. (2007) showed that grazing reduced species richness in freshwater habitats, fertilization reduced evenness, while the assemblages reacted differently in terms of species richness depending on the degree of evenness. These data from microphytobenthic assemblages need to be repeated in corresponding analyses using meiobenthos.
9.1 Community Structure and Diversity
375
Ecological diversity is a sensitive and highly reactive parameter that indicates the structural pattern of meiofaunal communities. It is the interdependence of numerous factors which makes the interpretation of diversity so difficult. Meiofaunal communities often have characteristically high taxonomic diversities compared with macrofauna and microfauna. “Up to now it has ultimately not conclusively been explained which factor or factorial combination is responsible for the high diversity of meiofauna” (Herman and Heip 1988). This is in contrast to microfauna, which have “a surprisingly modest global species richness” (Finlay et al. 1996), concurring with the wide, often global distributions of species. A species list, even if difficult to achieve, represents just an initial step in assessing diversity. What we need to know is not just “which species and how many individuals?” but rather “why, which pattern, and since when?” The list cannot explain the processes and changes that control the biodiversity; it cannot answer questions about uneven distributions, density variations, and the relationship to spatial and temporal seales. Moreover, there are major pitfalls in extrapolating the diversity of local patches to regional scales, which may lead to considerable miscalculations, especially when comparing deep-sea with shallow bottom data (Lambshead and Boucher 2003). The influence of different sample sizes must be statistically minimized in large-scale diversity assays (Boucher and Lambshead 1995). Perhaps the traditional diversity indices (e.g., the Shannon–Wiener diversity H’, or rarefaction calculations, with their inherent impairment by equitability, sample size and local features, are not appropriate (see Gray 2000)? Therefore, new methods have been developed to better compare values from various studies and regions (Rose et al. 2005). Margalef’s “species richness-weighted diversity index” has been suggested as a robust index of alpha-diversity by Boucher and Lambshead (1995) for large-scale comparisons of biotopes (Boucher and Lambshead 1995). For further statistical details the reader should consult the valuable compilations by Underwood (1989) and Underwood and Chapman (2005). For evolutionary aspects, a measure of diversity has been introduced, particularly for the assessment of wide-scale patterns and their changes in them: the “taxonomic distinctness index” (Clarke and Warwick 2001; Warwick and Clarke 2001). It is based on the taxonomic relatedness of the species and demonstrates that a community of distantly related species, each with a long and their independent evolutionary lineage, has evolved in a different ecological situation and has a higher level of diversity than a community of closely related species. The measure of distinctness highlights this difference, and thus weights the diversity of a sample comprising numerous genera as more diverse compared to a sample with the same number of species, but where species all belong to just one genus.
9.1.1
Processes of Recolonization
Disturbance and recolonization. Inherently linked to diversity, disturbances can enhance, impoverish or devastate meiofauna assemblages and their diversity (intermediate disturbance hypothesis, Huston 1979). Depending on their intensity and frequency,
376
9 Synecological Perspectives in Meiobenthology
disturbances create a mosaic of differently composed faunal patches depending on the state of the previous fauna and its capacity to recover and recolonize (Sect. 2.2.7). The first phases of the recolonization process will proceed rapidly, as shown experimentally (Hockin 1982a; De Troch et al. 2005b). As the suite of ecological niches becomes increasingly occupied, this process will slow down. Since the widths of niches gradually become narrower, during later phases higher specialization is needed for successful colonization. Thus, the character of the meiobenthos assemblage will gradually change from a mainly r-selected to a more K-selected one. However, there is also a spatial aspect to the colonization process: the more isolated the “defaunated island” and the larger the devastated area, the slower the process of recolonization from the surrounding undisturbed biotopes. As a result, the recolonizing meiofaunal assemblage will initially be relatively poor and low in diversity. These general features corroborate the zoogeographical “island theory,” but recolonization also depends on the degree of maturity and the complexity of the neighboring “donor” assemblages; whether they are dominated by opportunistic generalists or highly adapted specialists (Azovsky 1988). The nature and life histories of the meiofaunal groups (of various dispersive competence) also play a dominant role. Because of their active emergence, harpacticoids recover most rapidly, in spite of their physiological sensitivity (De Troch et al. 2005b; see Sect. 7.2.1). In contrast, the (usually) more “sediment-bound” nematodes have a lower potential for recolonization. However, overall the recolonization of meiofauna proceeds rather rapidly. Sherman and Coull (1980) even noted that an experimentally disturbed intertidal mudflat returned to pre-disturbance levels of meiofaunal populations within one (12 h) tidal cycle. Both copepod and nematode species composition and diversity was back to pre-disturbance values within four tidal cycles. The authors attributed this rapid recolonization to the close availability of “source” populations and to suspension and transport by tidal currents (see also Sect. 8.8 for recolonization after pollution events). Colonization rates of a few weeks to two months have also been reported from experiments in the lacustrine phytal zone (periphyton), where after wide initial fluctuations a stable meiofauna assemblage established after two months dominated by ostracods, harpacticoids and rotifers (Peters et al. 2007). However, recolonization of a devastated area with meiofauna is, to a considerable degree, also a matter of chance, since it is influenced by unpredictable events such as storms, irregular exposure, the prevalence of specific reproductive patterns in the area, seasonal and trophic conditions. These stochastic factors make it difficult to pin down the timescale for recovery after a disturbance event. The composition of the establishing meiofauna will never be exactly the same as it was before, although the general structural traits may be predictable (Rhoads and Young 1970). Large-scale, man-made disturbances (e.g., erosive forces due to climatic extremes, intense trawl-fishing and ever-increasing shore constructions, dikes and waterways) have an increasingly drastic effect on meiofaunal habitats. Maintenance dredging and habitat (beach) enhancement have become regular “remedial actions” by which huge masses of sediment are mechanically distorted. Amazingly, the meiofauna seems to recover much more quickly from intensive disturbances than the macrofauna, as field studies, micro-/mesocosm experiments and large-scale field
9.2 Community Structure and Size Spectra
377
surveys have shown (Schratzberger and Thiel 1995; Schratzberger et al. 2006; Bolam et al. 2006). Supported by meticulous statistical analyses, these works have disclosed that the impact of this mechanical sediment distortion is mitigated by the concomitant rich seeding of meiofauna in the slush water and intensive migration activities. After relatively short periods (weeks), the meiofauna assemblage recovered from an initially reduced biodiversity, while the abundance was subdued for a longer period (> 1 year) compared to the reference areas. In sandy habitats the recovery was quicker than in mud, and nematodes proved less affected than harpacticoids. In the freshly consolidated sediments, recolonization not only depended on sediment structure but also on the random settling, water transport capabilities and reproductive potentials of the taxa (see Sect. 7.2.1). More detailed reading: Rhoads and Young (1970); Woodin and Jackson (1979b); Gray (1978, 2000); Heip et al. (1988); Herman and Heip (1988); Clarke and Warwick (2001); Lambshead and Bouchet (2003); Carney 2007.
9.2
Community Structure and Size Spectra
The members of an ecosystem have body size spectra which reflect biotopical, structural and functional aspects. Hence, size spectra (the density of the fauna, mostly in biomass units, logarithmically plotted against their size) can reveal complex underlying ecological processes. Schwinghamer (1981a, 1983) showed that the marine benthos is separated not only by our formal mesh size criteria, but has an intrinsic trimodal structure when analysing the size spectrum. Consistent “troughs” delineate three discrete size groups, with the meiobenthos between the micro- and the macrobenthos (Fig. 9.1). The scrutiny of data from other marine investigations and areas has confirmed the validity of this benthic grouping. Moreover, it has resulted not only from calculations of body size or biomass, but also from assimilation efficiencies and respiration rates (Gerlach et al. 1985; Warwick et al. 1986a). It also conforms to the conclusion that in a community numerous species can co-exist when animals are of different sizes, i.e., high species richness can correlate with high abundance in different size classes (Finlay et al. 1996). The differences in size and mass scales observed when comparing organisms from benthic and pelagic habitats suggest a formative impact of the substratum. Schwinghamer’s (1981a) interpretation of the separation of the meiobenthic group makes use of is along the following line of thought. The animals must either to be small enough to live interstitially between the particles (mesobenthic meiofauna) or large and strong enough to push aside the sediment particles (endobenthic macrofauna). However, this size-related grouping does not separate macro- from meiobenthos only. Even within the meiobenthos, different animals live in different sediment types and have evolved different lifestyles (interstitial, epibenthic, burrowing, etc.) which require divergent size ranges (Tita et al. 1999). In the interstitial of sandy sediments, slender nematode species (body width class around 20 mm) with smaller individual biomasses prevail, while in muddy sediments the burrowing/pushing
378
9 Synecological Perspectives in Meiobenthology 103
Biovolume (cm3 x m −2)
102 101 100 10−1 10−2 10−3
0.5 µm
2
8
32
125
500 1
2
16
32
mm Equivalent Spherical Diameter
Fig. 9.1 Size spectra of benthic fauna. The values shown are from six sampling locations in a Canadian intertidal flat combined with data derived from eight other studies; weight data from the literature have been converted into volume data. (After Schwinghamer 1981a)
nematodes are broader (body width class around 35–45 mm) and heavier. Associated with these morphotypes are different feeding groups (microvores vs. epigrowth or predators) and metabolic types. The higher metabolic or respiratory ratios of the microvores in sands corresponds to their relatively long guts and indicate a different, lower-quality food spectrum. Conversely, the mud-dwelling larger species with shorter guts had lower metabolic ratios adapted to high-quality food (see also data from Kennedy 1994a for comparison; see Fig. 5.22). However, can a differentiation of meiobenthic size spectra by sediment type explain the trough towards the microbenthos? It has been argued that the underlying factors are of a more biological and evolutionary nature. Benthic assemblages are structured by various biological characteristics such as feeding behavior and life history modes. These, in turn, are affected by the spatial and temporal conditions of the environment. Following the principle of competitive displacement, Warwick (1989) and Warwick et al. (1986a) maintain that competition with the other benthic groups causes a delineation not only in size and biomass but probably also in a whole array of biological characteristics such as metabolic efficiency (see also Warwick et al. 2006). The demarcation towards the Protista (microbenthos) may be because the protists are mostly hapto-sessile organisms that adhere to sediment particles, while most meiobenthos are vagile. The numerous differentiating features are summarized in Table 9.1. It appears that the diverging size spectra result from mutually structuring effects that have evolved through biological interactions between larger and smaller
9.2 Community Structure and Size Spectra
379
Table 9.1 A biological delineation of meiobenthos vs. macrobenthos (Warwick 1984) Animal weight
< 45 mg
> 45 mg
Development Dispersal Generation time Reproduction Growth Trophic type
Direct, all benthic Mainly as adults Less than one year Mostly semelparous Attain an asymptotic final size Often selective particle feeders
Mobility
Motile
With planktonic stages As planktonic larvae More than one year Mostly iteroparous Life-long permanent growth Often non-selective particle feeders Also sedentary
benthos. As a consequence, today, the different benthic groups occupy different ecological niches, thus avoiding competition. This principal becomes evident in larvae. As long as larvae of macrobenthic species are of meiofaunal size, most of them occur as meroplankton in the open water. They do not settle prior to reaching an average size beyond the main prey size of the meiobenthos. Thus, they avoid contact with the meiobenthos, with its direct competition and predation. According to Warwick this life history feature is mainly responsible for the size separation of the benthic fauna. Consequently, in a diagram of size structure, the size range of meroplanktonic larvae of the macrobenthos fits exactly in the trough between the benthic meio- and macrofauna (Fig. 9.2). However, there are counter-arguments. If evolutionary pressures caused an ecological differentiation into separate size groups, there would be little interaction and competition between meio- and macrobenthos today. But the evidence contradicts this notion (Coull and Bell 1979; Bell and Coull 1980; Reise 1985; Watzin 1986). There is, indeed, a high rate of interactive effects (see Sect. 9.4.2) between temporary and permanent meiofauna of the same size class, which probably has a strong structuring impact on the macro- and meiobenthos. Following the “planktonic larva hypothesis,” benthic size spectra from polar seas, where macrofauna rarely have planktonic larvae, should have an unimodal, continuous curve without separating troughs, because larval “escape” from the benthal (separating the size spectrum) does rarely occur. However, a study with Warwick as a coauthor (Kendall et al. 1997) on polar benthic communities (Spitsbergen and the Barents Sea) showed a well-developed bimodal curve of size spectra with a trough between the meio- and the macrobenthos. The low endemism in the Arctic, with many invasive non-Arctic species and their planktonic larvae, was believed to have modified the pattern of adult–larval interactions and altered the competitive control between the meiobenthos and macrobenthos. A parallel investigation including larval plankton is needed to qualify this interpretation by Kendall et al. (1997). Burkovsky et al. (1994) in their study on the Arctic benthos of the White Sea shores also confirmed the division into micro-, meio- and macrobenthos. These authors interpreted the size separation as being a result of fundamental biological parameters such as generation time and locomotory activity. These are believed to create a hierarchical habitat heterogeneity in which the different groups perceive
9 Synecological Perspectives in Meiobenthology
No of species
Proportion of species
380 0.1
Northumberland
adults
0.05
0 Polychaete larvae
3 0 4
Northumberland larvae
0 2
6
10
14
18
22
26 30 weight class
Fig. 9.2 The size spectrum of benthic fauna in relation to that of planktonic larvae in the Northumberland area (Great Britain). (Warwick 1989)
their environment at different grades of resolution: related to their own body volume, the relative density of microfauna or meiofauna in a given volume of substratum is much lower than that of macrofauna. Similarly, in a given astronomical time period (e.g., one year), a ciliate has many more generations than a nematode or a crab. Hence, the “organismic” space and time of meiofauna is much more extended than that of macrofauna (see Fig. 9.3). Each of the three size categories has its own environmental perception, with differing chorological and chronological scales. Correspondingly Stead et al. (2003) concluded from the different responses of stream benthos to environmental factors that freshwater meiofauna and macrofauna assemblages live at different spatial and temporal scales, thus perceiving the environment “with a different grain.” In the brackish Baltic Sea, several studies revealed a complex picture of partly contradictory size spectra. Clues are probably low endemism, changing impacts of stress factors such as hypoxia, and the regionally differing invasions of non-domestic species in different parts of the Baltic. In a study of benthic communities from the southern Baltic Sea (Gulf of Gdansk), two disparate size groups of meio- and macrobenthos emerged, separated by a (fairly low) trough (Drgas et al. 1998). Although the authors conclude that biomass size spectra represent a “unifying concept in community ecology,” various other sites in the Eastern Baltic (Swedish Askö area) did not display this bimodal benthic size spectrum (Duplisea and Drgas 1999). For the lacustrine zoobenthos of Mirror Lake (USA), Strayer (1986) calculated a unimodal curve without any separation between meio- and macrofauna. He attributed this to the numerous oligochaetes and chironomid larvae, characteristic members of the freshwater meiobenthos; their size spectra exactly fill the trough between the typical meio- and macrobenthos. This would corroborate Warwick’s (1989) arguments: while many factors in lakes are comparable to those in the
9.2 Community Structure and Size Spectra
381
1024 512
CILIATES Year in generation periods
256 128 64 32 16 8
NEMATODES 4 2 1
MACROBENTHOS
10
102
103
104
105
106
Free space for each individual (related to body volume)
Fig. 9.3 “Organismic” time and space for various faunal size categories; for details see text. (After Burkovsky et al. 1994)
marine realm, the reproductive biology of the lacustrine benthos is essentially different, since planktonic larvae are largely lacking. Additionally, in freshwater much of the temporary meiofauna, such as many insect larvae, leave the aquatic biota after metamorphosis. Lack of direct competition prevented the evolution of disparate size spectra in lacustrine systems which resulted in a unimodal, continuous curve. Thus, the competitive situation between meiofauna and macrofauna in freshwater is fundamentally different from that in the marine sites studied by Schwinghamer (1981a, 1983) and Warwick (1984, 1989). However, size spectra analyses from other freshwater habitats do not conform to this idea. From a Piedmont stream (USA), Poff et al. (1993) revealed a clearly trimodal structure for metazoan size: oligochaetes and microcrustaceans provided the main meiobenthos, an introduced bivalve was the macrobenthic component, and fish represented a separate third group. The authors contend that this diverging structure, in contrast to the lacustrine studies by Strayer (1986), is because of their comprehensive sampling across all habitat types and faunal size ranges. However, the particular faunal composition of this stream, which is influenced by many local factors, may not be valid for all types of freshwater habitats. Woodward et al. (2005) interpreted marked size disparities in the benthos under trophic aspects. They suggested that different size groups represent fairly
382
9 Synecological Perspectives in Meiobenthology
independent “sub-webs” with a relatively low degree of trophic connection. Within these different compartments the larger size spectra are more severely affected by environmental perturbations, a notion confirmed by the different reactions of macro- and meiobenthos (Alongi 1985). Even though the factors governing the pattern of size spectra in benthic assemblages are, as yet, contradictory or unclear and apparently different in different ecosystems, the relatively straightforward parameter of “size” seems to have an ecological bearing beyond just the biological definition of meiobenthic size delineations. Size-spectral analyses, therefore, find increasing use in (meio)benthic studies on the energy flow through the system (see below). More detailed reading: Kennedy (1994a); Warwick et al. (1986a); Finlay et al. (1996b); Stead et al. (2003); Sheldon (2005); Woodward (2005).
Box 9.1 Diversity and Size Spectra: Indices of Community Structure The diversity and the size range of meiobenthic species are more than just abstract figures. They tell us about the life history, the nature of interactions with other organisms, and about the environmental stress status of the community-all of them parameters of high structural relevance. A biotope with high meiobenthic species richness will have a complex physical architecture, and favorable chemical and trophic conditions. The wealth of positive biotic interactions will counterbalance the impact of negative stress factors. Because of the minute sizes that define the meiobenthic world, our perception of these structural aspects is minimal; we need indicators to envisage, measure and compare them. These indicators can reveal that seemingly uniform sands or the monotonous deep-sea floor represent “havens” for meiobenthos, while exposed shores or organic-rich muds often are meiofaunal “deserts.” Diversity indices and size-related spectra reflect the complexity and hierarchy of community interactions, the ecological resilience and the colonization potential. However, the multitude of influencing factors makes them problematic. Their interpretation requires an assessment of the environmental situation, and they cannot substitute for good knowledge of the community composition. Characteristics of meiofaunal communities and their differentiation from other benthic communities are also reflected in the size spectra. Whether based on biomass, respiration or other size-related quantities, the meiobenthic range is often a clearly separated cluster in the benthic spectrum. As yet, the factors contributing to this pattern are not fully understood, but substrate and food interactions seem to play a major role. Despite numerous suggestions, why this pattern does not emerge from benthic studies of all ecosystems, marine and limnetic alike, is also unclear. However, the important message remains: there are biological and ecological characteristics differentiating meiofauna aside from their mesh size limits.
9.3 The Meiobenthos in the Benthic Energy Flow
9.3 9.3.1
383
The Meiobenthos in the Benthic Energy Flow General Considerations
The ecological role of meiofauna in the benthic ecosystem can only be assessed by measuring the flow of energy, from uptake to excretion and from reproduction to mortality. Static parameters such as abundance and biomass are only momentary reflections of this flow. Compared to the macrobenthos, assessments of production and energy flow in the meiobenthos are particularly important since the relevance of small animals lies in their high dynamics and turnover. In contrast to the macrobenthos, metabolic rates of single meiobenthic animals are difficult to measure. The energy budgets (Crisp 1984) are usually calculated using the summarizing parameters P = C − R − E (where P = production; C= consumption; R= respiration; E= excretion as feces or urine). The relations of meiobenthos to other faunal elements and its contribution to the energy flow through the benthic ecosystem can be assessed by measuring its population density (abundance, biomass), production and annual turnover. A description and critical evaluation of pertinent methods and their applicability, together with a detailed report on their inherent problems, is given by Feller and Warwick (1988) and Van der Meer et al. (2005). In addition, the accounts by Banse and Mosher (1980; mainly for macrofauna), Gray (1981), De Bovée (1987) and Warwick and Clarke (1993a) should be consulted. Compilations for freshwater meiobenthos in lakes can be found in Plante and Downing (1989) and Bergtold and Traunspurger (2005). The calculation of production in streams is discussed in Stead et al. (2005). A comprehensive compilation of benthic life history data, energy flow and pertinent literature can also be found in Brey (1990) and his Virtual Handbook (Brey 2001). With the focus on macrobenthos, this also presents numerous meiofaunal results, including conversion tables with search links to specific taxa. Therefore, except for some general aspects, methods will not be considered here in detail. The assessment of biomass from meiofaunal wet weight (wwt, or better “wet mass”) is difficult because of the inherently large range of error. Hence, biomass is mostly calculated from abundance using conversion factors (see Table 9.2). A nondestructive, semi-automatic method for calculating biomass applies digital microphotography and analytical computer graphics and allows fast and individual determinations of body volume from which mass values can be derived. The results, tested on thousands of nematodes and harpacticoids, did not differ much from tedious gravimetrical measurements (Baguley et al. 2004). However, biomass values at a given time and for a given area (“standing stock”) remain of limited ecological value if not connected with life history data which allow for extrapolations of the energy flow. Gerlach (1971) pointed out that meiofaunal biomass comprises just 3% of the overall biomass; however, the nutritional share of meiofauna within the food web is ~15%. The production, “the gain in organic substance per unit of time,” is a more relevant value when considering the ecological role of meiofauna. When calculating annual production, the P/B ratio
384
9 Synecological Perspectives in Meiobenthology
Table 9.2 Calculations of weight and derived parameters of meiofauna obtained by applying biometric conversion factors (compiled from Feller and Warwick 1988; Friedrich et al. 1996; Baguley et al. 2004; and other authors as indicated) Taxon Calculation Notes Reference A. Volume from length and widtha Nematodes Harpacticoids Nauplii Ostracods Turbellarians Polychaetes Oligochaetes Tardigrades Halacarids Kinorhynchs Gastrotrichs Isopods
L × W2 × C
Rachor (1975) (modified) C = 530 C = 230–560b C = 360 C = 450 C = 550 C = 530 C = 530 C = 614 C = 399 C = 295 C = 550 C = 230
B. Wet weight (wwt) from volume Meiofauna, most taxa V (nl) × 1.13* Nematodes + harpacticoids V (nl) × 1.064* Nematodes (p r2 L × 1.13*) − 10% Ciliates, alive p L W2 × 0.083 Ciliates, fixed p L W2 × 0.083 × 2.5 Rotifers 0.26 L W2 × 1.028* C. Dry weight (dwt) from wwt Nematodes Factor 0.25 Factor 0.20 Factor 0.15 Harpacticoids Factor 0.225 Factor 0.203 Nematodes + harpacticoids Factor 0.215 Nauplii Factor 0.225 Rotifers Factor 0.1 Meiofauna Factor 0.15
Wieser (1960) Baguley et al. (2004) Riemann et al. (1990) Friedrich et al. (1996) Friedrich et al. (1996)
Freshwater Mean value
D. Ash-free dry weight (adwt) or carbon content (C) from dwt Nematodes Factor 0.47 Factor 0.51 Harpacticoids Factor 0.46 Nauplii Factor 0.46 Rotifers Factor 0.1 E. Weight from length Nematodes
log10dwt = 2.47log10L −7.97
F. Carbon content from volume (V) Rotifers 0.26 L W2 × 0.08 G. Carbon content from wet weight Meiofauna Factor 0.116 Nematodes
Factor 0.124
Wieser (1960) Myers (1967) Schiemer (1982) Friedrich et al. (1996) Baguley et al. (2004) Banse (1982) De Bovée (1987) Baguley et al. (2004) Baguley et al. (2004) Friedrich et al. (1996) De Bovée (1987)
Friedrich et al. (1996) De Bovée and Labat (1993) Jensen (1984)
Note: L, length; W, width; r, radius; V, volume; C, factor. b Harpacticoid conversion factors depend on shape: pear-shaped, 400; semicylindrical, 560; depressed, 230 (Warwick and Gee 1984). *This value is the specific gravity a
9.3 The Meiobenthos in the Benthic Energy Flow
385
becomes important, especially in the often short-lived meiofauna. This ratio expresses the rate of turnover in a given time, and thus integrates over the different biomass and lifetime data of the various taxa (Table 9.3). Production and P/B ratio are, however, difficult to calculate since they require life-table parameters such as fecundity, mortality, generation time, and other data on population dynamics. Especially for meiobenthos, life-table parameters must still be extrapolated from only a few species studied in the laboratory. Moreover, life history cannot be considered a static process. For individuals of the same species, these parameters vary with environmental conditions and food supply (see Vranken et al. 1988b, for nematodes), and between species this variability is greater. Even in common species, insufficient life history data present a serious barrier to accurate quantitative calculations. “A great deal more data is needed before it becomes possible to refine the relationships and to establish … equations according to taxonomical and ecological groups” (Ceccherelli and Mistri 1991). The more detailed the data we have on single species, the more reliable our compilations become, making them more useful for testable ecological hypotheses and predictions. Measuring the annual production requires knowledge of the annual number of generations. This varies considerably among taxa (Table 9.4), and in many populations, growth of distinct generations (= cohorts) is not identifiable since reproduction is
Table 9.3 Annual P/B ratios calculated for some common meiobenthic species (Heip et al. 1985b; Heip 1995) Taxon P/B ratio y−1 Harpacticoida Huntemannia jadensis Tachidius discipes Microarthridion litorale Paronychocamptus nanus Canuella perplexa
3.8 9.3 18.0 24.5 11
Nematoda Oncholaimus oxyuris Paracanthonchus caecus Monhystrella parelegantula Chromadora nudicapitata
3–6 10.4 18.2 31.4
Ostracoda Cyprideis torosa
2.7
Table 9.4 Annual number of generations for some common nematode species (Heip et al. 1985a) Taxon Generations 1.6 Oncholaimus oxyuris 5 Monhystrella parelegantula 10 Rhabidits marina 13 Chromadorina germanica 15 Monhystera denticulata 17 Diplolaimelloides brucei
386
9 Synecological Perspectives in Meiobenthology
continuous and the generations overlap. The high variability in life history parameters results in inaccurate data when generalized for a large taxon like nematodes, harpacticoids, or even meiofauna. So far, we have often been forced to extrapolate our calculations of production rates from the few meiofaunal species that have been successfully cultured through generations. Life history studies have concentrated on representatives of the two most important meiobenthic taxa ecologically, harpacticoids and nematodes. Life history data are often calculated in context with toxicity testing and food web analyses. The main problem in obtaining reliable life history data for different species is twofold: (1) laboratory conditions may differ from natural field conditions, so we need field-generated life tables; (2) assessing life history data is an extremely tedious task, but it is still rarely considered a “modern” science that would generate financial support. Among other harpacticoid copepods, detailed autecological analyses of growth and production exist for Canuella perplexa. Using this species, Ceccherelli and Mistri (1991) demonstrated that direct measurements yielded results that were considerably different from those obtained through indirect calculations, which greatly underestimated production. Other harpacticoid species whose production and population development have been followed over time are: a decade-long study on Parastenhelia megarostrum from New Zealand (Hicks 1985), and a 15-month study on Enhydrosoma propinquum, Microarthidion littorale and Stenhelia bifidia in South Carolina, USA (Fleeger 1979). Also, Fleeger and Palmer (1982) and Morris and Coull (1992) have focused on the common harpacticoid Microarthridion littorale. The latter authors found a remarkable dominance of the naupliar stock, whose large interannual variations (due to predation and natural mortality) determined the overall population fate of the species. Morphologically closely related Tisbe species have been studied in detail, revealing differences in growth and reproduction (Battaglia 1957; Gaudy and Guerin 1977; Abu Rezg et al. 1997). The extensive growth and short generation times of the remarkable harpacticoid Drescheriella glacialis from Antarctic sea ice documented specific traits that were interpreted as adaptations to the low ambient temperatures (Bergmans et al. 1991). Many harpacticoid life tables have revealed the high sensitivities of reproductive and larval stages to toxicants (Strawbridge et al. 1992; Chandler et al. 2004a,b). Corresponding work on nematodes includes detailed studies on Chromadorina germanica (Tietjen and Lee 1977), Chromadorita tenuis (Jensen 1983) and Monhystera disjuncta (Vranken et al. 1988b). The number of species capable of being cultivated through several generations is steadily increasing. Numerous freshwater species have been studied intensively for their life history data; again the driving force is the use of test species for the impact of pollution (see compilation by Bergtold and Traunspurger 2006). In order to elude the laborious direct measurements of life history data, more generalized approaches have been developed which summarize existing data. Based on abundance, the values for wet and dry weight, carbon and energy content can be calculated applying conversion factors (Brey 1987, 1990; Ricciardi and Bourget 1998). Another approach is to lump individual data and calculate mean values (De Bovée and Labat 1993). When measuring the rate of respiration (semi-automatically), the
9.3 The Meiobenthos in the Benthic Energy Flow
387
energy consumption can be assessed (Baguley et al. 2004) or it can be calculated from body volume, which, in turn, can be derived from size or weight data (see Table 9.2). Regarding the different methodological approaches of the underlying studies and the different faunal properties that exist in the various sites, results and formulae obtained from these reductive and generalized compilations should not be overestimated. “Generalizations remain useful but also dangerous tools” (Vranken et al. 1988b). Most energy-flux diagrams which include meiobenthos and contain quantitative figures are not as accurate as the figures insinuate. Often calculated to the fourth digit after the decimal point, many of them are not mathematically meaningful and are probably best considered rough estimates with a rather loose relation to reality. Existing models are probably more useful for qualitative and comparative purposes than for retrieving realistic values of energy metabolized.
9.3.2
Assessing Production: Abundance, Biomass, P/B Ratio, Respiration
Following the general outlines above, we now provide some examples of how to assess the various parameters required to estimate production. They simultaneously illustrate the extreme variability involved and some of the inherent pitfalls. Since the (marine) meiofauna in almost all habitats is dominated by nematodes, the abundance values obtained for nematodes are often representative of the whole community and thus allow for some initial generalizations (see Figs. 5.15a,b; Sect. 5.6.1). This also applies to biomass: nematodes represent >90% of all living biomass in intertidal salt marshes, and even in the subtidal this figure reaches almost 80% (Sikora et al. 1977). 9.3.2.1 Abundance, density As shown in Table 9.5 the figures for meiofaunal density vary greatly, and authors that give average values should be aware of this difficulty. The review by McIntyre (1969) reports a range of between 30 and 30,000 ind. 10 cm−2. If 1,000–2,000 ind. 10 cm−2 can be assumed to be an average value integrated over all habitats (Coull and Bell 1979), the meiofauna would exceed the macrofauna in abundance by twoto threefold. Particularly high values for meiofauna (sometimes ten times greater than this range) have been recorded from silty mud and fine sand in tidal lagoons and flats (for harpacticoids, see the compilation in Table 25 of Heip et al. 1995). As sublittoral depths these extreme densities become reduced, and at greater depths they rarely exceed 2,000 ind. 10 cm−2. 9.3.2.2 Biomass With general values of about 1–2 g dwt m−2 and peaks at around 5 g (Coull and Bell 1979), the meiofauna of shallow littoral bottoms usually attain less than 10% of the
388
9 Synecological Perspectives in Meiobenthology Table 9.5 Meiofauna abundances from various habitats (compiled from McIntyre 1969 and some other authors) Habitat and locality Abundance (ind. 10 cm−2) Sandy beaches, tidal: West coast of Scotland West coast of Denmark North Sea shore in Germany Indian Ocean shores Australian east coast European beaches Arctic beach
1,000–4,000 750–1,900 200–800 1,000–10,000 ~470 360–4300 250–480
Sandy beaches, atidal: East coast of Sweden Kattegat (low tides)
200–1,000 ~500
Muddy to silty tidal sand, tidal flats: Sylt, North Sea Lynher Estuary, England Coast of Netherlands Vellar Estuary, India Swartkop Estuary, S. Africa Gironde Estuary, France European estuaries Mud, North Inlet, S. Carolina Sand, North Inlet, S. Carolina West coast of Canada Alaskan flat
Several thousand, <11,600 ~12,500 ~2,500 420–3,800 30,000–60,000 200–1,050 <14,500 856 641 <20,000 500–5,000
Shallow subtidal sediments: Sand, east coast of Denmark Sand, Baltic shore Sand, 1.5 m, North Sea North Sea, England Fine sand, Belgium Fine silty sand, Helgoland (North Sea) Antarctic shore Tropical seagrass bed Mud, Fladen Ground, North Sea Mud, English Channel Mud, Buzzards Bay, US east coast Mud, low brackish water, Finnish coast Mud, Baltic Sea Mud, Adriatic Sea
600–1,300 100–4,200 Several thousand, <9,200 <16,000 ~2,200, <5,400 ~4,000 6,200 on average <5,400 900–3,200 90–200 280–1,860 11–100 4,000–5,000, max 17,000 1,460–4,000
Deep-sea: Atlantic (Porcupine Bight) Western Pacific
500–1,500 100–1,000
corresponding macrofaunal values. In organically enriched sites, meiofauna can exceed 10 g dwt m−2. However, in ecologically extreme biotopes such as sandy beaches, brackish water regions and deep-sea bottoms (Thiel 1972; Tietjen 1992, see Sect. 8.3), it is initially the macrofauna that decreases in abundance, resulting in an increased relevance of the meiofauna (Fig. 9.4). In such cases the biomass relation between the two faunal groups can sometimes become 1:1.
9.3 The Meiobenthos in the Benthic Energy Flow
389
Abundance
Macrofauna 103 ind. x m−2
Meiofauna 106 ind. x m−2
10 9
Macrofauna
8
Meiofauna
7 6 5 4 3 2 1 0
mud
mixed
sand
Sediment type Biomass 400
Macrofauna g x m−2
Meiofauna g x m−2
350 300
10
250 200 5
150 100 50
0
0
mud
mixed
sand
Sediment type Fig. 9.4 Abundance and biomass (wet weight) of meiofauna vs. macrofauna in different sediment types of the eastern Baltic Sea. (After Ankar and Elmgren 1976)
Some technical comments on biomass values. If stated as wet weight (wwt) or better wet mass (wm), one should indicate whether the values are based on unfixed or formalin-fixed material. Fixation will enhance the weight values considerably (Wiederholm and Erikson 1977; Widbom 1984) and also cause changes in other weight parameters. Gravimetrically obtained biomass values should always be preferred, since values converted from body volume are usually massive underestimates (up to 50%; Udalov et al. 2005). Table 9.2 presents some formulae and conversion factors used in weight calculations for different meiofaunal groups, and it demonstrates the considerable variability. Length (L) measurements of several common nematodes have been converted into body mass (dwt) by De Bovée (1987) using the formula
390
9 Synecological Perspectives in Meiobenthology
log W (dwt) = 2.47 log L − 7.97. A detailed data list converting size and shape to weight for aquatic organisms is regrettably difficult to access (Chislenko 1968). For freshwater meiofauna, Kurashov (2002) calculated the wet weight (mg) from the length (mm) using the formula W = q × Lb, where q and b are conversion factors that are specific to each taxon. The average body mass of an adult nematode from tropical shallow sediments is 0.7 −3.0 mg g dwt (Grelet 1985); specimens from the deep Gulf of Mexico appear to have a similar weight but the very high standard variation reflects the extreme size range of nematodes (Baguley et al. 2004). From the Mediterranean a value of about 0.5 mg (dwt) has been reported (De Bovée 1987). In the latter study the average harpacticoid weighed 2.8 mg. More detailed body masses are given in Schwinghamer et al. (1986). With increasing water depth the average nematode biomass tends to decrease while that of harpacticoids increases (Baguley et al. 2004). A list of mean dry weight values for various meiofaunal groups (Faubel 1982) or single species (Huys et al. 1992) may prove useful. The most reliable weight parameter is the carbon content (C) or ash-free dry weight (adwt). With the development of modern carbon analyzers, the C contents of individual meiofaunal organisms can be directly and accurately determined. Jensen (1984b) found for meiofauna an average C content of about 12% wwt (12.4% for nematodes; see Table 9.2) when the material was fresh. In formalin-preserved specimens there was a carbon loss of 8–24% depending on storage time. The basis for further assessments of production is the caloric value, which is frequently derived from the C content. Direct determinations of meiofaunal caloric values are rare. Caloric values calculated from C values have only a small variance: ∼6 kcal g−1 adwt for nematodes and harpacticoids (see Coull 1999). Considering how demanding direct calorimetry (heat dissipation) is, conversion is the better approach to use (Sikora et al. 1977). As a rough equivalent, 1 mg C (or adwt) corresponds to 25 joules. Parameters often measured to quantify the overall energetic content of meiofauna without tedious counting are the ATP content, the dehydrogenase activity, and the activity of the electron transport system (ETS). The value of these parameters is their ability to designate processes in living cells only. The methods involved are well established in physiological studies, where they are based on isolated organisms under controlled conditions. However, using them in a field situation and applying them to complex multispecies communities creates severe interpretation problems due to non-reproducible fluctuations and inconsistent variations. Enzyme activity as well as electron transport are dynamic processes that depend on the physiological status, the age and other individual variables. This could explain the high intervariability that is often recorded in replicates, and it makes data interpretation a problem. If used for whole sediment cores, the necessary separation of animal-related values from bacteria-related ones is particularly difficult. Expressing the ATP content in terms of the more illustrative wet weight, Goerke and Ernst (1975) used a factor of 740 based on experiments with nematodes. The methodological procedures used to measure these energetic parameters are relatively simple and result in exact figures. However, considerable experience and a solid understanding of the often complicated
9.3 The Meiobenthos in the Benthic Energy Flow
391
underlying physical and (bio)chemical processes are required to reach a reasonable interpretation (see Sect. 2.1).
9.3.2.3 Generation time, P/B ratio The annual number of generations is difficult to obtain from field data. Gerlach (1971) estimated three nematode generations per year as a realistic figure, while Heip et al. (1985a) considered 5–15 annual generations to be a representative range for nematodes (see Table 9.4). For nematode populations from fine North Sea sands the number was 20. Phytal nematodes from Sargassum had 60 annual generations (Kito 1982). The minimum generation time under field conditions for the common nematode Monhystera disjuncta varied between 9 and 52 days depending on the temperature. In cultures, the species reached up to 23 generations per year (Vranken and Heip 1986). This high variability (Table 9.4) within the same species illustrates the limitations of using one representative value for “nematodes,” viz., of most meiofauna. This would question the log-linear relation between weight and P/B ratio given in Heip et al. (1982a). The harpacticoid Parastenhelia megarostrum had seven annual generations (Hicks 1985) and Fleeger (1979) reported 12 generations per year for Microarthridion littorale, nine for Enhydrosoma propinquum and five for Stenhelia bifidia. P/B ratio. According to the seminal paper by Banse and Mosher (1980), the annual P/B ratio is ecologically related to body mass. Heip et al. (1982a) showed that this relationship also agrees with meiofaunal data (Fig. 9.5a). Gerlach (1971, 1978) based his estimates of quantitative energy fluxes on an estimated annual P/B ratio of 9 for all meiofauna, which has since been used in many calculations. This value is close to the value of 8.4 reported by Warwick and Price (1979) on the basis of more detailed meiofaunal data and using the formula given by McNeil and Lawton (1970, see below). Vranken and Heip (1986) considered 20 a better estimate of P/B. Chardy and Dauvin (1992), in their production model of a sandy sediments from the English Channel, set the ratio at 15. For meiofauna from an Antarctic island an extreme range of between 4 and 69 was used (Vanhove et al. 2000). In entire communities that also suffer from predatory impact the P/B ratios are usually lower than those obtained in analyses of single species. The various P/B values are often derived from those few species whose life history data have been obtained in the laboratory (Table 9.3). For the harpacticoid Parastenhelia megarostrum, Hicks (1985) calculated an annual P/B ratio of 15, which is close to the P/B value of 18 given by Fleeger and Palmer (1982) for another common harpacticoid, Microarthridion littorale, but much higher than 2.5–4 listed in Schwinghamer et al. (1986) for “harpacticoid copepods.” Using field and laboratory studies, Heip (1976) based his calculations for the common harpacticoid Tachidius discipes on “eliminated biomass” (mortality) from a brackish pond, and resulted in a generalized P/B ratio of about 15, but emphasized the possibility of lower values for other meiofauna. Banse and Mosher (1980) suggested that meiofauna tend to have a considerably lower P/B than
9 Synecological Perspectives in Meiobenthology
100
5000
50
3000 µl O2 x h−1 g−1 wwt
annual P /B - ratio
392
10 5
culture field
1 −7
a
−6 −5 −4 −3 log body mass (kcal)
0.7
4
1000 500
1 0.1
b
b=
0.51.0
5 10 µg wwt
50100
500
Fig. 9.5 a The relationship of P/B ratio to body mass in meiofauna. The line represents the relationship proposed by Banse and Mosher (1980). b The relationship of respiration (at 20 °C) to body weight, emphasizing meiofaunal data. Enlarged circle to the left: the ciliate Tracheloraphis (a After Heip et al. 1982a; b after Vernberg and Coull 1974)
macrofauna because of low predation and relatively long generation times. This reasoning has been criticized and is not supported by most data on meiofauna. Rather, Gray (1981) mentioned a range as low as 0.1–5.5 referring to macrobenthic animals (see also Table 1 in Schwinghamer et al. 1986). Within freshwater nematodes, the annual P/B ratio varied between 5 and 58 (Bergtold and Traunspurger 2006). For various European lakes, the overall seasonal P/B ratio of the meiobenthic community has been calculated to be 4.8 (Kurashov 2002), twice the value for the corresponding macrobenthos. Even lower is the meiofaunal P/B in the deep-sea or stygobios, with annual ratios close to one. Ciliates from the shallow sublittoral of the Baltic Sea have been estimated to have a P/B of 260 (Sich 1990). As with microfauna, most short-lived meiofauna have several generations per year. Although the P/B per generation or cohort is less variable considering the seasonal fluctuations (Ceccherelly and Mistry 1991), the year is the usual time reference for meiobenthic P/B values. Regarding this extreme variability in the different meiofaunal groups and habitats, it has been maintained that the use of “a generalized value of nine is invalid” (Vranken and Heip 1986; Moens and Vincx 1997b). The ecological dependence of the P/B ratio on size or body mass (Heip et al. 1982a; Schwinghamer et al. 1986, see Fig. 9.5a) is the basis for the contention that the annual production in benthic communities can be calculated from biomass sensu Brey (1990), who summarized numerous data sets from benthic studies and provided the formula: log P/B = −0.2455 − 0.1663 log B − 0.2019 log W, where B is the mean annual biomass of a species and W the mean individual weight.
9.3 The Meiobenthos in the Benthic Energy Flow
393
The corresponding formula suggested by Vranken and Heip (1986) is: log P/B = −1.288 − 0.44 log W (dwt). Brey (1990) calculated the mean annual production of a benthic species as log P = −0.2578 + 0.9905 log W. For further elaboration of these calculations, see Brey (1990).
9.3.2.4 Respiration Respiration rates are often the basis for calculations of production. As much as biomass and generation time are interrelated, there is also a proportionality between weight and respiration rate. Respiration or metabolic activity depend on size and weight (Peters 1983), mostly defined by the power equation R = aWb (R= respiration in nl O2; W= weight in mg dwt; a= value for metabolic intensity, which is species-dependent; b= about 0.75 for nematodes and harpacticoids and also for most other poikilotherms, see Banse 1982; Heip et al. 1982a; Warwick and Price 1979). Calculated per mass unit (the mass-specific rate of oxygen consumption), respiration is usually inversely related to the animal’s total body mass (Fig. 9.5b). Because of the divergent body masses in meiobenthic animals, respiration rates of various meiobenthic taxa range over orders of magnitude (Vernberg and Coull 1974). While the interstitial ciliate Tracheloraphis consumed on average 4,500 ml O2 h−1 g−1 wwt, corresponding values for metazoan meiofauna were estimated to be around 1,500 ml (Gerlach 1971). Lasserre (1976), based on his own measurements and literature data, emphasized the large differences present, and reported values of between 500 and 1,700 ml for “high consumers” (harpacticoids) among the metazoan meiofauna, but only about 10% of this value (90–300 ml) for “low consumers.” The reduced consumption of the “low consumers” can be interpreted as an ecophysiological adaptation of euryoxic meiofauna, which are often found in oxygen-deficient habitats. For invertebrate macrofauna, values of only about 50 ml O2 h−1 g−1 wwt were measured. The resulting linear relation between weight and metabolism, calculated by various authors, is illustrated in Fig. 9.5b. Because of the divergent body masses in meiobenthic animals, respiration rates of various meiobenthic taxa range over orders of magnitude (Figs. 9.5b and 9.6a,b; Vernberg and Coull 1974). While the oxygen consumption of typical metazoan meiobenthos appears to mostly range between 1 and 10 ml O2 mg–1 dwt h–1 the interstitial ciliate Tracheloraphis consumed clearly more oxygen thus confirming the inverse size/respiration relation (Table 9.6a.) The divergent consumption rates
10.0 8.0
µl O2 x mg−1 dwt x h−1
(7)
6.0
8.0
4.0
6.0
2.0 (7)
(9)
(8)
(6)
4.0
(12)
(8)
1.0 0.8 0.6
(6) (8)
(10) (10)
2.0 (12)
0.4
(6) temperature C°
0.2
a
µl O2 x mg−1 dwt x h−1
10.0
0
5
10
15
20
25
1.0 0.8
b
salinity g / l 7
10 15 20 26
33 40 45
Fig. 9.6 a–b Respiration of Derocheilocaris remanei (Mystacocarida, Crustacea) at various temperatures (a) and salinities (b). (After Lassèrre and Renaud-Mornant 1971) Table 9.6a Respiration rates (µl O2 mg–1 dwt h–1) calculated for various meiobenthic groups and biotopes (conversion factors wwt to dwt: 0.2; C to dwt: 2.0) Taxon and remarks Respiration rate Reference Ciliate Tracheloraphis Metazoan Meiobenthos Macrobenthos Meiobenthos, “high consumers” Meiobenthos, “low consumers” Various harpacticoids Various harpacticoids “Heavy” nematodes, upper muddy tidal shore “Small” nematodes, lower sandy tidal shore Nematodes, Antarctic bight Nematodes, ostracods, foraminiferans, average value
22.5 7.5 0.25 2.5–8.5 0.45–1.5 2–14 6–6.5 2.9
Vernberg and Coull 1974 Gerlach 1971 Lassèrre 1976 Lassèrre 1976 Lassèrre 1976 Herman and Heip 1983 Vernberg et al. 1977 Tita et al 1999
2.75 3.2–4.9 4.0
Tita et al 1999 Vanhove et al. 1997 Moodley et al. 2008
Table 9.6b Respiration rates (µl O2 mg−1dwt h−1) calculated for various meiobenthic groups (Herman et al. 1985) Taxon or ecological group Respiration rate Nematoda Non-selective deposit-feeders Epigrowth feeders Omnivores and predators Ostracoda Cyprideis torosa Harpacticoida Canuella perplexa Mesochra lilljeborgi Tachidius discipes
1.77 2.56 3.78 1.78 3.72 10.47 12.59
9.3 The Meiobenthos in the Benthic Energy Flow
395
derived from direct measurements and literature data (Lassèrre 1976) also emphasize the large differences present among meiofauna. The reduced consumption of the “low consumers” can be interpreted as an ecophysiological adaptation of euryoxic meiofauna, which are often found in oxygen-deficient habitats. Invertebrate macrofauna with their much larger body masses consume per weight unit one order of magnitude less oxygen than meiobenthos. In compilation, there results a linear relation between weight and metabolism, calculated by various authors and illustrated in Fig. 9.5b. Despite low polar temperatures, the respiration rates observed for nematodes from a shallow Antarctic bight (Vanhove et al. 1997) stayed in the general range of Table 9.6a. For nematodes, Wieser et al. (1974) found a relation between oxygen consumption and trophic habits and discussed a proportionality between respiration rate and size of the buccal cavity: Bacteria-feeders (narrow buccal cavity), consumed less oxygen than predators (wide cavity). Heip et al. (1985) related nematode respiration to feeding types and calculated that selective deposit-feeders and epigrowthfeeders consumed less than half the amounts of predators or non-selective deposit-feeders, relations that were quite different from those given in Warwick and Price (1979). The proportional relation between body size and respiration was attributed to different habitat zones by Tita et al. (1999). The heavier nematodes from the upper shore with muddy sediments (mean individual biomass 0.8 mg dwt) consumed more oxygen than the smaller nematodes from the sandy low tide reaches (mean individual biomass 0.4 mg dwt). The recent respiration data by Moodley et al. (2008) were obtained with a novel, accurately measuring setup using a highly sensitive fluorogenic “optode” that allows measurements for individual meiofauna specimens of 1–5 mg dwt (ostracods, nematodes, foraminifera). The data obtained with the traditional Cartesian diver-technique (Lassèrre 1976; Vermberg et al. 1977; see below) yielded results in the same range. When we compare the overall metabolic contributions of microbenthos, meiobenthos and macrofauna, we observe wide variations. While it is clear that such variations occur between habitats and species, the metabolically important differences in turnover rates must also be considered—a factor often overlooked in annual calculations (Vernberg and Coull 1974). Warwick and Price (1979) suggested that, for a nematode community in an estuarine mud flat, each gram of nematode (wwt) respired about 6 L O2 per year and m2. However, only measurements of community respiration under in situ-conditions or in sediment microcosms, and if possible for long periods of time will reflect the natural interactions and disturbances (see for instance Lassèrre and Tournié 1984; Vanhove et al. 1997). In the deep-sea, these can best be performed in situ with a bell jar apparatus (Pfannkuche and Lochte 1990). The respiration rates even of small meiobenthic animals can be measured under variable environmental conditions (temperature, salinity) by either the traditional “Cartesian diver technique” (Lassèrre 1976, improved by Hamburger 1981; Shirayama 1992), the differential respirometer (Dixon 1979), or the polarographic microrespirometer using micro-oxygen electrodes in a closed system (Gnaiger
396
9 Synecological Perspectives in Meiobenthology
1983, 1991; Moens et al. 1996; Moens and Vincx 1997b). A thorough comparison of both methods was performed by Varó et al. (1993) and yielded comparable results using both gadgets. The most advanced microrespiration technique allowing non-invasive and exact measurements on a minute scale is based on optodes equipped with fluorogenic tips (Moodley et al. 2008; see Sect. 2.1.4). All respiration techniques are methodologically hampered by problems of low reproducibility and require careful calibration and confirmation via replicate runs. Respiration of the test animals often interferes with that of the bacteria developing in the test chamber, and must be adjusted through “blank runs.” The impact of methodological errors on the measurement of respiration has been reviewed by Malan and McLachlan (1991). As with other processes in living organisms, respiration depends on individual metabolism, with all its compensation reactions (oxyregulators) and irregular changes. Intermittent phases that use an anaerobic metabolism are surprisingly common in the meiobenthos, which confirms the statements of Sikora and Sikora (1982) and Revsbech and Jørgensen (1986) about the frequent anoxia to which most of the meiobenthos is exposed (see Sect. 8.4). Population density and excretion of metabolites (see Boaden 1989a) must also be considered when interpreting values of oxygen consumption. Different environmental factors (salinities, temperatures; Fig. 9.6) strongly influence respiration rates, as does handling of the animals (Vernberg et al. 1977). Moreover, and rather obviously, respiration rates vary considerably between the species in each taxonomic group (Table 9.6), making the extrapolation of results from single-species respiration experiments to the whole group or to community production problematic. The high weight-specific respiration rates of meiofauna (as compared to macrofauna) given by Gerlach (1971) are estimates, but they indicate that the metabolic rate of the meiobenthos is perhaps five times higher than that of the macrobenthos, i.e., meiofaunal weight equivalents consume five times more energy than the equivalent macrofauna do. The meiobenthic contribution has been found to be particularly high in organically enriched sediments (Duplisea and Hargrave 1996). In contrast, Kurashov (2002) reported that in freshwater lakes the production of the meiobenthos (calculated via respiration) did not depend on the trophic potential of the sediment. The technical complexity of measuring respiration supports the application of methods to calculate respiration mathematically using size, volume and weight or volume parameters. Once the animals have been measured, their weights (dwts) can be calculated using conversion factors (see Table 9.2). Assuming a linear relationship between weight and respiration (see above), the following formula usually applies (Heip et al. 1982a): R = a.dwtb (where a is the species-specific constant value of metabolic expenditure; b the weight-dependent regression coefficient, usually 0.75 for small poikilotherms, the “three quarters rule”).
9.3 The Meiobenthos in the Benthic Energy Flow
397
Logarithmically transformed: logR[nl O2 h−1] = log a + b log (dwt) Another pathway is afforded by the “biovolume method,” where the volume (V) is obtained from the weight and the following formula for mean respiration is used: logR = log a + b log V (coefficients a and b, derived for nematodes: log a = −0.052; b = 0.92, see Warwick and Price 1979). The production can be calculated from values for respiration using formulae suggested by McNeill and Lawton (1970): logP[kcal m−2y−1] = 0.8262 logR − 0.0948 or by Humphreys (1979): logP[kcal m−2y−1] = 1.069 logR − 0.601 Again using conversion factors, the resulting energy values can be expressed as weight units: 1g[ash-free dwt] = 5.6 kcal or 23 kj (Winberg 1971), or 25kJ (Schiemer 1982). Emphasizing nematodes: 1g[ash-free dwt] = 6.1 kcal (Sikora et al. 1977) = 5.3 kcal (Ceccherelli and Mistri 1991)
9.3.3
The Energetic Divergence Between Meiofauna and Macrofauna
Compared to the macrofauna, the meiofauna usually has a low standing stock biomass. Nevertheless even in tidal flats and in the shallow subtidal, where the macrofauna is relatively abundant, the shorter turnover of the meiobenthos generates a high production that frequently exceeds that of the macrofauna (Fig. 9.7). In beaches, meiofauna are even more clearly favored (Table 9.7). Contrastingly, in the upper reaches of a beach and in estuaries, the contribution of meiofauna plus microbes to the total metabolism can reach up to 97%. An increased metabolic relevance of the meiobenthos compared to the macrobenthos has also been recorded from deep-sea bottoms. Contrastingly, in freshwater streams, the contribution of meiobenthic organisms to metazoan biomass and production is rather small (Hakenkamp and Morin 2000) and the predation pressure exerted by macrobenthic organisms on meiofauna is low (Schmid-Araya and Schmid 2000).
398
9 Synecological Perspectives in Meiobenthology 80
60
Meiofauna Macrofauna
40
20
0 %
a
Biomass
Production
80
60 Bacteria Meiofauna Macrofauna
40
20
0 %
b
Biomass
Production
Fig. 9.7a–b Comparison between bacteria, meiofauna and macrofauna in terms of biomass and production. a Intertidal silty sand flat, Canada. b Generalized calculations from numerous data sets. (a After Schwinghamer et al. 1986; b After Warwick et al. 1979)
Calculations of meiofauna (with nematodes and harpacticoids dominating) indicate that they consume twice the amount of carbon and produce four times as much as the ambient macrofauna, even though the meiofauna comprises only half the biomass of the macrofauna (Warwick et al. 1979). Nematodes have a turnover rate that is 21 times higher than that of the macrobenthos in the same biotope. For an average benthic marine biotope, it has been generalized that the meiobenthos produces between 10 and 25% of the total energy budget and the microbenthos (protozoans and microalgae) adds at least the same proportion, while the macrobenthos plays a subordinate energetic role (see Table 9.7; Munro et al. 1978; Schwinghamer et al. 1986). Hence, the data presented in Fig. 9.7a for a tidal mudflat are typical of the inverse relations between weight and production in meiofauna and macrofauna. The production and flow of energy is the domain of
9.3 The Meiobenthos in the Benthic Energy Flow
399
Table 9.7 Comparison of macrofauna and meiofauna in terms of biomass and production Productive Biomass Production P/B ratio share (%) A. Exposed beach, South Africa Macrofauna Meiofauna Bacteria
(g dwt beach m−1) 241 227 663
(g dwt beach m−1y−1) 603 2,267 19,892
B. Tidal flats, Netherlands
(g C m−2)
(g C m−2 y−1)
Macrofauna Meiofauna + Microfauna
27 1
27 65
C. Tidal flats, salt marsh, Canada
(kcal m−2)
(kcal m−2 y−1)
Macrofauna Meiofauna Bacteria
14 12 0.5–5
29 171 150–1,500
D. Sublittoral silty sand, generalized
(g dwt m−2)
(g dwt m−2 y−1)
Macrofauna, depositfeeders Meiofauna Microfauna + bacteria
6
E. Mirror Lake, US, oligotrophic
2.5 10 30
3 10 87
1 65
29.3 70.7
2 12.5 292
≈2 ≈ 10 ≈ 86
12
2
9.4
1.5 5
11 104
10 20
8.7 82
(g dwt m−2)
(g dwt m−2 y−1)
Macrofauna 1.8 9.1 5.0 65 (insect larvae) Meiofauna 0.6 4.9 8.2 35 Note: The five subtables A– E have differing units of measurement. A is from Koop and Griffiths (1982); B is from Kuipers et al. (1981); C is from Schwinghamer et al. (1986); D is from Gerlach (1978); E is from Strayer (1985)
the smaller benthos, the storage of energy is characteristic of the slower-growing macrobenthos. This argument indicates that the more conspicuous macrofauna has only a moderate productive significance, an aspect often not adequately considered in ecology textbooks. More detailed reading: Gerlach (1971, 1978); Warwick et al. (1979), Banse and Mosher (1980); Banse (1982); Heip et al. (1982a), Heip et al. (1985b), Crisp (1984); Schwinghamer et al. (1986); Feller and Warwick (1988); Brey (1990); Warwick and Clarke (1993a); Moens and Vincx (1997b); Danovaro et al. (2002); Van der Meer et al. (2005); freshwater, Plante and Downing (1989); Bergtold and Traunspurger (2006)
400
9 Synecological Perspectives in Meiobenthology
Box 9.2 Meiofauna: Highly Productive? How Do We Know? Organisms such as meiofauna are beyond our normal range of perception, and so their energetic role is usually underestimated. Among the parameters required to assess the energy budget of an organism (Production = Consumption – Respiration – Excretion) only P and R can be reasonably accurately determined for animals as small as meiofauna, while quantitative information on C and E is scarce. Direct laboratory cultures of a few species provide the basis for life-history tables from which generation time, turnover and biomass values are derived. But does a laboratory offer natural conditions? Length and weight data are used to calculate production using formulae and conversion factors that are often derived from macrobenthos. Another approach is to measure respiration and then extrapolate these data to production. But what about regulative metabolism, intermittent anaerobiosis and fluctuations in oxygen uptake under variable environmental conditions? The high innate divergence of generation times among species and the variable impacts of environmental and physiological factors introduce a high degree of uncertainty into measurements of production. The yearly productive capacity of the meiobenthos is often compared with that of the macrobenthos. However, without distinct cohorts, without knowledge of the reproductive inputs and predatory losses in natural systems, what is the real productive output of a given population or even a community after a year? Mathematically, the repeated use of generalized conversion factors multiplies inherent inaccuracies. Aware of these pitfalls, we need to avoid estimates, extrapolations and exceptions and move towards more reproducible results of predictive value. Energy flow does not show up in the low biomass of meiobenthic organisms. However, numerous comparative benthic surveys from marine and lacustrine areas, shallow and deep-water sites, and from sands and muds, all suggest a significant contribution of the meiobenthos to production and energy flow. Its throughput of carbon can be orders of magnitude higher than its standing stock. This critical note shall encourage reliable quantitative investigations, preferably combining laboratory and field/community studies, that cover the entire benthic spectrum.
9.4
The Position of Meiofauna in the Benthic Ecosystem: A Compilation of Energy Fluxes
The connections between faunal compartments in ecosystems are usually illustrated by energy flux diagrams. Assessments of the role of the meiofauna in these diagrams, however, are hampered by a lack of ecological information on many meiobenthic groups. As stated above, data on life history and trophic demands cannot be generalized since they are obtained from only a few species. Generalizations
9.4 The Position of Meiofauna in the Benthic Ecosystem
401
for entire benthic groups like “nematodes” or “meiofauna” entail a high degree of inaccuracy. Energy webs in which meiofauna are linked to almost all trophic compartments (compare Elmgren 1978) are of little insightful value. The nutritional basis of a faunal community or the trophic value of a sediment obtained from bulk measurements of nutrients (e.g., total organic carbon and nitrogen) cannot give a sufficiently differentiated estimate because of the trophic specificity and trophic flexibility of many meiofauna. In order to overcome over-simplified and often biased generalizations, more refined methods have been developed and are increasingly being applied (e.g., microsensors in respiration experiments). One modern tool used to assess the transfer of energy through meiofauna is the application of fluorescent tracers to potential food items and subsequent analyses of gut contents (e.g., bacteria: Epstein and Rossel 1995b, Starink et al. 1994a, 1996; prey for macrofauna: Feller 2006). The specific composition of fatty acids in meiofauna, the “fatty acid signature,” is another component that is increasingly being used to reveal and quantify the flux and consumption of food of meiobenthic organisms. Different food items (diatoms, bacteria, protists, meiofauna) have different structures of fatty acids (especially of polyunsaturated fatty acids, e.g., C18:1ω-9 for bacteria, C20:5ω-3 for diatoms). These fatty acid structures can serve not only as qualitative markers but also as quantitative determinants of dietary composition (Giere et al. 1991; Iverson et al. 2004; Caramujo et al. 2008; Leduc 2008). Measurements of isotopic quotients (δ 13C and δ 15N) allow us to trace and even quantify the pathway of trophic energy. In invertebrates, each transfer from one trophic level to the next will enrich the “isotopic signature” by 2.2% (in vertebrates by 3.4%). However, is this often-reported value, based on DeNiro and Epstein (1978, 1981), a reliable estimator for dietary analyses? Since the actual degree of enrichment from prey to consumer shows considerable discrepancy from the theoretical value, these C- and N-quotients should be taken as guidelines only. For instance, the relatively light δ-values of North Sea copepods suggest a considerable amount of chemoautotrophic bacterial food (Franco et al. 2008). Also, techniques using labeled food (e.g., with 13C) can indicate the trophic cascades in the natural environment (Van Oevelen et al. 2006a,b) and can even differentiate between the meiofauna in different sediment layers. Nematodes in the surface sediment preferred freshly deposited detritus, while deeper-dwelling nematodes fed on older material. The nematodes of a tidal mudflat studied by Riera et al. (1996) preferred microphytobenthos to sedimenting detritus, but in another tidal mudflat C and N isotopes indicated that nematodes and harpacticoids ingested more detritus than microphytobenthos (Couch 1989). Labeling with radioactive isotope tracers such as 14 C-carbon dioxide/3H-thymidine or 14C-acetate/3H-thymidine permits the quantification of trophic uptake rates (Montagna 1983; Montagna and Bauer 1988). These authors demonstrated that meiofaunal communities affect bacterial numbers to only a limited degree, consuming just 3.6% of their biomass. Although radioisotopic techniques have some methodological pitfalls (bias from bulk freezing: Dannheim et al. 2007) and interpretational ambiguities, especially when measured under field conditions, the results do agree that meiofauna grazing does not have a considerable overall effect on bacteria and microalgal production (see below).
402
9 Synecological Perspectives in Meiobenthology
An early method of tracing the trophic relations between macro- and meiobenthic species was based on a modified serological antigen–antibody reaction (Feller et al. 1979; Feller and Warwick 1988; Feller et al. 1990). This method used antigens precipitated from the stomach contents of suspected predators with antibodies developed (in the serum of rabbit blood) for possible meiobenthic prey organisms (in rabbit blood serum). Due to improved methods of extracting DNA from small individuals (Schizas et al. 1997), the most promising and widely applicable tools used in the study of food relationships and energetic transfer today are molecular biology and genetic probes that can directly trace the organisms involved. The value of molecular biological methods is that they can be applied to animals taken from the field or kept under largely natural conditions. Nutritional selectivity, a characteristic of many meiofauna, and one that differs even among closely related species, can locally and temporarily modify average values of meiofauna consumption and has been reported to sometimes even exhaust a particular trophic source. Denaturing gradient gel electrophoresis (DGGE) studies on bacterivorous nematodes even showed a selective uptake of bacteria. The worms could change their selected diet when mixed cultures were offered. Trophic specialization was directly shown in feeding experiments with different prey species (turbellarians) (Watzin 1985; Sect. 2.2.7). On the other hand, microcosms with several species of (monhysterid) nematodes yielded a different food uptake than single species setups (De Mesel et al. 2004). Trophic selectivity and flexibility is advantageous from an evolutionary perspective since it may prevent interspecific competition among both meiofaunal species and their prey. However, it increases the difficulties to quantify food uptake.
9.4.1
The Meiofauna as Members of the “Small Food Web”
Meiofauna, microorganisms and detritus have mutually facilitating and inhibiting links that closely integrate meiofauna into a “detrital trophic complex” (Tenore and Rice 1980) with numerous transitions and interactions. The interwoven energetic pathways in this “small food web” (Kuipers et al. 1981; Fig. 9.8) can only be analyzed when they are differentiated using refined methods. Detritus. The central role of detritus in the energy budget of meiofauna (Escaravage et al. 1989; see Sect. 2.2.1) is demonstrated by the position of the most important meiobenthic group, the nematodes, which are considered to be primarily linked to the detritus/bacteria-based food chain. Freshly sedimented phytodetritus derived from phytoplankton seems to be so nutritionally valuable that it is preferred by many meiofauna, which provides an interesting link to the pelagic ecosystem (Ólafsson and Elmgren 1997; Buffan-Dubau and Carman 2000). Seagrass detritus, an important detrital component, is both mechanically and chemically degraded by meiofauna (nematodes). This process can enhance the mineralization process by up to 50% and increased the incorporation rate by polychaete macrofauna considerably (Tenore et al. 1977; Findlay and Tenore 1982). Hence, under field conditions, the detrital complex
9.4 The Position of Meiofauna in the Benthic Ecosystem
Organic input
403
Primary production
Nutrients
Macrobenthos
Small Macrofauna Larvae and juveniles Meiofauna Microfauna
Carnivores
Bacteria The ⬙small food web
Fig. 9.8 The small food web, integrating meiofauna in a substantial compartment of the benthic energy flux. (After Kuipers et al. 1981)
is tightly linked to the activities of the other partners in the detrital food web, especially microorganisms. Watling (1991) argues that much of the total organic carbon is bound into humic polymers and not available to the animals. He suggests techniques that can separate the digestible fractions of the bulk parameter “detritus.” Also, among the amino acids, only the enzymatically degradable ones are relevant. Bacteria. Whether in marine shallow sites or in creeks and lakes, bacteria contribute a huge amount of production, which is usually not limited by the consumption of meiofauna, nor does it control them. In a North Sea tidal flat, enrichment experiments with 13C (Van Oevelen et al. 2006a,b) showed that bacterial production primarily represents a sink in the food web. Most bacterial biomass is respired or remineralized, the valuable nutrients being recycled back to bacterial biomass. Only 3% are consumed by the meiobenthos and about 20–24% by the macrobenthos, respectively. Metazoans cover <10% of their carbon requirements with bacterial carbon. For (estuarine) meiofauna, Heip et al. (1995) stated in their Table 26 that only 0.5% or less of bacterial and algal stock was removed per hour. A marginal grazing rate of just a few percent of the microbial stock has also been calculated from many other marine sites (Epstein 1997a; Kemp 1990; Montagna 1995; Danovaro 1996). Similarly, in freshwater ecosystems (streams, creeks and lakes), only a small fraction of the bacterial production is ingested (Borchardt and Bott 1995; Bergtold and Traunspurger 2005; Traunspurger et al. 1997). Although methodological inaccuracies
404
9 Synecological Perspectives in Meiobenthology
may contribute to a significant underestimate of meiobenthic grazing on bacteria (Moens et al. 1999c), the meiobenthos as a whole appears to rely on carbon sources other than bacteria, so that the bulk of the bacterial production is not transferred to higher trophic levels. This is not to say that momentary grazing rates are not impressive at times, e.g., 25% per day (Carman, pers. commun.), 10–40% (Moriarty et al. 1985), 5,000–10,000 bacteria per nematode and hour (Traunspurger, unpubl.). However, the overall impact on the bacterial assemblage is small because of their rapid turnover rates (hours or days). Perhaps deep-sea bottoms, with their reduced bacterial production, are an exception. Here, one can postulate a structuring effect of meiofaunal grazing (see Sect. 8.3). But what about the regulatory impact of meiofauna on bacteria, as found in several studies (Nilsson et al. 1991; Riera et al. 1996)? Indirect effects, e.g., competition between different faunal compartments (e.g., ciliates and meiofauna) for the common food resource of “bacteria” (Bergtold and Traunspurger 2005) are probably of relevance. Grazing of meiofauna on bacteria stimulates bacterial activity by keeping the population in its reproductive log-phase. Also, meiofaunal excretions are known to stimulate bacterial growth (Walters and Moriarty 1993; Carman 1994; Montagna 1995; Traunspurger et al. 1997). The amount of preying on bacteria varies in relation to the availability of other food sources (e.g., microphytobenthos), so that seasonal shifts in grazing intensities and local aggregations of bacteria can complicate the overall picture. The various interactions between bacteria and meiofauna can be summarized as follows: - Meiofaunal grazing has a stabilizing effect on bacterial populations, removing senescent cells and maintaining cell growth in an exponential phase (Yingst and Rhoads 1980; Meyer-Reil and Faubel 1980; Montagna 1995; Traunspurger et al. 1997). Secretion of N- and P-containing dissolved metabolites of meiofauna fosters the growth of bacteria. The bacterial stock, of which normally only a minor percentage is metabolically active, is stimulated by harpacticoid aggregations (Carman 1994). - “Gardening” effects of meiofauna enhance bacterial stocks. Excreted mucus, which is colonized by microorganisms, are cropped by meiofauna (Riemann and Schrage 1978; Moens et al. 2005; de Troch et al. 2005). - Meiofaunal bioturbation activates geochemical fluxes (Aller and Aller 1992; Murray et al. 2002). In particular, the diffusion rate for oxygen becomes activated, enlarging the oxic habitats of aerobic bacteria and many meiofauna. Through the minute mucus-stabilized burrows and trails of nematodes, ostracods and harpacticoids, bacterial growth is promoted (Fenchel 1996). - The mechanical breakdown of detrital particles by meiofauna supports bacterial decomposition (Tenore et al. 1977). - Decaying meiofauna serve as food for bacteria. - Specialized meiofauna serve as a symbiotic partner for bacteria (see Sect. 8.4.2). Summarizing, meiobenthic communities are tightly, though often indirectly, coupled to bacteria. Their grazing rates are in equilibrium with the rapid microbial turnover, which implies that meiobenthos can significantly affect microbial community
9.4 The Position of Meiofauna in the Benthic Ecosystem
405
structure, and, in turn, bacterial spatial and temporal fluctuations will influence the distribution and abundance of meiobenthos (Walters and Moriarty 1993; Montagna 1995; Traunspurger et al. 1997). Microalgae. Meiofauna are often found to feed preferably on benthic microalgae (mainly diatoms). While harpacticoids are usually regarded as phytobenthosbased (see Rudnick 1989), nematodes also locally feed on diatoms as their main food source (Riera et al. 1996). At least temporarily, the severe grazing pressure exerted by meiofauna with the ingestion of a significant portion of the diatom production may reduce diatom growth (Hentschel and Jumars 1994; Borchardt and Bott 1995), thus controlling the microphytobenthic stock sporadically (Kemp 1990; Blanchard 1991; Montagna 1995). On the other hand, in seagrass beds, where microphytobenthos can account for 25% of the carbon production (Danovaro et al. 1996), the microalgal biomass seems to be so high that it is hardly affected by meiobenthic grazing (Walters and Moriarty 1993). Using labeled carbon, Middelburg et al. (2000) determined that microalgae-derived carbon was rapidly transferred to all members of the benthic community in intertidal flats. Maximum transfer to bacteria occurred in one day. This important field experiment underlined the central role of microphytobenthos in moderating the carbon flow in coastal sediments. However, the results did not suggest that the microalgal stock was controlled by meiobenthos. Overall, the conclusion that diatoms are of limited trophic significance for meiobenthos (Montagna et al. 1983; Nilsson et al. 1991; Epstein 1997a) is hard to generalize and quantify considering the manifold trophic specializations of meiobenthos on microalgae (Rzeznik-Orignac et al. 2008). Ciliates. The trophic role of ciliates as a link mediating between bacteria and meiofauna is ambiguous. Intensive grazing effects on bacteria were recorded from clean sand, but not from mud; massive seasonal and local variations are typical (Kemp 1988). In streambeds, ciliates grazed 15% of the bacterial production (Kemp 1990). Despite their strong grazing on bacteria, often accounting for 10–25% of the bacterial stock (Epstein et al. 1992; Kemp 1990), ciliates rarely seem to exert a direct control. Ciliates also graze intensively on flagellates, removing on average 17% of the standing stock (Epstein et al. 1992). Although ciliate grazing on the diatom stock is on average only 6% of the total stock, some massively ingested microalgal species can be selectively controlled by ciliates. Higher up in the trophic web, ciliates are massively preyed upon by meiobenthos, which can regulate the growth of protists (Epstein and Gallagher 1992; Hamels et al. 2002). This grazing pressure caused a patchy distribution of meiofauna attracted by aggregations of ciliates (Blackburn and Fenchel 1999). On the other hand, pronounced fluctuations in meiobenthic abundance may be caused by food shortages. In experiments with freshwater species, indications of trophic competition between ciliates and nematodes for the bacterial food they have in common have been found (Bergtold et al. 2005). The complex interactions within the “small food web” outlined above are a characteristic that emerges from all pertinent studies (Figs. 9.7, 9.8). Nevertheless, the results are partly contradictory and mostly only locally valid, so they often defy attempts to summarize and generalize them. Fenchel (1978) emphasizes the productive importance of the small benthos, stating that the micro- and meiofauna in
406
9 Synecological Perspectives in Meiobenthology
the sandy sublittoral contribute 40% of the total benthic metabolism. Meiofauna can temporally and locally structure microbenthic communities, but often the pathways are indirect and variable. Montagna (1995) summarized: “it is certain the meiofauna will never run out of food;” “bacterial production and turnover is more than sufficient to satisfy the demands of the meiofauna population,” and “meiofauna are not grazing so fast that they are limited by microalgal food.” Nevertheless, it is clear that the meiofauna has an impact on microbiologically mediated processes. But is the small food web a “parallel world” that lacks ties to the macrobenthos? We saw in the previous chapters that such ties are often concealed or indirect. Any differentiation beyond simple straightforward predator–prey relations makes it fairly complicated to study them. More detailed reading: Tenore (1977a); Fenchel (1978); Tietjen (1980a); Montagna et al. (1983); Montagna (1995); Kemp (1990); Blanchard (1991); Heip et al. (1995); Danovaro et al. (1996); Epstein (1997a,b); Van Oevelen 2006b
9.4.2
Links Between the Meiofauna and the Macrofauna
When we compile the rich literature on meiofauna–macrofauna interactions, we find evidence of two contrasting contentions. (1) According to the “dead-end hypothesis,” the energy fixed by meiobenthic organisms is often not transmitted significantly to higher trophic levels (McIntyre 1969), viz., to the benthic macrofauna and mobile nekton. As a consequence, meiobenthic communities appear to be regulated by bottom-up effects. (2) According to the “integration hypothesis,” a great deal of the meiofaunal trophic energy is channeled to higher levels, and the populations are massively reduced through the consumption by macrofauna. Thus, since it is an integrated part of the trophic flux to the macrobenthos, the meiobenthos is mostly regulated by top-down effects. So which of these contentions is correct? There is good scientific evidence for both! Again we must differentiate. According to the “dead-end conception,” the meiofauna only maintains close energetic links to the detritus/bacteria complex discussed above (see Sect. 2.2), while most of the predatory macrobenthos feeds on small macrobenthic prey or indiscriminately ingests microorganisms together with the sediments taken up. Only a small proportion (20%?) of the trophic requirements of macrofauna are covered by consuming meiofauna (Gerlach 1978), which is not large enough to regulate meiobenthic communities, considering their intense production. Many subsequent studies have confirmed a weakly or moderately negative effect of macrofauna, especially of epibenthic predators, on most meiofauna (among others: Reise 1979; Service et al. 1992; Nilsson et al. 1993; Kennedy 1993; Schrijvers et al. 1997; from freshwater: Hakenkamp and Morin 2000; Schmid-Araya and Schmid 2000). However, among many other authors, Coull and Bell (1979) have argued that these conclusions cannot be generalized. Before we turn to the “integrated hypothesis,” we should consider the “general foraging theory,” which can elucidate some features involved in this controversy.
9.4 The Position of Meiofauna in the Benthic Ecosystem
407
Trophic relations always evolve as a trade-off between the energy gained by the consumer from the food ingested and energy used by the consumer for to gain the food. The size- or (better) weight-related energetic effort determines the degree of predaceous interactions. A general scaling rule says that the sustenance of one predator with a weight unit of 1 requires about 100 weight units of prey (Carbone and Gittleman 2002). Therefore, a relative size coherence between predator and prey is advantageous and usual, while big disparities are exceptional. According to the “packaging model” (Lackey 1936), for an animal of intermediate size between the macrofauna and the meiofauna, i.e., in the size range of 5 to about 15 mm, it is energetically much more parsimonious to ingest one meiofaunal organism rather than numerous dispersed microorganisms of (combined) the same energetic value (Sikora et al. 1977). The concentration of trophic energy in one larger meiobenthic particle could even compensate for the loss of energy usually attributed to shifting from one trophic level to the next (approximating one order of magnitude). Thus, small macrofauna and nekton (small bottom-feeding fish species, juvenile stages of larger fishes or crustaceans) will get an energetic advantage from preying on meiofauna. Many juvenile fish prefer and grow well by feeding on harpacticoid copepods, but this prey would not be profitable for an adult fish. Many young shrimps (length below 10 mm) prey voraciously on meiofauna, while the adults prefer larger food items (Nilsson et al. 1993). This ontogenetic shift in food preference has been documented many times (see reviews by Gee 1989; Coull 1990, 1999). The ecological rules of the energy trade-off mentioned above apply to the directed uptake of single prey items. Of course, they do not apply to the massive uptake of food achieved through bulk ingestion (deposit-feeders) or trapping (mucus secretion). Numerous field and laboratory studies demonstrate a clear predatory impact of juvenile fish on mainly harpacticoid meiofauna (see reviews by Coull 1990, 1999; Heip et al. 1995). From enclosure experiments in the North Sea tidal flats, Reise (1987) concluded that, among several other factors, it was predation pressure by macroinverbetrates that structured the populations of meiobenthos. From experimental and field studies in the eastern Baltic Sea emerged a strong and specific predation of small fish on meiofauna (Aarnio and Bonsdorff 1993). Experiments in Australian and African mangrove muds (Coull et al. 1995; Olafsson and Ndaro 1997) demonstrated that harpacticoids were strongly preyed upon in the surficial layers at both locations, while nematodes, especially those in the deeper layers, were less affected in Africa, but were heavily preyed upon in Australia. An ontogenetic shift in meiofauna prey selection has been noted in the diets of many postlarval fish. The youngest stages fed on epi- and hyperbenthic meiofauna (mainly harpacticoids), while the older fish preferred the meiobenthos in the sediment (McCall and Fleeger 1995). Here, the initially preferred harpacticoid species were later replaced by other harpacticoid species which were then supplemented with nematodes. Pihl (1985) quantified the meiobenthic prey consumed by small fishes and shrimps: up to 60% of the harpacticoid copepods and 90% of the ostracod standing stock were ingested. Danovaro et al. (2007) estimated that at Mediterranean sites studied in exclosure experiments, about 75% of the metazoan meiofauna in soft bottoms is channeled to higher trophic levels.
408
9 Synecological Perspectives in Meiobenthology
How should we interpret the controversial scenarios described here? The results do not exclude each other because there are many more levels of interaction between meiofauna and macrofauna than just predation. However, even the nutritional aspects per se must be differentiated and the circumstances considered (Coull 1990). - The methodological problem: Manipulative experiments in the field (exclosure gadgets) may influence factors other than predator exclusion, such as sediment structure and current regime, leading to erroneous conclusions (“caging effect”). Laboratory experimental systems are inherently artificial. - The biological differences: It appears from most studies that, among the prevalent meiobenthic groups, harpacticoid copepods and ostracods are preferred by most predators, although the nutritive value of nematodes is probably at least as high as the values for these two groups (e.g., Nilsson et al. 1993; Yozzo and Smith 1995; Ólafsson and Ndaro 1997; but see Coull 1999 for a discussion of probable fatty acid differences in copepods and nematodes). In contrast to the writhing burrowers, such as nematodes, crustaceans’ flitting motions in the more surficial sediment layers probably make them attractive and available to predators. In addition small annelids that are temporarily and permanently of meiobenthic size are also preferred by predators, especially at muddy sites (Kennedy 1993), while turbellarians seem little affected by consumption. - Nature of the predator: Vision- or touch-oriented predators (young epibenthic fishes, shrimps, some wading birds, chironomid larvae in freshwater) grasp meiobenthic organisms selectively, while deposit-feeding crabs, priapulids, polychaetes, ophiurids and some birds plough through the upper sediment layers, ingesting all of the digestible matter present rather indiscriminately (e.g., Schrijvers et al. 1997). These different predatory patterns will affect the meiobenthic assemblages differently. Populations from the deeper strata are usually less affected/reduced than the surface dwellers (Tita et al. 2000b). - Influence of habitat structure: In the more permeable and oxygenated sandy habitats, meiofauna occur fairly homogeneously in the upper centimeters, while oxygen depletion in muds concentrates most meiofauna into the layer very close to the surface. This would explain the reduced predatory impact in most studies from sandy sites compared to muds (e.g., Smith and Coull 1987; McLachlan and Romer 1990). The inverse relation reported by Schratzberger and Warwick (1999) for Carcinus may be caused by the concurring effects of predation and disturbance. In sediments with tubes and culms protruding from it, predation pressure by macrofauna is less intense; increased habitat complexity has a protective effect (Walters et al. 1996). On rocky surfaces with their high spatial variability (algal cover), predator exclusion did not affect the number of meiobenthic taxa, while in the experiments in muds the number and taxa and the average abundance increased in muds, indicating a higher impact of predation (Danovaro et al. 2007). Aside from direct predation, macrofauna consume a great deal of meiofauna by indirect “bulk uptake.” Many infaunal macrobenthos consume a considerable amount of meiobenthic animals, indiscriminately ingesting them with the sediment. Feeding behavior, analyses of esophagus content and exclosure experiments have shown that
9.4 The Position of Meiofauna in the Benthic Ecosystem
409
meiofauna can be temporarily important in the diets of wading birds and water fowl on beaches and mud flats, resulting in significant reductions in meiofauna (Sutherland et al. 2000). Specific structural adaptations of the bill and tongue in sandpipers (Calidris) even seem to support feeding on the meiofauna adhering to biofilms (Elner et al. 2005). Gaston (1992) analyzed the esophagus contents of some ducks after 20 min of feeding in an intertidal mud flat. Several thousand epibenthic harpacticoids and ostracods were apparently selectively ingested per bird. The meiofaunal contributions to the diets of the huge flocks of migratory wading birds passing through the North Sea tidal flats twice a year need to be investigated in quantitative detail. Competition for foods that macro- and meiofauna have in common has been suggested to lead to a non-predatory reduction. When small clams (Macoma) grazed down a microalgal cover, the meiofauna was clearly reduced (Ólafsson et al. 2005; see similar conclusion by Carlén and Ólafsson 2002). In Australian coral reef sand, an inverse relation between harpacticoid and holothurian density was interpreted as competition for microalgae (Moriarty et al. 1985). An indirect effect is the increase in the uptake rate of macrobenthic deposit-feeders in the presence of meiobenthos. This nutritive interaction appears to be related to the mechanical and chemical degradation of detritus by meiofauna (Tenore et al. 1977b; Tenore and Rice 1979). Meiofauna breaking down macrofaunal fecal pellets mechanically enhance the bacterial decomposition rate and thus support the regeneration of a nutritious sediment for deposit-feeding macrobenthos. Intensive bioturbative disturbance can reduce meiofaunal abundance considerably (see Sect. 2.1.3). An increase in chemical fluxes and structural complexity are proposed to be responsible for the reversed effect, macrofaunal activity promoting meiofaunal diversity and abundance (see Sect. 2.2.7). The fluctuations in predatory interactions are well demonstrated by a transitional interaction between permanent meiofauna and larval macrofauna termed the “bottleneck effect” (Bell and Coull 1980): Newly settled macrobenthic juveniles, particularly polychaetes, molluscs and crustaceans developing from planktonic larvae represent temporary meiofauna. In this phase they are directly exposed to the strong feeding pressure exerted by the permanent meiofauna (e.g., in the Baltic Sea, see Elmgren 1978). Meiofauna are known to be very effective utilizers of food (Warwick 1989), reducing macrobenthic offspring considerably. As the recruits grow, the situation reverses and many macrobenthic juveniles begin feeding on meiofauna (Reise 1979; Bell 1980; Watzin 1985, 1986; Tipton and Bell 1988; Gee 1989; Coull 1990; Service et al. 1992). However, in experiments designed to directly test the bottleneck effect, to scrutinize the predation pressure exerted by settling or recently settled macrofauna on meiofauna, Zobrist and Coull (1992, 1994) found no evidence of a bottleneck effect. On the other hand, the results of field studies in the western Mediterranean (Danovaro et al. 1993) can be interpreted as consequences of the bottleneck effect. Just after the settling of macrobenthic recruits (temporary meiofauna), predaceous nematodes and turbellarians increased, while the stock of juvenile polychaetes collapsed. Experimental evidence for a negative interaction (predation or/and disturbance?) between settling larvae of the polychaete Hydroides and populations of the omnivorous harpacticoid copepod Tisbe japonica (Dahms et al. 2004) emphasizes
410
9 Synecological Perspectives in Meiobenthology
the inhibition effects between macro- and meiofauna. Whether meiofauna exert a structuring effect on new settlings of macrofauna is debatable. The magnitude of the bottleneck effect may well depend on local circumstances. For conclusions on the regulative impact of macrofauna on meiofauna, we probably need to differentiate. Only in particularly clear cases is the transfer of meiofaunal energy to higher levels massive enough to reduce meiofauna. Where meiofauna (mostly motile harpacticoids and ostracods) are concentrated in the surficial layers, like in muddy sediments, and easily accessible to numerous predators (small fish, shrimps, birds), the predatory pressure is direct and intensive (Smith and Coull 1987). In many other cases (sandy substrates, deeper layers, sites with protective structures), the consumption rates by macrofauna are low and the regulative impact remains moderate. The “dead end hypothesis” holds here: the meiobenthos represents a rather isolated biotic element whose biomass is mostly remineralized in a short-cut cycle rather than transferred to macrofaunal levels. In these cases, direct competition and predatory pressure between the two benthic groups remain rather irrelevant. While we would like to extrapolate and possibly even quantify the roles of these two scenarios in meiobenthic energy transfer, we lack reliable data from field observations that separate predation from disturbance or other indirect effects. Results from experimental setups in culture tanks cannot substitute for field data, but they can be used to refine such data. We need observations that are performed frequently over periods long enough to account for tidal or diurnal changes, at sampling and control sites that are spaced closely enough to circumvent the smallscale patchy aggregations typical of many meiofauna. More reliable than assessing the increase in meiofauna “a negativo” after the exclusion of macrofauna is the direct analysis of the predators’ gut contents. These tedious analyses must be performed quickly enough to match the rapid digestion processes—nematodes are digested within an hour, leaving no traces (Scholz et al. 1991). In the complex food web, it appears that there are no simple answers to whether top-down or bottom-up control applies, whether meiofauna represent an energetic dead end or a trophic source for macrobenthos. More detailed reading: Gerlach (1978); Reise (1979); Bell (1980); Gee (1989); Coull (1990); McCall and Fleeger (1995); Walters et al. (1996); Hakenkamp and Morin (2000); Carbone and Gittleman (2002).
9.4.3
Meiofauna as an Integrative Benthic Complex
Considering the complexity of a specialized benthic food web, can we define the position and the role of the meiobenthos? Only circumstantial answers are possible, since “the” benthic food web does not exist. Shaped by exposure, sediment type, food supply and other environmental variables, the benthic communities are so different that generalizations are problematic. As we saw before, predation efficiency can change in each species with food availability, season, tide and sediment
9.4 The Position of Meiofauna in the Benthic Ecosystem
411
structure. Indirect effects can blur the impacts of more direct relations. One general principle, however, is that of scales and of size spectra. If size disparities between consumers and prey become too large, the direct preying interactions will loosen, and so the food web tends to become “compartmentalized” and fairly independent sub-webs will develop (Woodward et al. 2005). The meiobenthos is known to be tightly linked to the “small food web” (Kuipers et al. 1981), confirming the scaling rule mentioned above. The meiofaunal biomass can be similar to the bacterial biomass, but most of the bacterial biomass is not transferred to higher levels, so the microbial food web represents a sink of energy (Kemp 1990). Menn (2002a) studied the benthos and the food relations in boreal shores, comparing erosive and dissipative shore types (see Sect. 2.1.3). In the erosive beach, the “small food web” dominated with a diverse meiofauna (mainly turbellarians and harpacticoids) and without close links to the scarce macrofauna (Fig. 9.9). Conversely, the dissipative shore, with its rich organic matter, was characterized by tight links between the preying macrobenthos/shorebirds and meiofauna
Erosive beach organic
supply
impoverished ⬙large food web
diverse ⬙small food web high organic fluxes- no pool
low C/N ratio oxic conditions
Dissipative beach
pply
ic su
n orga
rich ⬙large food web
nematode-dominated ⬙small food web
high C/N ratio
organic pool
anoxic conditions
Fig. 9.9 Integration of the meiobenthos into the food webs associated with different types of tidal shores—a comparison of erosive and dissipative beaches. (After Menn 2002a)
412
9 Synecological Perspectives in Meiobenthology predation (-)
Juvenile Macrofauna
sediment changes (-) structure effects (+)
Adult Macrofauna
nu trie nt
-) n( tio da -) n( pre tio e ti mp co
reg en str era uc tio tur n( ee +) ffe cts pre (+) da dis tion ( loc ati ) on (-)
gregarious settlement (+)
Permanent Meiofauna
Fig. 9.10 Schematic compilation of the positive and negative interactions between meiofauna, temporary meiofauna (juvenile macrofauna), and macrofauna. (After Bell and Coull 1980)
(mainly nematodes). In these marine biotopes, with their rich fauna of various sizes, the more accessible meiobenthos is tightly linked to juvenile and indirectly to adult macrobenthos/nekton (Figs. 9.9, 9.10), so the meiobenthic energy will be channeled to higher levels to a considerable extent. Depending on shore types, these links would put the meiobenthos in a central position in the entire benthic food web, connected to both the macrofauna and the complex of detritus and microorganisms (Fig. 9.11). Moreover, these interconnections are known to enhance the stability and carrying capacity of this ecosystem (Woodward et al. 2005), which could explain the high resilience of meiobenthic communities to disturbances (Alongi 1985). The conclusion drawn by Chardy and Dauvin (1992) about coastal habitat types seems to be representative: after studying a coastal site in northern France and incorporating numerous other data sets, they stated that “meiofauna is an important link between the bacteria plus detritus and the carnivorous level and cannot be considered as an independent food web.” In limnic ecosystems, analyses of energy fluxes performed by Strayer and Likens (1986) demonstrated that half of the overall energy passes through the meiobenthos, but it is then transferred almost completely (80%) to predacious macrobenthic larvae of tanypodid midges, and thus ultimately exported from the system. On the other hand, the study of Bergtold and Traunspurger (2005) of a European lake emphasizes the strong connection between the members of the small food web (Fig. 9.8): a “comprehensive understanding of meiobenthic population dynamics is only possible if bacteria and protozoa are included.”
9.4 The Position of Meiofauna in the Benthic Ecosystem
Allochthonous sources
413
Phytoplankton 0.90
Particulate organic carbon 2.32
209
Zooplankton 0.10
25 Bacteria 42
Detritus
Microphytobenthos 0.50
10
0.135 83.1
2.5
7.8
Bacteria 1.02
Mixed
Deposit feeders 0.11
24.3
feeders 0.51
41.6
Meiofauna
0.83
3.6
0.28
0.46
2.16 10
3.75
Macrofauna 1.46
4.02
Fig. 9.11 The central role and position of the meiobenthos in a simplified carbon flow model from a subtidal fine sand community in the English Channel. Numbers beside arrows reflect mean annual trophic fluxes (g C m−2 y−1); numbers in circles or squares indicate mean values of biomass (g C m−2); “earthed” arrows indicate losses by respiration (losses by natural mortality and defaecation not indicated); the gray underlay indicates the benthic system. (After Chardy and Dauvin 1992)
414
9 Synecological Perspectives in Meiobenthology
Box 9.3 Meiobenthos: Mediating Between the Productive Base and the Consumptive Top? The microbenthos and its enormous metabolic and productive capacity are unequivocally the energetic basis for all benthic food webs—nobody would doubt that. At the other end of the scale, it is evident that the macrobenthos/ nekton holds a strong and important trophic position as the main consumer, and as such receives much attention. So what is the role of the intermediate meiobenthos? Numerous links connect it to the microbiota as well as to the macrobiota, but how can we prove and quantify this? Isotopic signatures help to track the fluxes of energy. Fluorescent tracers and chemical markers are additional tools that can be used to elucidate the pathways and the role of meiofauna in the food web. It is rarely direct predation that makes the meiofauna impressive. In the “small food web,” with its inexhaustible productive capacity, the meiobenthos attains a “catalytic” rather than a consumptive role. It keeps the processes going at high intensity, attracts bacteria, activates fluxes, and sometimes regulates diatom and ciliate production. In the “large food web,” the role of meiobenthos is similar. But the meiofauna stocks will rarely (perhaps for some epibenthic fishes in muddy habitats) suffice to satiate macrofauna. However, at certain life stages, when macrobenthic juveniles are in the immediate size range of meiofauna, there can be competitive interactions of regulative relevance. Where sediments are depleted of digestible organic matter, as in the deep sea or in clean sands, the resulting competitive food situation favors meiofauna over macrofauna through their effective utilization of food. Overall, do the energetic pathways lead to a disconnection of the meiofauna from the benthic mainstream; is the meiobenthos using a “parallel road net?” Once they have reached the level of the meiofauna, the energetic avenues do indeed appear to reverse, creating small recycling circuits. However, there are some important side-tracks. These are not straightforward and direct, but they have sometimes and locally a regulative function for the overall net. In the future, refined studies, especially those performed under field conditions, will demonstrate that more of these meiobenthic side-tracks lead to the main roads, thus integrating the energetic role of the meiobenthos into the system.
Admittedly, we still lack enough accurate data to be able to provide a unifying statement about the often controversial results on the role of meiofauna in the ecological complex. By constructing ecosystem models it has been possible to demonstrate that there are links between the meiobenthos and other compartments of the system, but there are also large gaps. At present, these models may locally be valid but are often based more on assumptions than on calculations. Future research will lead to numerous, often significant, corrections to the quantitative
9.4 The Position of Meiofauna in the Benthic Ecosystem
415
relations. For further calculations of energy flow we need more data on the food spectrum and additional details of the life histories of meiofauna. In the future advanced computation and computer-assisted modeling will be more feasible than it is today. This will yield more generally valid and consequently more relevant information on the meiobenthos and its ecological impact. Only then will the role of meiofauna be realistically assessed in ecosystem models of predictive value. More detailed reading: Coull and Bell (1979); Schwinghamer et al. (1986); Vranken and Heip (1986); Warwick (1987); Coull (1990, 1999); Chardy and Dauvin (1992); Heip et al. 1995; Woodward et al. (2005); freshwater, Hakenkamp and Morin (2000); Schmid-Araya and Schmid (2000).
Chapter 10
Retrospect on Meiobenthology and Outlook on New Approaches and Future Research
Summarizing the short history of meiofaunal studies, this closing chapter on meiobenthology is certainly somewhat subjective. Meiobenthology can be largely grouped into periods with dominant research “trends.” There are, of course, many individual publications that do not really fit into this gross categorization of meiofaunal research, but most of the papers seem to follow certain lines which are defined by the scientific standard of the time and by contentions about fields of major importance. ●
●
●
●
●
Nineteenth century until the end of the 1950s. Taxonomic and faunistic papers; microscope studies; establishment of the special character of many meiobenthic animals; attempts to define meiobenthos “coenoses” (stable communities) based on sediment structure. 1960 into the 1970s. Descriptive ecology of the meiobenthic habitat (abundance, distribution); analysis of abiotic factors (restricted to the technically more simple parameters) and their distributional impacts; first experimental work on meiobenthic animals. 1970s to the beginning of the 1980s. Experimental work, tolerance–preference tests; first studies on “difficult” abiotic parameters like the oxygen/sulfide complex and its impact on the distribution; biotic and trophic factors (predator–prey interactions); ultrastructure studies with respect to phylogenetic implications. 1980s into the 1990s. Manipulative ecology as an aid to understanding the role of meiofauna in benthic ecosystems; calculations of production and energy flow; meiofauna in special biotopes like the deep-sea and sulfidic environments; recolonization and dispersive mechanisms of meiofauna, impact of pollution on meiofauna; life history assays. End of 1990s to 2010(?). Three major domains appear dominant. (1) Biodiversity, its assessment and conservation: characterizing “hot spots” of meiofauna biodiversity; work on diversity-relevant factors (habitat complexity); impact of natural and anthropogenic disturbance on diversity. (2) Incorporating the potential of computerization: compilation of computerassisted databases and networks on meiofauna; development of electronic identification keys.
O. Giere, Meiobenthology, 2nd edition, doi: 10.1007/b106489, © Springer-Verlag Berlin Heidelberg 2009
417
418
10 Retrospect on Meiobenthology
(3) Molecular genetics: phylogenetic relations and evolution of taxa; distributional patterns and their genetic background; coherence of remote populations; genetic changes under extreme conditions; genetic alterations induced by pollutants (ecotoxicogenomics); tracing trophic pathways. These are just a few aspects that the powerful tool of molecular genetics is already contributing to meiobenthology. Will they become the main research areas in this century? Many other topics will require continuing studies even after their peaks of interest, but, in general, there is a sequence of problems that are addressed in meiobenthology (and in other fields of biology): What is there? → Where should we look for it? → How much is there? → Why is it there? → What are the resources and threats? → What is its ecological role? → How do we cultivate or protect it? → Which relationships are present (phylogenetic, zoogeographical, ecological, under molecular scrutiny)? Many of the unsolved problems in meiobenthology are of not only academic interest but are also of relevance in a more general or applied context. The widths of the research fields provide important work for every interested meiobenthologist. Below, some topics that (subjectively) appear to be important and feasible are listed (not in any order of relevance): Diversity, distribution, zoogeography: – Investigate the relation between ecosystem functioning and biodiversity – Explain small-scale distributional patterns of meiofauna as a reaction to complex abiotic factors (e.g., local hydrodynamic fluxes, oxygen supply) or to structural complexity – Understand the dispersive processes in meiofauna that would help us to explain the notorious problem of meiofaunal zoogeographical distribution (aggregates as rafts?) – Study the relation between refuge biotopes (deep sea, caves, continental groundwater) and evolution on a genetic basis Ecology, life history: – Reveal details of reproductive biology, especially in key taxa with aberrant modes, e.g., ostracods or nematodes – Produce reliable, preferably field-generated, data on life history and population dynamics of relevant species in order to calculate meiobenthic production and assess energy flow in the system – Refine our knowledge in laboratory work and cross-check the results with corresponding data from the field situation – Tackle the problem of life under low-oxic or anoxic conditions (thiobios) and assess its ecological role – Ascertain the role of meiofauna in the small food web, especially their partitioning in decomposing organic debris – Establish the role of biofilms and exopolymer excretions for meiofauna
10 Retrospect on Meiobenthology
419
– Scrutinize the role of heterotrophic flagellates between bacteria and meiofauna – Refine (e.g., by multifactorial experimental designs or in microcosms) the impacts of complex biotic features like competition, trophic relations, facilitation and amensalism (manipulative ecology) – Study the structural/behavioral adaptations of meiofauna organisms to gain a better understanding of their micro-ecology and small-scale reactions Applied sciences, pollution: – Study the impact of environmental constraints and pollution on biodiversity – Determine the potential of large meiofauna culture plants as a food source for aquaculture – Use molecular signatures of meiofauna for ecological and pollution problems – Learn more about the subterranean meiofauna and the related problems of groundwater pollution by pesticides and pharmaceutical products and understand the disruptions caused by endocrine substances – Assess the consequences of pollution, mainly in relation to life history changes and recolonization processes Physiology: – Increase our understanding of the (eco)physiology of meiobenthos, which is still in its infancy (using micromethods, mass extraction of meiofauna); metabolic processes under permanent anoxia are of particular relevance here Evolution, phylogeny, morphology: – Reveal the phylogenetic links that bridge the gaps between large taxa by applying molecular techniques – Use morphological and ultrastructural details in order to compare structural with genetic changes For these goals, we need innovative and often experimental studies using sophisticated methods. Some promising approaches and techniques are suggested below, and most of them have already been applied to meiofauna. Molecular genetics: ●
●
DNA extraction from small tissue samples, even just one harpacticoid (Schizas et al. 1997) DNA barcoding of single markers for rapid genetic identification of species-rich communities and for genetic monitoring in ecological surveys; development of successful PCR methods for the notoriously “difficult” nematodes; exploitation of genetic information from archived material (Bhadury et al. 2006a,b,c; Meldal et al. 2007); genome and transscriptome analyses (nematodes: Parkinson et al. 2004; oligochaetes: Woyke et al. 2006)
420 ●
●
●
10 Retrospect on Meiobenthology
Using molecular differences for phylogenetic investigations (gastrotrichs: Manylov and Vladychenskaya 2004; foraminiferans: Pawlowski et al. 1994; harpacticoids: Rocha-Olivares et al. 2001; nematodes: Meldal et al. 2007) Diversity of taxa based on r RNA (28S); assessment of differences in the context of ecological analyses (Markmann and Tautz 2005) Changes in single-locus gene frequencies for biogeographic and ecotoxicological purposes (Staton et al. 2001; Watanabe and Iguchi 2006)
Microscopic studies: ●
●
●
●
●
●
●
●
●
Fluorometric enzyme immunoassays for monitoring molting and growth cycles of crustaceans (Block et al. 2003) Microscopic interference patterns for the study of osmoregulation and osmotic stress (Forster 1998) Application of fluorescent markers for investigating trophic pathways with fluorescence microscopy (Lindegarth et al. 1991) Three-dimensional time-lapse microscopy for the continuous monitoring of cell development and embryology (Hejnol and Schnabel 2006) Confocal laser scanning microscopy and immunohistochemistry for the 3Dvisualization of complex internal organs such as the nervous system (Müller 2006) Using fluorochromes in two-photon laser scanning microscopy for in situ characterization and visualization of exopolymers and biofilms (Neu et al. 2002) Specific fluorescent staining of carbohydrates and proteins (Schumann and Rentsch 1998) Confocal laser scanning microscopy for assessing the effects of toxicants in vivo (Chandler and Volz 2004; Volz and Chandler 2004) Developmental biology and reproductive modes of turbellarians (Peter et al. 2001)
Physiological approaches: ●
●
●
●
●
Mass extraction of meiofauna for physiological assays (Couch 1988; RzeznikOrignac et al. 2004) New extraction technique in the microgram range for the gas-chromatographic analysis of pesticides (Wirth et al. 1994) Measurement of respiration with polarographic microelectrodes using just a few specimens (Moens et al. 1996a; Moodley et al. 2008) Establishing specific fatty acid signatures in order to estimate the prey proportions in food webs (Watanabe et al. 1983; Iverson et al. 2004) Physiology and biochemistry of nematodes (Perry and Wright 1998)
Isotopic marking for ecological studies: ●
Radioisotopic labeling for analyzing food transfer (Widbom and Frithsen 1995; Riera et al. 1996; Ólafsson et al. 1999; De Troch et al. 2005a)
Computer-based methods: ●
Computer-assisted identification of large numbers of meiofauna (nematodes) (Diederich et al. 2000)
10 Retrospect on Meiobenthology ● ●
421
Computerized keys for harpacticoids (see Wells 2007) Computer-produced illustrations (Bouck and Thistle 1998)
Other relevant procedures: ●
●
Mathematical principles underlying the distribution patterns of meiofauna (Pfeifer et al. 1996) X-ray imaging of sediment cores to diagnose bioturbation (Diaz et al. 1994)
The statement that “the easy part of meiofaunal research is done” (Coull and Palmer 1984) should remain our dictum, stimulating us for future tasks. However, it does not seem to be simply a matter of “easy” and “difficult.” We need to structure our research better. Selecting exotic sites and studying the local abundance of meiofaunal taxa will not do in the future. To ensure that we do not get lost in details and miss the mainstream, we must focus on topics of general concern, utilizing potent and sophisticated methods. The wider public will turn their attention to the meiobenthos when we understand that we must present meiobenthology not just as a fascinating scientific field, but also as an extremely useful one for solving important problems. Climatic change, the issue that dominates our future, must be specifically addressed in meiobenthos research (see for instance Richmond et al. 2007). “The values of meiofauna research for assessing ecosystem health need to be advocated” (Coull, pers. commun.). The advantages of meiofauna as opposed to macrofauna for monitoring and interpreting pollution problems should be emphasized more clearly; disadvantages such as complicated identification can be overcome by creating computer-based illustrated keys that are accessible to each meiobenthologist and perhaps designed for specific tasks and regions. At least for ecology-oriented studies, “keystone taxa” must be ascertained and research concentrated on them. These are not necessarily the most abundant taxa; rather, they are the taxa that structure the processes in ecosystems—the depletion of these “engineering taxa” would have severe, cascading effects. The close ties between the meiobenthos of marine and freshwater biotopes are of eminent importance for both fields (Palmer 1990b; Rundle et al. 2002). When comparing data we should use the same terminology and methodology and look beyond the borderlines. The psychological problem inherent in meiobenthology — examining animals of microscopical size unfamiliar even to many biologists — could be overcome by emphasizing the general aspects of our findings rather than looking for idiosyncrasies. These steps could lead meiobenthology out of the somewhat isolated “corner” where specialists discuss their problems away from the biological and ecological mainstream. They steps would bring meiobenthology closer to the general concerns of the public and make it an issue to know and care about. More detailed reading: Coull and Palmer 1984, Coull and Giere 1988.
422
10 Retrospect on Meiobenthology
Box 10.1 The Meiobenthos Dilemma: Promoting the Invisible During several decades of work on meiobenthos, I have had occasion to consider and discuss with colleagues a major problem that is apparently inherent to meiobenthology: the human mind seems to be adapted to consider only conspicuous items as impressive and relevant. Meiofauna are not impressively large or tasty, and they are not even dangerous—they are simply small. Meiofauna, organisms beyond our normal range of perception, are therefore intuitively uninteresting to most people, even to some in the scientific community, despite the productive capacity, ecological adaptability and environmental sensitivity of these tiny creatures. Indeed, meiobenthology has been completely ignored at recent marine congresses (European Marine Biology Symposium 2007) and in recent books (Pitcher et al. 2007 on sea mount ecology, Thomas et al. (2003) on sea ice). We few dedicated specialists must struggle to draw public attention, financial support and young talent to the study of meiofauna. Only when we advocate the specific values of meiobenthology will we keep in contact with the mainstream. I am not convinced the situation will change soon. Even in good times, meiobenthologists are a small group, and in bad times meiobenthic research is always at risk of becoming extinct. Since meiobenthology often depends on a few scientists and their “schools,” retirements and fluctuations in research activities and support can limit productive viability. In many respects the scenario resembles that of rare, isolated species— meiobenthologists are endangered! How can we strengthen and safeguard the situation? We should not allow our research to be marginalized. We should focus on problems of general relevance (e.g., pollution, climate change, ecosystem health), and demonstrate that many scientific issues can be optimally solved by working meiofauna: this is the way to attract general recognition. ● We should closely tie our research to general benthology, interact with general benthologists at common meetings, and discuss problems with them; the North American Benthological Society is a good example. ● We should publish results of general relevance in general, internationally renowned biological journals. ● We should concentrate our efforts and create big, cooperating groups with international partners, thus enhancing the “multiplication factor” considerably. This attracts the awareness of the public. ● We should seek to financially support cooperation with those nations in which meiobenthology is in its infancy or is not even represented. In most cases these countries are as meiobenthologically pristine as their scholars are enthusiastic; important discoveries can result. ● ●
There is only one thing we should not do: lose our easy-going way of exchanging news and ideas at meetings and the enthusiasm for our work on these tiny but fascinating creatures.
References
Aarnio K, Bonsdorff E (1993) Seasonal variation in abundance and diet of the sand goby Pomatoschistus minutus (Pallas) in a northern Baltic Archipelago. Ophelia 37: 19–30 Aarnio K, Bonsdorff E (1997) Passing the gut of juvenile flounder, Platichthys flesus (L.): differential survival of zoobenthic prey species. Mar Biol 129: 11–14 Aarnio K, Bonsdorff E, Norkko A (1998) Role of Halicryptus spinulosus (Priapulida) in structuring meiofauna and settling macrofauna. Mar Ecol Prog Ser 163: 145–153 Abu Rezg TS, Yule AB, Teng SK (1997) Ingestion, fecundity, growth rates and culture of the harpacticoid copepod, Tisbe furcata, in the laboratory. Hydrobiologia 347: 109–118 Adrianov AV, Malakhov VV (1994) Kinorhyncha: structure, development, phylogeny and taxonomy. Nauka Publishing, Moscow Agatha S, Spindler M, Wilbert N (1993) Ciliated Protozoa (Ciliophora) from Arctic sea ice. Acta Protozoologica 32: 261–268 Åkesson B (1973) Reproduction and larval morphology of five Ophryotrocha species (Polychaeta, Dorvilleidae). Zool Scr 2: 145–155 Åkesson B (1975) Bioassay studies with polychaetes of the genus Ophryotrocha as test animals. In: Koeman JH, Strik JJ (eds) Sublethal effects of toxic chemicals on aquatic animals. Proceedings of the Swedisch-Netherlands Symposium, Wageningen, The Netherlands, pp. 121–135 Åkesson B (1977) Parasite–host relationships and phylogenetic systematics. The taxonomic position of dinophilids. Mikrofauna Meeresboden 61: 19–28 Alekseev VR, De Stasio B, Gilbert JJ (eds) (2007) Diapause in aquatic invertebrates: theory and human use. Springer, Berlin Alkemade R, Wielemaker A, Hemminga MA (1992) Stimulation of decomposition of Spartina anglica leaves by the bacterivorous marine nematode Diplolaimelloides bruciei (Monhysteridae). J Exp Mar Biol Ecol 159: 267–278 Allan JD, Castillo MM (2007) Stream ecology. Structure and function of running waters. 2nd ed., Springer, Dordrecht, Netherlands, p. 436 Aller RC (1988) Benthic fauna and biogeochemical processes in marine sediments: the role of burrow structures. In: Blackburn TH, Sørensen J (eds) Nitrogen cycling in coastal marine environments. Wiley, Chichester, pp. 301–338 Aller RC, Aller JY (1992) Meiofauna and solute transport in marine muds. Limnol Oceanogr 37: 1018–1033. Aller RC, Yingst JY (1978) Biogeochemistry of tube-dwelling: A study of the sedentary polychaete Amphitrite ornata (Leidy). J Mar Res 36: 201–254 Aller RC, Yingst JY (1980) Relationships between microbial distributions and the anaerobic decomposition of organic matter in surface sediments of Long Island Sound, USA. Mar Biol 56: 29–40 Aller RC, Yingst JY (1985) Effects of the marine deposit-feeders Heteromastus filiformis (Polychaeta), Macoma balthica (Bivalvia), and Tellina texana (Bivalvia) on averaged sedimentary solute transport, reaction rates, and microbial distributions. J Mar Res 43: 615–645 Alongi DM (1985) Effects of physical disturbance on population dynamics and trophic interactions among microbes and meiofauna. J Mar Res 43: 351–364 423
424
References
Alongi DM (1986) Quantitative estimates of benthic protozoa in tropical marine systems using silica gel: a comparison of methods. Estuar Coast Shelf Sci 23: 443–450 Alongi DM (1987a) Intertidal zonation and seasonality of meiobenthos in tropical mangrove estuaries. Mar Biol 95: 447–458 Alongi DM (1987b) The influence of mangrove derived tannins on intertidal meiobenthos in tropical estuaries. Oecologia (Berl) 71: 537–540 Alongi DM (1990a) The ecology of tropical soft-bottom benthic ecosystems. Oceanogr Mar Biol Ann Rev 28: 381–496 Alongi DM (1990b) Community dynamics of free-living nematodes in some tropical mangrove and sandflat habitats. Bull Mar Sci 46: 358–373 Alongi DM (2008) The dynamics of tropical mangrove forests. Springer, Berlin Alongi DM, Tietjen JH (1980) Population growth and trophic interactions among free-living marine nematodes. In: Tenore KR, Coull BC (eds) Marine benthic dynamics. University of South Carolina Press, Columbia, pp. 151–166 Alongi DM, Boesch DF, Diaz RJ (1983) Colonization of meiobenthos in oil-contaminated subtidal sands in the lower Chesapeake Bay. Mar Biol 72: 325–335 Alongi DM, Tenore KR (1985) Effect of detritus supply on trophic relationships within experimental benthic food webs. I. Meiofauna-polychaete (Capitella capitata (Type I) Fabricius) interactions. J Exp Mar Biol Ecol 88: 153–166 Alongi DM, Sasekumar A (1992) Benthic communities. In: Robertson AL, Alongi DM (eds) Tropical mangrove ecosystems. American Geophysical Union, Washington, DC, pp. 137–171 Alsterberg C, Sundbäck K, Larson F (2007) Direct and indirect effects of an antifouling biocide on benthic microalgae and meiofauna. J Exp Mar Biol Ecol 351: 56–72 Andersen TJ, Pejrup M (2002) Biological mediation of the settling velocity of bed material eroded from an intertidal mudflat, the Danish wadden sea. Estuar Coast Shelf Sci 54: 737–745 Anderson JG, Meadows PS (1969) Bacteria on intertidal sand grains. Hydrobiologia 33: 33–46 Andrew NL, Mapstone BD (1987) Sampling and the description of spatail pattern in marine ecology. Oceanogr Mar Biol Ann Rev 25: 39–90 Angelier E (1953) Recherches écologiques et biogéographiques sur la faune des sables submergés. Arch Zool Exp Gén 90: 37–161 Ankar S (1977) Digging profile and penetration of the Van Veen grab in different sediment types. Contr Askö Lab 16: 1–22 Ankar S, Elmgren R (1976) The benthic macro- and meiofauna of the Askö – Landsort area (Northern Baltic proper). A stratified random sampling survey. Contrib Askö Lab Univ Stockholm 11: 115 Ansari ZA, Ingole B (2002) Effects of an oil spill from MV Sea Transporter on intertidal meiofauna at Goa, India. Mar Pollut Bull 44: 396–402 Ansari ZA, Parulekar AH (1993) Environmental stability and seasonality of a harpacticoid copepod community. Mar Biol 115: 279–286 Ansari ZA, Sreepada RA, Matondkar SGP, Parulekar AH (1993) Meiofaunal stratification in relation to microbial food in a tropical mangrove mud flat. Trop Ecol 34: 204–216 Aramayo V, Enríquez E, Gutiérrez D, Quipúzcoa L, Marquina R (2007) Meiofaunal communities off the central Peruvian coast under the weak 2002/03 El Niño period. In: Santos PJP dos (ed) 13th International Meiofauna Conference. The Organizing Committee, Recife, Brazil, P 23 Archer D, Devol A (1992) Benthic oxygen fluxes on the Washington shelf and slope: a comparison of in situ microelectrode and chamber flux measurements. Limnol Oceanogr 37: 614–62. Archer D, Emerson S, Reimers C (1989) Dissolution of calcite in deep-sea sediments: pH and O2 microelectrode results. Geochim Cosmochim Acta 53: 2831–2845 Arlt G (1988) Temporal and spatial meiofauna fluctuations in an inlet of the South West Baltic (Darss-Zingst Bodden chain) with special reference to the Harpacticoida (Copepoda, Crustacea). Int Rev Ges Hydrobiol 73: 297–308 Arlt G (1993) Composition and seasonal variations in the tropical shallow subtidal meiofauna of a coral reef lagoon near Massawa (Red Sea, Eritrea). In: Eleftheriou A, Ansell AD, Smith CJ
References
425
(eds) Biology and ecology of shallow coastal waters. Olsen & Olsen, Fredensborg, Denmark, pp. 101–106 Arlt G, Müller B, Warnack KH (1982) On the distribution of meiofauna in the Baltic sea. Int Rev Ges Hydrobiol 67: 97–111 Armonies W (1986) Plathelminth abundance in North Sea salt marshes: environmental instability causes high diversity. Helgol Meeresunters 40: 229–240 Armonies W (1988a) Active emergence of meiofauna from intertidal sediment. Mar Ecol Prog Ser 43: 151–159 Armonies W (1988b) Hydrodynamic factors affecting behaviour of intertidal meiobenthos. Ophelia 28: 183–194 Armonies W (1988c) Plathelminth fauna of North Sea salt marshes is strongly affected by salinity. Progr Zool 36: 505–509 Armonies W (1988d) Common pattern of plathelminth distribution in North Sea salt marshes and in the Baltic Sea. Arch Hydrobiol 111: 625–636 Armonies W (1989a) Semiplanktonic Plathelminthes in the Wadden Sea. Mar Biol 101: 521–528 Armonies W (1989b) Meiofaunal emergence from intertidal sediment measured in the field: significant contribution to nocturnal planktonic biomass in shallow waters. Helgol Meeresunters 43: 29–43 Armonies W (1989c) Occurrence of meiofauna in Phaeocystis seafoam. Mar Ecol Prog Ser 53: 305–309 Armonies W (1990) Short-term changes of meiofaunal abundance in intertidal sediments. Helgol Meeresunters 44: 375–386 Armonies W (1994) Drifting meio- and macrobenthic invertebrates on tidal flats in Königshafen (German Wadden Sea). Helgol Meeresunters 48: 299–320 Arnaud PM, Poizat C, Salvini-Plawen LV (1986) Interstitial gastropods. In: Botosaneanu L (ed) Stygofauna Mundi. Brill/Backhuys, Leiden, The Netherlands, pp. 153–176 Arroyo NL, Maldonado M, Pérez-Portela R, Benito J (2004) Distribution patterns of meiofauna associated with a sublittoral Laminaria bed in the Cantabrian Sea (north-eastern Atlantic). Mar Biol 144: 231–242 Arroyo NL, Aarnio K, Bonsdorff E (2006a) Drifting algae as a means of re-colonizing defaunated sediments in the Baltic Sea. A short-term microcosm study. Hydrobiologia 554: 83–95 Arroyo NL, Maldonado M, Walters K (2006b) Within- and between-plant distribution of harpacticoid copepods in a North Atlantic bed of Laminaria ochroleuca. J Mar Biol Ass UK 86: 309–316 Asmus RM, Bauerfeind E (1994) The microphytobenthos of Königshafen—spatial and seasonal distribution on a sandy tidal flat. Helgol Meeresunters 48: 257–276 Astm (2004) Standard guide for conducting renewal microplate-based life-cycle toxicity tests with a marine meiobenthic copepod. American Society for Testing and Materials, West Conshohocken, PA, USA Athersuch J, Horne DJ, Whittaker JE (1989) Marine and brackish water ostracods (Superfamilies Cypridacea and Cytheracea). Brill, Leiden, p. 343 (Synopses of the British Fauna, vol 43) Atilla N, Fleeger JW, Finelli CM (2005) Effects of habitat complexity and hydrodynamics on the abundance and diversity of small invertebrates colonizing artificial substrates. J Mar Res 63: 1151–1172 Austen MC, McEvoy AJ (1997a) The use of offshore meiobenthic communities in laboratory microcosm experiments: response to heavy metal contamination. J Exp Mar Biol Ecol 211: 247–261 Austen MC, McEvoy AJ (1997b) Experimental effects of Tributylin (TBT) contaminated sediments on a range of meiobenthic communities. Environ Pollut 96: 435–444 Austen MC, Somerfield PJ (1997) A community level sediment bioassay applied to an estuarine heavy metal gradient. Mar Environ Res 43: 315–328 Austen MC, Warwick RM (1995) Effects of manipulation of food supply on estuarine meiobenthos. Hydrobiologia 311: 175–184
426
References
Austen MC, Widbom B (1991) Changes in and slow recovery of a meiobenthic nematode assemblage following a hypoxic period in the Gullmar Fjord basin, Sweden. Mar Biol 111: 139–145 Austen MC, Widdicombe S (2006) Comparison of the response of meio- and macrobenthos to disturbance and organic enrichment. J Exp Mar Biol Ecol 330: 96–104 Austen MC, Warwick RM, Rosado MC (1989) Meiobenthic and macrobenthic community structure along a putative pollution gradient in southern Portugal. Mar Poll Bull 20: 398–404 Austen MC, McEvoy AJ, Warwick RM (1994) The specificity of meiobenthic community responses to different pollutants: results from microcosm experiments. Mar Pollut Bull 28: 557–563 Ax P (1963) Die Ausbildung eines Schwanzfadens in der interstitiellen Sandfauna und die Verwertbarkeit von Lebensformcharakteren für die Verwandtschaftsforschung. (Mit Beschreibungen zweier neuer Turbellarien aus den Ordnungen Acoela und Seriata). Zool Anz 171: 51-76 Ax P (1966) Die Bedeutung der interstitiellen Sandfauna für allgemeine Probleme der Systematik, Ökologie und Biologie. Veröff Inst Meeresforsch Bremerhaven Sbd. 2: 15–65 Ax P (1969) Populationsdynamik, Lebenszyklen und Fortpflanzungsbiologie der Mikrofauna des Meeressandes. Verh Deutsch Zool Ges Innsbruck 1968, Zool Anz Suppl. 32: 66–113 Ax P (1985) The position of the Gnathostomulida and Platyhelminthes in the phylogenetic system of the Bilateria. In: Conway MS, George JD, Gibson R, Platt HM (eds) The origins and relationshsips of lower invertebrates. Clarendon, Oxford, pp. 168–180 Ax P (2008) Plathelminthes aus Brackgewässern der Nordhalbkugel. Steiner, Stuttgart Ax P, Armonies W (1987) Amphiatlantic identities in the composition of the boreal brackish water community of Plathelminthes. A comparison between the Canadian and European Atlantic coast. Microfauna Mar 3: 7–80 Ax P, Armonies W (1990) Brackish water Plathelminthes from Alaska as evidence for the existence of a boreal brackish water community with circumpolar distribution. Microfauna Mar 6: 7–110 Azam F (1998) Microbial control of oceanic carbon flux: the plot thickens. Science 280: 694–696 Azovsky AI (1988) Colonization of sand islands by psammophilous ciliates: the effect of microhabitat size and stage of succesion. Oikos 51: 48–56 Azovsky AI, Saburova MA, Chertoprood ES, Polikarpov IG (2005) Selective feeding of littoral harpacticoids on diatom algae: hungry gourmands? Mar Biol 148: 327–337 Babbitt CC, Patel NH (2005) Relationships within the Pancrustacea: examining the influence of additional Malacostracan 18S and 28S rDNA. Crustacean Issues 16: 275–294 Bagarinao T (1992) Sulfide as an environmental factor and toxicant: tolerance and adaptations in aquatic organisms. Aquat Toxicol 24: 21–62 Baguley JG, Hyde LJ, Montagna PA (2004) A semi-automated digital microphotographic approach to measure meiofaunal biomass. Limnol Oceanogr Methods 2: 181–190 Bak RPM, Nieuwland G (1989) Seasonal fluctuations in benthic protozoan populations at different depths in marine sediments. Neth J Sea Res 24: 37–44 Bale AJ, Kenny AJ (2005) Sediment analysis and seabed characterisation. In: Eleftheriou A, McIntyre A (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 43–86 Balian EV, Lévèque C, Segers H, Martens K (eds) (2008) Freshwater animal diversity assessment. Hydrobiologia 595: 1–637 Balsamo M, Guidi L, Pierboni L, Marotta R, Todaro MA, Ferraguti M (2007) Living without mitochondria: spermatozoa and spermatogenesis in two species of Urodasys (Gastrotricha, Macrodasyida) from dysoxic sediments. Invert Biol 126: 1–9 Banse K (1982) Mass-scaled rates of respiration and intrinsic growth in very small invertebrates. Mar Ecol Prog Ser 9: 281–297 Banse K, Mosher S (1980) Adult body mass and annual production/biomass relationships of field populations. Ecol Monogr 50: 355–379 Barbuto M, Zullini A (2005) The nematode community of two Italian rivers (Taro and Ticino). Nematology 7: 667–675
References
427
Barnard JL (1969) The families and genera of marine gammaridean Amphipoda. Bull US Natn Hist Mus 271: 1–535 Barnard JL, Barnard CM (1983) Freshwater Amphipoda of the world. I. Evolutionary patterns. II. Handbook and bibliography. Hayfield Associates, Mt. Vernon, VA Barnes DKA (2002) Invasions by marine life on plastic debris. Nature 416: 808–809 Barnes H, Stanbury FA (1948) The toxic action of copper and mercury salts both separately and when mixed on the harpacticoid copepod, Nitocra spinipes (Boeck). J Exp Biol 25: 270–275 Barnes PO (1973) An in situ interstitial water sampler for use in unconsolidated sediments. DeepSea Res 20: 1125–1128 Barnett BE (1980) A physico-chemical method for the extraction of marine and estuarine benthos from clays and resistant muds. J Mar Biol Ass UK 60: 255 Barnett PRO, Watson J, Connelly D (1984) Multiple corer for taking virtually undisturbed samples from shelf, bathyal and abyssal sediments. Oceanologica Acta 7: 399–408. Barry JP, Buck KR, Lovera C, Kuhnz L, Whaling PJ (2005) Utility of deep sea CO2 release experiments in understanding the biology of a high-CO2 ocean: Effects of hypercapnia on deep sea meiofauna. J Geophys Res 110: 09–12 Bartolomaeus T, Ax P (1992) Protonephridia and Metanephridia—their relation within the Bilateria. Z zool Syst Evolutforsch 30: 21–45 Bartolomaeus T, Purschke G (eds) (2005) Morphology, molecules, evolution and phylogeny in Polychaeta and related taxa. Springer, Berlin Bartsch I (1974) Ein Beitrag zur Systematik, Biologie und Ökologie der Halacaridae (Acari) aus dem Litoral der Nord- und Ostsee. II. Ökologische Analyse der Halacaridenfauna. Abh Verh naturwiss Ver Hamb 17: 9–53 Bartsch I (1979) Halacaridae (Acari) von der Atlantikküste Nordamerikas. Beschreibung der Arten. Mikrofauna Meeresboden 79: 1–62 Bartsch I (1989) Marine mites (Halacaridea: Acari): a geographical and ecological survey. Hydrobiologia 178: 21–42 Bartsch I (1996) Halacarids (Halacaroidea, Acari) in freshwater. Multiple invasions from the Paleozoic onwards? J Nat Hist 30: 67–99 Bartsch I (2004) Geographical and ecological distribution of marine genera and species (Acari: Halacaridae). Exper Appl Acarol 34: 37–58 Bartsch I (2006) Halacaroidea (Acari): a guide to marine genera. Organ Divers Evolut 6: (2) 103–108 Bates JW (1997) The slide-sealing compound “Glyceel”. J Nematol 29: 565–566 Battaglia B (1957) Ricerche sul ciclo biologico di Tisbe gracilis (T. Scott), (Copepoda, Harpacticoida), studiato in condizioni di laboratorio. Archo Oceanogr Limnol 11: 29–46 Battaglia B, Beardmore JA (1978) Marine organisms. Genetics, ecology and evolution. NATO. Advanced Research Institute, Venice 1977. Plenum, New York Battaglia B, Bisol PM, Varotto V (1978) Variabilité génétique dans des populations marine et lagunaires de Tisbe holothuriae (Copepoda, Harpacticoida). Arch Zool Exp Gén 119: 251–264 Beeton AM (1984) The world’s great lakes. J Great Lakes Res 10: 106–113 Bejanaro AC, Chandler GT (2003) Reproductive and developmental effects of atrazine on the estuarine meiobenthid copepod Amphiascus tenuiremis. Environ Toxicol Chem 22: 3009–3016 Bejanaro AC, Maruya KA, Chandler TG (2004) Toxicity assessment of sediments associated with various land-uses in coastal South Carolina, USA, using a meiobenthic copepod bioassay. Mar Pollut Bull 49: 23–32 Bejarano AC, Chandler GT, Decho AW (2005) Influence of natural dissolved organic matter (DOM) on acute and chronic toxicity of the pesticides chlorothalonil, chlorpyrifos and fipronil on the meiobenthic estuarine copepod Amphiascus tenuiremis. J Exp Mar Biol Ecol 321: 43–57 Bejarano AC, Chandler GT, He LJ, Cary TL, Ferry JL (2006a) Risk assessment of the National Institute of Standards and Technology petroleum crude oil standard water accommodated
428
References
fraction: further application of a copepod-based, full life-cycle bioassay. Environ Toxicol Chem 25: 1953–1960 Bejarano AC, Chandler GT, He L, Coull BC (2006b) Individual to population level effects of South Louisiana crude oil water accommodated hydrocarbon fraction (WAF) on a marine meiobenthic copepod. J Exp Mar Biol Ecol 332: 49–59 Bell SS (1979) Short and long-term variation in a high marsh meiofauna community. Estuar Coast Mar Sci 9: 331–350 Bell SS (1980) Meiofauna-macrofauna interactions in a high salt marsh habitat. Ecol Monogr 50: 487–505 Bell SS, Coen LD (1982) Investigations on epibenthic meiofauna I. Abundances on and repopulation of the tube-caps of Diopatra cuprea (Polychaeta: Onuphidae) in a subtropical system. Mar Biol 67: 303–309 Bell SS, Coull BC (1980) Experimental evidence for a model of juvenile macrofauna-meiofauna interactions. In: Tenore KR, Coull BC (eds) Marine benthic dynamics. University of South Carolina Press, Columbia, SC, pp. 179–192 Bell SS, Hicks GRF (1991) Marine landscapes and faunal recruitment: a field test with seagrasses and copepods. Mar Ecol Prog Ser 73: 61–68 Bell SS, Sherman KM (1980) A field investigation of meiofaunal dispersal: tidal resuspension and implications. Mar Ecol Prog Ser 3: 245–249 Bell SS, Hicks GRF, Walters K (1989) Experimental investigations of benthic reentry by migrating meiobenthic copepods. J Exp Mar Biol Ecol 130: 291–303 Berelson WM, Townsend T, Heggie D, Ford P, Longmore A, Skyring G, Kilgore T, Nicholson G (1999) Modelling bio-irrigation rates in the sediments of Port Phillip Bay. Mar Freshwat Res 50: 573–579 Berg CJ, Adams NL (1984) Microwave fixation of marine invertebrates. J Exp Mar Biol Ecol 74: 195–199 Berg P, Rysgaard S, Funch P, Sejr MK (2001) Effects of bioturbation on solutes and solids in marine sediments. Aquat Microb Ecol 26: 81–94 Berge JA, Leinaas HP, Sandoy K (1985) The solitary bryozoan, Monobryozoon limicola Franzén (Ctenostomata), a comparison of mesocosm and field samples from Oslofjorden, Norway. Sarsia 70: 91–94 Bergmans M, Dahms H-U, Schminke HK (1991) An r–strategist in Antarctic pack ice. Oecologia (Berl) 86: 305–309 Bergtold M, Günter V, Traunspurger W (2005) Is there competition among ciliates and nematodes? Freshw Biol 50: 1351–1359 Bergtold M, Traunspurger W (2004) The benthic community in the profundal of Lake Brunnsee: seasonal and spatial patterns. Arch Hydrobiol 160: 527–554 Bergtold M, Traunspurger W (2005) Benthic production by micro-, meio-, and macrobenthos in the profundal zone of an oligotrophic lake. J N Am Benthol Soc 24: 321–329 Bergtold M, Traunspurger W (2006) Production of freshwater nematodes. In: Eyualem-Abebe E, Andrássy I, Traunspurger W (eds) Freshwater nematodes. Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, pp. 94–104 Bernhard JM (1996) Microaerophilic and facultative anaerobic benthic foraminifera: a review of experimental and ultrastructural evidence. Rev Paléobiol 15: 261–275 Bernhard JM (2000) Distinguishing live from dead foraminifera: methods review and proper application. Micropalaeontology 46: 38–46 Bernhard JM, Bowser SS (1992) Bacterial biofilms as a trophic resource for certain benthic Foraminifera. Mar Ecol Prog Ser 83: 263–272 Bernhard JM, Bowser SS (1996) Novel epifluorescence microscopy method to determine life position of Foraminifera in sediments. J Micropalaeontol 15: 68 Bernhard JM, Sen Gupta BK (2002) Foraminifera of oxygen-depleted environments. In: Sen Gupta BK (ed) Modern Foraminifera. Kluwer , Dordrecht, pp. 201–216 Bernhard JM, Buck KR, Farmer MA, Bowser SS (2000) The Santa Barbara Basin is a symbiotic oasis. Nature 403: 77–80
References
429
Berninger U-G, Epstein SS (1995) Vertical distribution of benthic ciliates in response to the oxygen concentration in an intertidal North Sea sediment. Aquat Microb Ecol 9: 229–236 Bertolani R (2001) Evolution of the reproductive mechanisms in tardigrades—a review. Zool Anz 240: 247–252 Bett BJ, Vanreusel A, Vincx M, Soltwedel T, Pfannkuche O, Lambshead PJD, Gooday AJ, Ferraro T, Dinet A (1994) Sampler bias in the quantitative study of deep-sea meiobenthos. Mar Ecol Prog Ser 104: 197–203 Beyers SC, Mills EL, Stewart PL (1978) A comparison of methods of determining organic carbon in marine sediments with suggestions for a standard method. Hydrobiologia 58: 43–47 Bezerra TNC, De Mesel I, Bouillon S, Vanreusel A, Moens T (2007) Diversity and structure of nematode communities across mangrove and seagrass vegetations at Gazi Bay, Kenya. In: Santos PJP dos (ed) Abstr 13th Int Meiof Conf, Recife Brazil, O 7 Bhadury P, Austen MC, Bilton DT, Lambshead PJD, Rogers AD, Smerdon GR (2005) Combined morphological and molecular analysis of individual nematodes through short-term preservation in formalin. Mol Ecol Notes 5: 965–968 Bhadury P, Austen MC, Bilton DT, Lambshead PJD, Rogers AD, Smerdon GR (2006a) Development and evaluation of a DNA-barcoding approach for the rapid identification of nematodes. Mar Ecol Prog Ser 320: 1–9 Bhadury P, Austen MC, Bilton DT, Lambshead PJD, Rogers AD, Smerdon GR (2006b) Molecular detection of marine nematodes from environmental samples—overcoming eukaryotic interference. Aquat Microb Ecol 44: 97–103 Bhadury P, Austen MC, Bilton DT, Lambshead PJD, Rogers AD, Smerdon GR (2006c) Exploitation of archived marine nematodes—a hot lysis extraction protocol for molecular studies. Zool Scr 36: 93–98 Bhadury P, Austen MC, Bilton DT, Lambshead PJD, Rogers AD, Smerdon GR (2008) Evaluation of combined morphological and molecular techniques for marine nematode (Terschellingia spp.) identification. Mar Biol 154: 509–518 Bick A, Arlt G (2005) Intertidal and subtidal soft-bottom macro- and meiofauna of the Kongsfjord (Spitsbergen). Polar Biol 28: 550–557 Billerbek M, Werner U, Bosselmann K, Walpersdorf E, Huettel M (2006) Nutrient release from an exposed intertidal sand flat. Mar Ecol Prog Ser 316: 35–51 Bird AF, Bird J (1991) The structure of nematodes. Academic, San Diego, CA Black KS (1997) Microbiological factors contributing to erosion resistance in natural cohesive sediments. In: Burt N, Parker R, Watts J (eds) Cohesive sediments. Wiley, Chichester, pp. 231–244 Blackburn N, Fenchel T (1999) Influence of bacteria, diffusion and shear on micro-scale nutrient patches, and implications for bacterial chemotaxis. Mar Ecol Prog Ser 189: 1–7 Blanchard GF (1990) Overlapping microscale dispersion patterns of meiofauna and microphytobenthos. Mar Ecol Prog Ser 68: 101–111 Blanchard GF (1991) Measurement of meiofauna grazing rates on microphytobenthos—is primary production a limiting factor? J Exp Mar Biol Ecol 147: 37–46 Block DS, Bejarano AC, Chandler TG (2003) Ecdysteroid concentrations through various lifestages of the meiobenthic harpacticoid copepod, Amphiascus tenuiremis and the benthic estuarine amphipod, Leptocheirus plumulosus. Gen Comp Endocrin 132: 151–160 Blome D, Schleier U, Bernem K-HV (1999) Analysis of the small-scale spatial patterns of free-living marine nematodes from tidal flats in the East Frisian Wadden sea. Mar Biol 133: 717–726 Blomqvist S (1985) Reliability of core sampling of soft bottom sediment—an in situ study. Sedimentology 32: 605–612 Blomqvist S (1990) Sampling performance of Ekman grabs—in situ observations and design improvements. Hydrobiologia 206: 245–254 Blomqvist S (1991) Quantitative sampling of soft-bottom sediments: problems and solutions. Mar Ecol Prog Ser 72: 295–304 Blomqvist S, Abrahamsson B (1985) An improved Kajak-type gravity core sampler for soft bottom sediments. Schweiz Z Hydrol 47: 81–84
430
References
Blomqvist S, Abrahamsson B (1987) A device for rapid sectioning of soft bottom sediment cores. Schweiz Z Hydrol 49: 393–396 Blott SJ, Pye K (2001) Gradistat: a grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth Surf Process Landforms 26: 1237–1248 Bluhm BA, Gradinger R, Piraino S (2007) First record of sympagic hydroids (Hydrozoa, Cnidaria) in Arctic coastal fast ice. Polar Biol 30: 1557–1563 Blumenthal T, Davis RE (2004) Exploring nematode diversity. Nat Genet 36: 1246–1247 Boaden PJS (1962) Colonization of graded sand by interstitial fauna. Cah Biol Mar 3: 245–248 Boaden PJS (1964) Grazing in the interstitial habitat: a review. In: Crisp (ed) Grazing in terrestrial and marine environments. Blackwell Science , Oxford, pp. 299–303 Boaden PJS (1968) Water movement—a dominant factor in interstitial ecology. Sarsia 34: 125–136 Boaden PJS (1974) Three new thiobiotic Gastrotricha. Cah Biol Mar 15: 367–378 Boaden PJS (1975) Anaerobiosis, meiofauna and early metazoan evolution. Zool Scr 4: 21–24 Boaden PJS (1977) Thiobiotic facts and fancies (aspects of the distribution and evolution of anaerobic meiofauna). Mikrofauna Meeresboden 61: 45–63 Boaden PJS (1989a) Adaptation of intertidal sand meiofaunal oxygen uptake to temperature and population density. Sci Mar 53: 329–334 Boaden PJS (1989b) Meiofauna and the origins of the Metazoa. Zool J Linnean Soc 96: 217–227 Boaden PJS (1996) Habitat provision for meiofauna by Fucus serratus epifauna with particular data on the flatworm Monocelis lineata. Mar Ecol PSZN 17: 67–75 Boaden PJS, Erwin DG (1971) Turbanella hyalina versus Protodriloides symbioticus: a study in interstitial ecology. Vie Milieu, Suppl. 22: 479–492 Boaden PJS, Platt HM (1971) Daily migration patterns in an intertidal meiobenthic community. Thalass Jugosl 7: 1–12 Bock E, Wilderer PA, Freitag A (1988) Growth of Nitrobacter in the absence of dissolved oxygen. Wat Res 22: 245–250 Bodin P (1988) Results of ecological monitoring three beaches polluted by the ‘Amoco Cadiz’ oil spill: development of meiofauna from 1978–1984. Mar Ecol Prog Ser 42: 105–123 Bodin P (1991) Perturbations in the reproductive cycle of some harpacticoid copepod species further to the Amoco Cadiz oil spill. Hydrobiologia 209: 245–258 Bodin P (1997) Catalogue of the new marine harpacticoid copepods (1997 edition), Doc Trav Inst Sci Nat Belg, Brussels Bodin P, Boucher D (1983) évolution à moyen terme du méiobenthos et des pigments chlorophylliens sur quelques plages polluées par la marée noire de l’Amoco Cadiz. Oceanol Acta 6: 321–332 Boeckner MJ, Proctor HC (2008) Impact of clams on marine meiofaunal assemblages in soft sediment: digging for answers. In: Santos PJP dos (ed) Abstr 13th Int Meiof Conf, Recife Brazil, O 10 Boesgaard TM, Kristensen RM (2001) Tardigrades from Australian marine caves. With a redescription of Actinarctus neretinus (Arthrotardigrada). Zool Anz 240: 253–264 Boisseau JP (1957) Technique pour l’ etude de la faune interstitielle des sables. C R Congr Savent Paris Départment Bordeaux, pp. 117–119 Bolam SG, Schratzberger M, Whomersley P (2006) Macro- and meiofaunal recolonisation of dredged material used for habitat enhancement: temporal patterns in community development. Mar Pollut Bull 52: 1746–1755 Bongers T (1990) The maturity index: an ecological measure of environmental disturbance based on nematode species composition. Oecologia 83: 14–19 Bongers T, Alkemade R, Yeates GW (1991) Interpretation of disturbance-induced maturity decrease in marine nematode assemblages by means of the maturity index. Mar Ecol Prog Ser 76: 135–142 Bongers T, Van de Haar J (1990) On the potential of basing an ecological typology of aquatic sediments on the nematode fauna: an example from the river Rhine. Hydrobiol Bull 24: 37–45
References
431
Booij K, Helder W, Sundby B (1991) Rapid redistribution of oxygen in a sandy sediment induced by changes in the flow velocity of the overlying water. Neth J Sea Res 28: 149–165 Borchardt MA, Bott TL (1995) Meiofaunal grazing of bacteria and algae in a piedmont stream. J N Am Benthol Soc 14: 278–298 Botosaneanu L (1986a) Stygofauna Mundi. A faunistic, distributional, and ecological synthesis of the world fauna inhabiting subterranean waters (including the marine interstitial). Brill/ Backhuys, Leiden, p. 740 Botosaneanu L (1986b) Spelaeogriphacea. In: Botosaneanu L (ed) Stygofauna Mundi. A faunistic, distributional, and ecological synthesis of the world fauna inhabiting subterranean waters (including the marine interstitial). Brill/Backhuys, Leiden, p. 493 Botosaneanu L, Holsinger JR (1991) Some aspects concerning colonization of the subterranean realm—especially of subterranean waters: a response to Rouch & Danielopol, 1987. Stygologia 6: 11–39 Bottjer DJ, Hagadorn JW, Dornbos SQ (2000) The Cambrian substrate revolution. Geol Soc Amer (GSA) Today 10: 1–9 Bou C (1974) Les méthodes de récolte dans les eaux souterraines interstitielles. Ann Spéléol 29: 611–619 Boucher G (1980) Impact of Amoco Cadiz oil spill on intertidal and sublittoral meiofauna. Mar Pollut Bull 11: 95–100 Boucher G (1985) Long-term monitoring of meiofauna densities after the Amoco Cadiz oil spill. Mar Pollut Bull 16: 328–333 Boucher G (1990) Pattern of nematode species diversity in temperate and tropical subtidal sediments. Mar Ecol PSZN 11: 133–146 Boucher G, Chamroux S (1976) Bacteria and meiofauna in an experimental sand ecosystem. I. Material and preliminary results. J Exp Mar Biol Ecol 24: 237–249 Boucher G, Lambshead PJD (1995) Ecological biodiversity of marine nematodes in samples from temperate, tropical and deep-sea regions. Conserv Biol 9: 1594–1604 Bouck L, Thistle D (1998) A computer-assisted method for producing illustrations for taxonomic descriptions. Vie Milieu 49: 101–105 Boudreau BP, Jørgensen BB (eds) (2001) The Benthic boundary layer. Oxford University Press, Oxford Bouguenec V, Giani N (1989) Biological studies upon Enchytraeus variatus Bougenec & Giani 1987 in breeding cultures. Hydrobiologia 180: 151–165 Bouillon J, Grohmann PA (1990) Pinushydra chiquitita gen. et sp. nov. (Cnidaria, Hydrozoa, Athecata), a solitary marine mesopsammic polyp. Cah Biol Mar 31: 291–305 Bouquegneau JM, Joiris C (1988) The fate of stable pollutants—heavy metals and organochlorines in marine organisms. Adv Comp Environ Physiol 2: 219–247 Boutin C, Coineau N (1991) Regression model, “modèle biphase” d’évolution et origine des microorganismes stygobies interstitiels continentaux. Rev Micropaléontol 33: 303–322 Bowman TE, Iliffe TM (1985) Mictocaris halope, a new unusual peracaridian crustacean from marine caves on Bermuda. J Crust Biol 5: 58–73 Boxshall GA, Halsey SH (2004) An introduction to copepod diversity. Vol I + II, Scion Publ, Bloxham, Oxfordshire, UK, p. 966 Bracken M, Gonzalez-Dorantes C, Stachowicz J (2007) Whole-community mutualism: associated invertebrates facilitate a dominant habitat-forming seaweed. Ecology 88: 2211–2219 Bradshaw C, Kumblad L, Fagrell A (2006) The use of tracers to evaluate the importance of bioturbation in remobilising contaminants in Baltic sediments. Estuar Coast Shelf Sci 66: 123–134 Brandt A et al. (2007) First insights into the biodiversity and biogeography of the Southern Ocean deep sea. Nature 447: 307–311 Bretschko G, Klemens WE (1986) Quantitative methods and aspects in the study of the interstitial fauna of running waters. Stygologia 2: 297–316 Brey T (1990) Estimating productivity of macrobenthic invertebrates from biomass and mean individual weight. Meeresforsch 32: 329–343
432
References
Brey T (2001) Population dynamics in benthic invertebrates. A virtual handbook. Version 01.2. Alfred Wegener Institute for Polar and Marine Research, Germany (see http://www.awibremerhaven.de/Benthic/Ecosystem/FoodWeb/Handbook/main.html) Bright M, Giere O (2005) Microbial symbiosis in Annelida. Symbiosis 38: 1–45 Brinkhurst RO (1982a) Oligochaeta. In: Parker SP (ed) Synopsis and classification of living organisms. McGraw Hill, New York, pp. 50–61 Brinkhurst RO (1982b) Evolution in the Annelida. Can J Zool 60: 1043–1059 Brinkhurst RO (1984) The position of the Haplotaxidae in the evolution of oligochaete annelids. Hydrobiologia 115: 25–36. Brinkhurst RO (1991) Phylogenetic analysis of the Tubificinae (Oligochaeta, Tubificidae. Can J Zool 69: 392–397 Brown RJ, Rundle SD, Hutchinson TH, Williams TD, Jones MB (2003) Small-scale detritusinvertebrate interactions: influence of detrital biofilm composition on development and reproduction in a meiofauna copepod. Arch Hydrobiol 157: 117–129 Brown RJ, Rundle SD, Hutchinson TH, Williams TD, Jones MB (2005) A microplate freshwater copepod bioassay for evaluating acute and chronic effects of chemicals. Environ Toxicol Chem 24: 1528–1531 Brown TJ, Sibert JR (1977) The food of some benthic harpacticoid copepods. J Fish Res Bd Can 34: 1028–1031 Brunke M (1999) Colmation and depth filtration within streambeds: retention of particles in hyporheic interstices. Intern Rev Hydrobiol 84: 99–117 Brunke M, Gonser T (1997) The ecological significance of exchange processes between rivers and groundwater. Freshwat Biol 37: 1–33 Bryan JR, Riley JP, LeWilliams B (1976) A Winkler procedure for making precise measurements of oxygen concentration for productivity and related studies. J Exp Mar Biol Ecol 21: 191–197 Bryant C (1991) Metazoan life without oxygen. Chapman & Hall, London, New York, p.291 Buchanan JB (1971) Sediments. In: Buchanan JB, Kain JM (eds) Measurement of the physical and chemical environment. Blackwell Science, Oxford, pp. 30–52 Buck KR, Barry JP (1998) Monterey Bay cold seep infauna: quantitative comparison of bacterial mat meiofauna with non- seep control sites. Cah Biol Mar 39: 333–336 Buffan-Dubau E, Carman KR (2000) Diel feeding behavior of meiofauna and their relationship with microalgal resources. Limnol Oceanogr 45: 381–395 Buffan-Dubau E, Castel J (1996) Diel and seasonal vertical distribution of meiobenthic copepods in muddy sediments of a eutrophic lagoon (fish ponds of Arcachon Bay). Hydrobiologia 329: 69–78 Bunke D (1967) Zur Morphologie und Systematik der Aeolosomatidae Beddard 1895 und Potamodrilidae nov.fam. (Oligochaeta). Zool Jb Syst 94: 187–368 Burd BJ, Nemec A, Brinkhurst RO (1990) The development and application of analytical methods in benthic marine infaunal studies. Adv Mar Biol 26: 169–247 Burdige DJ (2002) Sediment pore waters. In: Hansell DA, Carlson CA (eds) Biogeochemistry of marine dissolved organic matter. Academic, London, pp. 612–664 Burgess R (2001) An improved protocoll for separating meiofauna from sediments using collodial silica sols. Mar Ecol Prog Ser 214: 161–165 Burkovsky LV, Azovsky A, Mokiyevsky VO (1994) Scaling in benthos: from microfauna to macrofauna. Arch Hydrobiol, Suppl. 99: 517–535 Burton SM, Rundle SD, Jones MB (2001) The relationship between trace metal contamination and stream meiofauna. Environ Pollut 111: 159–167 Bussau C, Schriever G, Thiel H (1995) Evaluation of abyssal metazoan meiofauna from a manganese nodule area of the eastern South Pacific. Vie Milieu 45: 39–48 Butman CA (1986) Larval settlement of soft-sediment invertebrates: the spatial scales of pattern explained by active habitat selection and the emerging role of hydrodynamical processes. Oceanogr Mar Biol Ann Rev 25: 113–166 Butman CA (1989) Sediment-trap experiments on the importance of hydrodynamic processes in distributing settling larvae in near-bottom waters. J Exp Mar Biol Ecol 134: 37–88
References
433
Cadée GC (2001) Sediment dynamics by bioturbating organisms. In: Reise K (ed) Ecological comparisons of sedimentary shores. Springer, Berlin, pp. 127–148 Cai WJ, Reimers CE (1993) The development of pH and pCO2 microelectrodes for studying the carbonate chemistry of pore waters near the sediment water interface. Limnol Oceanogr 38: 1762–1773 Cammen LM (1982) Effect of particle size on organic content and microbial abundance within four marine sediments. Mar Ecol Prog Ser 9: 273–280 Canfield DE (1998) A new model for Proterozoic ocean chemistry. Nature 393: 450–452 Cannon LRG (1986) Turbellaria of the world: a guide to families and genera. Queensland Museum, Brisbane, Australia, p. 136 Caramujo M-J, Boschker HTS, Admiraal W (2008) Fatty acid profiles of algae mark the development and composition of harpacticoid copepods. Freshw Biol 53: 77–90 Carbone C, Gittleman JL (2002) A common rule for the scaling of carnivore density. Science 295: 2273–2276 Carey jr AG (1985) Marine ice fauna: Arctic. In: Horner RA (ed) Sea ice biota. CRC Press, Boca Raton, FL, pp. 173–190 Carey jr AG (1992) The ice fauna in the shallow southwestern Beaufort Sea, Arctic Ocean. J Mar Syst 3: 225–236 Carey AG, Montagna PA (1982) Arctic sea ice faunal assemblage: first approach to description and source of the underice meiofauna. Mar Ecol Prog Ser 8: 1–8 Carey PG (1992) Marine interstitial ciliates. An illustrated key. Chapman & Hall, London, New York, p 351 Carlén A, Ólafsson E (2002) The effects of the gastropod Terebralia palustris on the infaunal communities in a tropical mud-flat in East Africa. Wetlands Ecol Manag 10: 303–311 Carlton JT, Geller JB (1993) Ecological roulette: the global transport of nonindigenous marine organisms. Science 261: 78–82 Carlton JT, Hodder J (1995) Biogeography and dispersal of coastal marine organisms: experimental studies on a replica of a 16th century sailing vessel. Mar Biol 121: 721–730 Carman KR (1994) Stimulation of marine free-living and epibiotic bacterial activity by copepod excretions. FEMS Microbiol Ecol 14: 255–262 Carman KR, Thistle D (1985) Microbial food partitioning by three species of benthic copepods. Mar Biol 88: 143–148 Carman KR, Todaro MA (1996) Influence of polycyclic aromatic hydrocarbons on the meiobenthic copepod community of Louisiana salt marsh. J Exp Mar Biol Ecol 198: 37–54 Carman KR, Fleeger JW, Pomarico SM (1997) Response of a benthic food web to hydrocarbon contamination. Limnol Oceanogr 42: 561–571 Carman KR, Thistle D, Fleeger JW, Barry JP (2004) Influence of introduced CO2 on deep-sea metazoan meiofauna. J Oceanogr 60: 767–772 Carney RS (2007) Use of diversity estimations in the study of sedimentary benthic communities. Oceanogr Mar Biol Annu Rev 45: 139–172 Carr RS, Chapman DC, Presley BJ, Biedenbach JM, Robertson L, Boothe P, Kilada R, Wade T, Montagna P (1996) Sediment porewater toxicity assessment studies in the vicinity of offshore oil and gas production platforms in the Gulf of Mexico. Can J Fish Aquat Sci 53: 2618–2628 Cary TL, Chandler GT, Volz DC, Walse SS, Ferry JL (2004) Phenylpyrazole insecticide fipronil induces male infertility in the estuarine meiobenthic crustacean Amphiascus tenuiremis. Environ Science Technol 38: 522–528 Casanova J-P, Perez Y (2000) A dwarf Spadella (Chaetognatha) from Bora Bay (Miyako Island, Japan). Cah Biol Mar 41: 137–141 Castel J (1992) The meiofauna of coastal lagoon ecosystems and their importance in the food web. Vie Milieu 42: 125–135 Casu M, Curini-Galletti M (2006) Genetic evidence for the existence of cryptic species in the mesopsammic flatworm Pseudomonocelis ophiocephala (Rhabditophora: Proseriata). Biol J Linn Soc 87: 553–576
434
References
Ceccherelli VU, Mistri M (1991) Production of the meiobenthic harpacticoid copepod Canuella perplexa. Mar Ecol Prog Ser 68: 225–234 Cedhagen T (1989) A method for disaggregating clay concretions and eliminating formalin smell in the processing of sediment samples. Sarsia 74: 221–222 Chandler GT (1986) High-density culture of meiobenthic harpacticoid copepods within a muddy sediment substrate. Can J Fish Aquat Sci 43: 53–59 Chandler GT (1989) Foraminifera may structure meiobenthic communities. Oecologia 81: 354–360 Chandler GT (1990) Effects of sediment-bound residues of the pyrethroid insecticide Fenvalerate on survival and reproduction of meiobenthic copepods. Mar Environ Res 29: 65–76 Chandler GT (2004) Standard guide for conducting renewal microplate-based life-cycle toxicity tests with a marine meiobenthic copepod. In: Bailey SJ et al. (eds) ASTM E 2317–04, American Society for Testing and Materials, West Conshohocken, PA, USA, p. 15 Chandler GT, Fleeger JW (1984) Tube-building by a marine meiobenthic harpacticoid copepod. Mar Biol 82: 15–19 Chandler GT, Fleeger JW (1987) Facilitative and inhibitory interactions among estuarine meiobenthic harpacticoid copepods. Ecology 68: 1906–1919 Chandler GT, Green AS (2001) Development-stage specific life-cycle bioassay for assessment of sediment-associated toxicant effects on benthic copepod production. Environ Toxicol Chem 20: 171–178 Chandler GT, Volz DC (2004) Semiquantitative confocal laser scanning microscopy applied to marine invertebrate ecotoxicology. Mar Biotechnol 6: 128–137 Chandler GT, Shirley TC, Fleeger JW (1988) The tom-tom corer: a new design of the Kajak corer for use in meiofauna sampling. Hydrobiologia 169: 129–134 Chandler GT, Cary TL, Bejarano AC, Pender J, Ferry JL (2004a) Population consequences of Fipronil and degradates to copepods at field concentrations: an integration of life cycle testing with Leslie matrix population modeling. Environ Sci Technol 38: 6407–6414 Chandler GT, Tawnya LC, Volz DC, Walse SS, Ferry JL, Klosterhaus SL (2004b) Fipronil effects on estuarine copepod (Amphiascus tenuiremis) development, fertility, and reproduction: a rapid life-cycle assay in 96-well microplate format. Environ Toxicol Chem 23: 117–124 Chao A, Chazdon RL, Colwell RK, Shen T-J (2006) Abundance-based similarity indices and their estimation when there are unseen species in samples. Biometrics 62: 361–371 Chao A, Shen TJ (2003) Nonparametric estimation of Shannon’s index of diversity when there are unseen species in sample. Environ Ecol Statist 10: 429–443 Chapman PM (1986) Sediment quality criteria from the sediment quality triad: an example. Environ Toxicol Chem 5: 957–964 Chappuis PA (1942) Eine neue Methode zur Untersuchung der Grundwasserfauna. Acta Sci Math Nat Univ Franzisco-Josephina 6: 1–7 Chardy P, Dauvin J-C (1992) Carbon flows in a subtidal fine sand community from the western English Channel: a simulation analysis. Mar Ecol Prog Ser 81: 147–161 Chen GT, Herman RL, Vincx M (1999) Meiofauna communities from the Straits of Magellan and the Beagle Channel. Sci Mar 63, Suppl. 1: 123–132 Chen J-Y, Bottjer DJ, Oliveri P, Dornbos SQ, Gao F, Ruffins S, Chi H, Li C-W, Davidson EH (2004) Small bilaterian fossils from 40 to 55 million years before the Cambrian. Science 305: 218–222 Chertoprud ES, Chertoprud MV, Garlitskaya LA, Azovsky AI, Kondar DV (2007) Spatial variability of the structure of the Harpacticoida (Copepoda) crustacean assemblages in intertidal and shallow water zones of European seas. Oceanology 47: 51–59 Chester R (2002) Marine geochemistry. Blackwell, Oxford Chesunov AV (2006) Biologiya morskikh nematod (Biology of marine nematodes). Tovarishchestvo Nauchnyk Izdani KMK, Moscow, p. 367 Child CA (1988) Pycnogonida. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 423–424 Chinnadurai G, Fernando OJ (2007) Meiofauna of mangroves of the southeast coast of India with special reference to the free-living marine nematode assemblage. Estuar Coast Shelf Sci 72: 329–336
References
435
Chislenko LL (1968) Nomograms for calculation of the weight of aquatic organisms according to body size and shape. (in Russian) Nauka, Leningrad Chitwood BG, Chitwood MB (1974) An introduction to nematology. Sect. 1, Anatomy. University Park Press, Baltimore, p. 334 Christensen B, O’Connor FB (1958) Pseudofertilization in the genus Lumbricillus (Enchytraeidae). Nature 181: 1085–1086 Christie H, Berge JA (1995) In situ experiments on recolonization of intertidal mudflat fauna to sediment contaminated with different concentrations of oil. Sarsia 80: 175–185 Chua KE, Brinkhurst RO (1973) Evidence of interspecific interactions in the respiration of tubificid oligochaetes. J Fish Res Board Can 30: 617–622 Clarke A (1992) Is there a latitudinal diversity cline in the sea? Trends Ecol Evol 7: 286–287 Clarke KR (1993) Non-parametric multivariate analyses of changes in community structure. Aust J Ecol 18: 117–143 Clarke KR, Warwick RM (1998) A taxonomic distinctness index and its statistical properties. J Appl Ecol 35: 523–531 Clarke KR, Warwick RM (2001b) A further biodiversity index applicable to species lists: variation in taxonomic distinctness. Mar Ecol Prog Ser 216: 265–278 Clausen C (2000) Gastrotricha Macrodasyoida from the Tromsø region, Northern Norway. Sarsia 85: 357–384 Clément P (1993) The phylogeny of rotifers: molecular, ultrastructural and behavioural data. Hydrobiologia 255–256: 527–544 Clément P, Wurdak E (1991) Rotifera. In: Harriso FW, Ruppert EE (eds) Microscopic anatomy of invertebrates. Vol. 4. Aschelminthes. Wiley-Liss, Wilmington, USA, pp. 219–296 Cline J (1969) Spectrophotometric determination of hydrogen sulfide in natural waters. Limnol Oceanogr 14: 454–458 Coineau N (2000) Adaptations to interstitial groundwater life. In: Wilkens H, Culver DC, Humphreys WF (eds) Subterranean ecosystems. Elsevier, Amsterdam, pp. 189–210 Coineau Y, Haupt J, Delamare Deboutteville C (1978) Un remarquable exemple de convergence écologique: l’adaptation de Gordialycus tuzetae (Nematalycidae, Acariens) à la vie dans les interstices des sables fins. C R Acad Sci Paris 287: 883–886 Colacino JM, Kraus DW (1984) Hemoglobin-containing cells of Neodasys (Gastrotricha, Chaetonotida). II. Respiratory significance. Comp Biochem Physiol 79 A: 363–369 Colombini I, Berti R, Nocita A, Chelazzi L (1996) Foraging strategy of the mudskipper Periophthalmus sobrinus Eggert in a Kenyan mangrove. J Exp Mar Biol Ecol 197: 219–235 Commito JA, Tita G (2002) Differential dispersal rates in an intertidal meiofauna assemblage. J Exp Mar Biol Ecol 268: 237–256 Condé B (1965) Présence de Palpigrades dans le milieu interstitiel littoral. C R Acad Sci Paris 261: 1898–1900 Connell JH, Slatyer RD (1977) Mechanisms of succession in natural communities and their role in community stability and organisation. Amer Natur 111: 1119–1144. Conrad JE (1976) Sand grain angularity as a factor affecting colonization by marine meiofauna. Vie Milieu B 26: 81–198 Conway Morris S (1998) Early metazoan evolution: reconciling paleontology and molecular biology. Am Zool 38: 867–877 Conway Morris S, George JD, Gibson R, Platt HM (1985) The origins and relationships of lower invertebrates. Clarendon, Oxford Cook AA, Bhadury P, Debenham NJ, Meldal BHM, Blaxter ML, Smerdon GR, Austen MC, Lambshead PJD, Rogers AD (2005) Denaturing Gradient Gel Electrophoresis (DGGE) as a tool for the identification of marine nematodes. Mar Ecol Prog Ser 291: 103–113 Cook AA, Lambshead JPD, Hawkins LE, Mitchella N, Levin LA (2000) Nematode abundance at the oxygen minimum zone in the Arabian Sea. Deep-Sea Res II 47: 75–85 Cook PL (1963) Observations on live lunulitiform zoaria of Polyzoa. Cah Biol Mar 4: 407–413. Cook PL (1966) Some “sand fauna” polyzoa (Bryozoa) from Eastern Africa and the Northern Indian Ocean. Cah Biol Mar 7: 207–223
436
References
Cook PL (1988) Bryozoa. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 438–443 Coomans A (2000) Nematode systematics: past, present and future. Nematology 2: 3–7 Copley J, Flint H, Ferrero T, Van Dover C (2007) Diversity of meiofauna and free-living nematodes in hydrothermal vent mussel beds on the northern and southern East Pacific Rise. J Mar Biol Ass UK 87: 1141–1152 Corliss JO (1974) The changing world of ciliate systematics: historical analysis of past efforts and a newly proposed phylogenetic scheme of classification for the protistan phylum Ciliophora. Syst Zool 23: 91–138 Corliss JO (1975) Taxonomic characterization of the suprafamilial groups in a revision of recently proposed schemes of classification for the phylum Ciliophora. Trans Am Microsc Soc 94: 224–267 Corliss JO (1979) The ciliated Protozoa: characterization, classification, and guide to the literature. Pergamon, Oxford Costello MJ, Bouchet P, Emblow CS, Legakis A (2006) European marine biodiversity inventory and taxonomic resources: state of the art and gaps in knowledge. Mar Ecol Prog Ser 316: 257–268 Couch CA (1988) A procedure for extracting large numbers of debris-free, living nematodes from muddy marine sediments. Trans Am Microsc Soc 107: 96–100 Couch CA (1989) Carbon and nitrogen stable isotopes of meiobenthos and their food resources. Estuar Coast Shelf Sci 28: 433–442 Coull BC (1970) Shallow water meiobenthos of the Bermuda platform. Oecologia (Berl) 4: 325–357 Coull BC (1973) Estuarine meiofauna: a review: trophic relationships and microbial interactions. In: Stevenson LH, Colwell ARR (eds) Estuarine microbial ecology. University of South Carolina Press, Columbia, SC, pp. 499–512 Coull BC (1985) Long-term variability of estuarine meiobenthos: an 11 year study. Mar Ecol Prog Ser 24: 205–218. Coull BC (1986) Long-term variability of meiobenthos: value, synopsis, hypothesis generation and predictive modelling. Hydrobiologia 142: 271–279 Coull BC (1990) Are members of the meiofauna food for higher trophic levels? Trans Am Microsc Soc 109: 233–246 Coull BC (1999) Role of meiofauna in estuarine soft-bottom habitats. Austr J Ecol 24: 327–343 Coull BC, Bell SS (1979) Perspectives of marine meiofaunal ecology. In: Livingston RJ (ed) Ecological processes in coastal and marine systems. Plenum, New York, pp. 189–216 Coull BC, Chandler GT (1992) Pollution and meiofauna: field, laboratory, and mesocosm studies. Oceanogr Mar Biol Annu Rev 30: 191–271 Coull BC, Giere O (1988) The history of meiofaunal research. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 14–17 Coull BC, Grant J (1981) Encystment discovered in a marine copepod. Science, Wash 212: 342–344 Coull BC, Palmer MA (1984) Field experimentation in meiofaunal ecology. Hydrobiologia 118: 1–19 Coull BC, Wells JBJ (1983) Refuges from fish predation: experiments with phytal meiofauna from the New Zealand rocky intertidal. Ecology 64: 1599–1609 Coull BC, Ellison RL, Fleeger JW, Higgins RP, Hope WD, Hummon WD, Rieger RM, Sterrer WE, Thiel H (1977) Quantitative estimates of the meiofauna from the deep sea of North Carolina. Mar Biol 39: 233–240 Coull BC, Bell SS, Savory AM, Dudley BW (1979) Zonation of meiobenthic copepods in a southeastern United States salt marsh. Estuar Coast Mar Sci 9: 181–188 Coull BC, Hicks GRF, Wells JBJ (1981) Nematode/copepod ratios for monitoring pollution: a rebuttal. Mar Pollut Bull 12: 378–381 Coull BC, Greenwood JG, Fielder DR, Coull BA (1995) Subtropical Australian juvenile fish eat meiofauna: experiment with winter whiting Sillago maculata and observations on other species. Mar Ecol Prog Ser 125: 13–19
References
437
Craib JS (1965) A sampler for taking short undisturbed marine cores. J Cons Perm Int Explor Mer 30: 34–39 Craik GJS (1980) Simple method for measuring the relative scouring of intertidal areas. Mar Biol 59: 257–260 Creed EL, Coull BC (1984) Sand dollar, Mellita quinquiesperforata (Leske), and sea pansy, Renilla reniformis (Cuvier) effects on meiofaunal abundance. J Exp Mar Biol Ecol 84: 225–234 Creutzberg F, Wapenaar P, Duineveld G, Lopez Lopez N (1984) Distribution and density of the benthic fauna in the southern North Sea in relation to bottom characteristics and hydrographic conditions. Rapp P-V Réun Cons Int Explor Mer 183: 101–110 Crezée M (1976) Solenofilomorphidae (Acoela), major component of a new turbellarian association in the sulfide system. Int Rev Ges Hydrobiol 61: 5–129 Crisp DJ (1984) Energy flow measurements. In: Holme NA, McIntyre AD (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 284–372 Crisp DJ, Mwaiseje B (1989) Diversity in intertidal communities with special reference to the Corallina officinalis community. Sci Mar 53: 365–372 Crowe JH, Oliver AE, Tablin F (2002) Is there a single biochemical adaptation to anhydrobiosis? Integr Comp Biol 42: 497–503 Cullen DJ (1973) Bioturbation of superficial marine sediments by interstitial meiobenthos. Nature 242: 323–324 Culver SJ, Buzas MA (2000) Global latitudinal species diversity gradient in deep-sea benthic Forminifera. Deep-Sea Res I 47: 259–275 Curini-Galletti M, Campus P, Delogu V (2008) Theama mediterranea sp. nov. (Platyhelminthes, Polycladida), the first interstitial polyclad from the Mediterranean. Ital J Zool 75: 77–83 Dahms HU (1990) Naupliar development of Harpacticoida (Crustacea, Copepoda) and its significance for phylognetic systematics. Microfauna Mar 6: 169–272 Dahms H-U (1991) Erster Nachweis eines Harpacticoiden (Copepoda) mit zystenloser Diapause. First indication of nonencysted diapause for Harpacticoida (Copepoda). Verh Dtsch Zool Ges 84: 442–443 Dahms H-U (1992) Metamorphosis between naupliar and copepodid phases in the Harpacticoida. Phil Trans R Soc Lond B 335: 221–236 Dahms H-U (1993) Copepodid development in Harpacticoida (Crustacea, Copepoda). Microfauna Mar 8: 195–245. Dahms H-U, Bergmans M, Schminke HK (1990) Distribution and adaptations of sea ice inhabiting Haracticoida (Crustacea, Copepoda) of the Weddell Sea (Antarctica. Mar Ecol PSZN 11: 207–226 Dahms H-U, Harder T, Qian P-Y (2004) Effects of meiofauna on macrofauna recruitment: settlement inhibition of the polychaete Hydroides elegans by the harpacticoid copepod Tisbe japonica. J Exp Mar Biol Ecol 311: 47–61 Dahms H-U, Harder T, Qian PY (2007) Selective attraction and reproductive performance of a harpacticoid copepod in a response to biofilms. J Exp Mar Biol Ecol 341: 228–238 Danielopol DL (1989) Groundwater fauna associated with riverine aquifers. J N Am Benthol Soc 8: 18–35 Danielopol DL (1990a) On the interest of the “Cytherissa” project and on the present state of researches. Bull Inst Géol Bassin d’Aquitaine, Bordeaux 47: 15–26 Danielopol DL (1990b) The origin of the anchialine cave fauna—the “deep sea” versus the “shallow water” hypothesis tested against the empirical evidence of the Thaumatocyprididae (Ostracoda). Bijdr Dierk 60: 137–143 Danielopol DL (1991) Spatial distribution and dispersal of interstitial Crustacea in alluvial sediments of a backwater of the Danube at Vienna. Stygologia 6: 97–110 Danielopol DL, Niederreiter R (1987) A sampling device for groundwater organisms and oxygen measurement in multi-level monitoring wells. Stygologia 3: 252–263 Danielopol DL, Niederreiter R (1990) New sampling equipment and extraction methods for meiobenthic organisms.. Bull Inst Béol Bassin d’Aquitaine, Bordeaux 47: 277–286 Danielopol DL, Rouch R (1991) L’adaptation des organismes au milieu aquatique souterrain. Réflections sur l’apport des recherches écologiques récentes. Stygologia 6: 129–142
438
References
Danielopol DL, Marmonier P, Boulton AJ, Bonaduce G (1994) World subterranean ostracod biogeography: dispersal or vicariance? Hydrobiologia 287: 119–129 Danielopol DL, Pospisil P, Rouch R (2000a) Biodiversity in groundwater: a large-scale view. Trends Ecol Evolut 15: 223–224 Danielopol DL, Pospisil P, Dreher J, Mösslacher F, Torreiter P, Geiger-Kaiser M, Gunatilaka A (2000b) A groundwater ecosystem in the Danube wetlands at Wien (Austria). In: Wilkens H, Culver DC, Humphreys WF (eds) Subterranean ecosystems. Elsevier, Amsterdam, pp. 481–511 Dannheim J, Struck U, Brey T (2007) Does sample bulk freezing affect stable isotope ratios of infaunal macrozoobenthos?. J Exp Mar Biol Ecol 351: 37–41 Danovaro R (1996) Detritus-bacteria-meiofauna interactions in a seagrass bed (Posidonia oceanica) of the NW Mediteranean. Mar Biol 127: 1–13 Danovaro R, Fraschetti S (2002) Meiofaunal vertical zonation on hard-bottoms: comparison with soft-bottom meiofauna. Mar Ecol Prog Ser 230: 159–169 Danovaro R, Fraschetti S, Belgrano A, Vincx M, Curini-Galletti M, Albertelli G, Fabiano M (1993) The potential impact of meiofauna on the recruitment of macrobenthos in a subtidal coastal benthic community of the Ligurian Sea (nort-western Mediterranean): a field study. In: Eleftheriou A, Ansell AD, Smith CJ (eds) Biology and ecology of shallow coastal waters. Olsen & Olsen, Fredensborg, Denmark, pp. 115–122 Danovaro R, Fabiano M, Vincx M (1995) Meiofauna response to the Agip Abruzzo oil spill in subtidal sediments of the Ligurian Sea. Mar Pollut Bull 39: 133–145 Danovaro R, Gambi C, Mirto S (2002) Meiofaunal production and energy transfer efficiency in a seagrass bed (Posidonia oceanica) of the Western Mediterranean. Mar Ecol Prog, Ser 234: 95–104 Danovaro R, Scopa M, Gambi C, Fraschetti S (2007) Trophic importance of subtidal metazoan meiofauna: evidence from in situ exclusion experiments on soft and rocky substrates. Mar Biol 152: 339–350 Dash MC, Cragg JB (1972) Selection of microfungi by Enchytraeidae (Oligochaeta) and other members of the soil fauna. Pedobiologia 12: 282–286 Dauer DM, Luckenbach MW, Rodi AJ (1993) Abundance biomass comparison (ABC method)— effects of an estuarine gradient, anoxic hypoxic events and contaminated sediments. Mar Biol 116: 507–518 Davey JT, Watson PG, Bruce RH, Frickers PE (1990) An instrument for the monitoring and collection of the vented burrow fluids of benthic infauna in sediment microcosms and its application to the polychaetes Hediste diversicolor and Arenicola marina. J Exp Mar Biol Ecol 139: 135–149 De Beer D, Wenzhöfer F, Ferdelman TG, Boehme SE, Huettel M, Beusekom JEEV, Boettcher M, Musat N, Dubilier N (2005) Transport and mineralization rates in North Sea sandy intertidal sediments, Sylt-Rømø Basin, Wadden Sea. Limnol Oceanogr 50: 113–127 De Bovée F (1987) Biomasse et équivalents énergétiques des Nématodes libres marins. Cah Biol Mar 28: 367–372 De Bovée F, Labat JP (1993) A simulation model of a deep meiobenthic compartment: a preliminary approach. Mar Ecol PSZN 14: 159–173 De Bovée F, Soyer J (1974) Cycle annuel quantitatif du méiobenthos des vases terrigènes côtières. Distribution verticale. Vie Milieu B 24: 141–157 De Bovée F, Yacoubi-Khebiza M, Coineau N, Boutin C (1995) Influence du substrat sur la répartition des Crustacés stygobies interstitiels du Haut-Atlas occidental (The influence of sediment granulometry in the distribution of the interstitial stygobiotic crustaceans in the western High Atlas). Intern Rev Ges Hydrobiol 80: 453–468 De Brouwer JFC, Stal LJ (2001) Short-term dynamics in microphytobenthos distribution and associated extracellular carbohydrates in surface sediments of an intertidal mudflat. Mar Ecol Prog Ser 218: 33–44 De Broyer C, Hecq J-H, Vanhove S (2001) Life under the ice: biodiversity of the Southern Ocean. The Belgica expedition centennial: perspectives on Antarctic science and history. Brussels University Press, Brussels, pp. 271–286 De Deckere EMGT, Tolhurst TJ, De Brouwer JFC (2001) Destabilization of cohesive intertidal sediment by infauna. Estuar Coast Shelf Sci 53: 665–669
References
439
De Deckker P, Forester RM (1988) The use of Ostracoda to reconstruct continental plaeoenvironmental records. In: De Deckker Pea (ed) Ostracoda in the earth sciences. Elsevier, Amsterdam, pp. 175–197 De Jonge VN, Bouwman LA (1977) A simple density separation technique for quantitative isolation of meiobenthos, using colloidal Ludox TM. Mar Biol 42: 143–148 De Jonge VN, Colijn F (1994) Dynamics of microphytobenthos biomass in the Ems estuary. Mar Ecol Prog Ser 104: 185–196 De Ley P, Blaxter M (2004) A new system for Nematoda: combining morphological characters with molecular trees, and translating clades into ranks and taxa. Nematology 2: 633–653 De Ley P, Decraemer W, Eyualem-Abebe E (2006) Introduction: summary of present knowledge and research addressing the ecology and taxonomy of freshwater nematodes. In: EyualemAbebe E, Andrássy I, Traunspurger W (eds) Freshwater nematodes. Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, U.K., pp. 3–30 De Mesel I, Derycke S, Moens T, Van der Gucht K, Vincx M, Swings J (2004) Topdown impact of of bacterivorous nematodes on the bacterial community: a microcosm study. Environ Microbiol 6: 733–744 De Rosa R, Grenier JK, Andreeva T, Cook CE, Adoutte A, Akam M, Carroll BS, Balavoine G (1999) Hox genes in brachiopods and priapulids and protostome evolution. Nature 399: 772–776 De Smet WH (2002) A new record of Limnognathia maerski Kristensen & Funch, 2000 (Micrognathozoa) from the sub-sub-Antarctic Crozet Islands, with description of the trophi. J Zool Lond 258: 381–383 De Troch M, Fiers F, Vincx M (2000) Range extension and microhabitat of Lightiella incisa (Cephalocarida). J Zool 251: 199–204 De Troch M, Fiers F, Vincx M (2002) Niche segregation and habitat specialisation of harpacticoid copepods in a tropical seagrass bed. Mar Biol 142: 345–355 De Troch M, Gurdebeke S, Fiers F, Vincx M (2001) Zonation and structuring factors of meiofauna communities in a tropical seagrass bed (Gezi Bay, Kenya). J Sea Res 45: 45–61 De Troch M, Steinarsdottir M, Chepurnov V, Olafsson E (2005a) Grazing on diatoms by harpacticoid copepods: species-specific density dependent uptake and microbial gardening. Aquat Microb Ecol 39: 135–144 De Troch M, Vandepitte L, Raes M, Suàrez-Morales E, Vincx M (2005b) A field colonization experiment with meiofauna and seagrass mimics: effect of time, distance and leaf surface area. Mar Biol 148: 73–86 De Zio S, Grimaldi P (1966) Ecological aspects of Tardigrada distribution in South Adriatic beaches. Veröff Inst Meeresforsch Bremerhaven, Suppl. 2: 87–94 Debenham NJ, Lambshead PJD, Ferrero TJ, Smith CR (2004) The impact of whale falls on nematode abundance in the deep sea. Deep-Sea Res I 51: 701–706 Decho AW (1990) Microbial exopolymer secretions in ocean environments: their role(s) in food webs and marine processes. Ocenogr Mar Biol Ann Rev 28: 73–153 Decho AW, Fleeger JW (1988a) Microscale dispersion of meiobenthic copepods in response to food-resource patchiness. J Exp Mar Biol Ecol 118: 229–243 Decho AW, Fleeger JW (1988b) Ontogenetic feeding shifts in the meiobenthic copepod Nitocra lacustris. Mar Biol 97: 191–197 Decho AW, Lopez GR (1993) Exopolymer microenvironments of microbial flora: multiple and interactive effects on trophic relationships. Limnol Oceanogr 38: 1633–1645 Decraemer W, Neira C, Backeljau T (2005) A new species of Glochinema (Epsilonematidae: Nematoda) from the oxygen minimum zone off Baja California, NE Pacific and phylogenetic relationships at species level within the family. Cah Biol Mar 46: 105–126 DeFlaun MF, Mayer LM (1983) Relationships between bacteria and grain surfaces in intertidal sediments. Limnol Oceanogr 28: 873–881. Delamare Deboutteville C (1960) Biologie des eaux souterraines littorales et continentales. Vie Milieu, Suppl. 9, p. 740
440
References
Delamare Deboutteville C, Chappuis P-A (1951) Presence de l’ordre des Mystacocarida Pennak et Zinn dans le sable des plages du Roussillon: Derocheilocaris remanei n. sp. C R Acad Sci Paris 233: 437–439 DeNiro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in animals. Geochim Cosmochim Acta 42: 495–506 Deniro MJ, Epstein SS (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochim Cosmochim Acta 45: 341–351 Deprez T (2006) NeMys: an all-round database system for biological information. MarBEF Newsletter 4: 31–32, Digital Version (open file) (see http://nemys.ugent.be/index.asp?c=67) Derycke S, Remerie T, Vierstraete A, Backeljau T, Vanfleteren J, Vincx M, Moens T (2005) Mitochondrial DNA variation and cryptic speciation within the free-living marine nematode Pellioditis marina. Mar Ecol Prog Ser 300: 91–103 Derycke S, Backeljau T, Vlaeminck C, Vierstraete A, Vanfleteren J, Vincx M, Moens T (2007a) Spatiotemporal analysis of population genetic structure in Geomonhystera disjuncta (Nematoda, Monhysteridae) reveals high levels of molecular diversity. Mar Biol 151: 1799–1812 Derycke S, Van Vynckt R, Vanaverbeke J, Vincx M, Moens T (2007b) Colonisation patterns of Nematoda on decomposing algae in the estuarine environment: community assembly and genetic structure of the dominant species Pellioditis marina. Limnol Oceanogr 52: 992–1001 Derycke S, Fonseca G, Vierstraete A, Vanfleteren J, Vincx M, Moens T (2008) Disentangling taxonomy within the Pellioditis marina (Nematoda, Rhabditidae) species complex using molecular and morphological tools. Zool J Linn Soc 152: 1–15 Derycke S, Fonseca G, Vierstraete A, Vanfleteren J, Vincx M, Moens T (2008) Disentangling taxonomy within the Pellioditis marina (Nematoda, Rhabditidae) species complex using molecular and morphological tools. Zool J Linn Soc 152: 1–15 D’Hondt J-L (1971) Gastrotricha. Oceanogr Mar Biol Ann Rev 9: 141–192 Di Sabatino A, Gerecke R, Martin P (2000) The biology and ecology of lotic water mites (Hydrachnidia). Freshw Biol 44: 47–62 Di Sabatino A, Martin P, Gerecke R, Cicolani B (2002) Hydrachnidia (water mites). In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys, Leiden, pp. 105–133 Diaz RJ, Rosenberg R (1995) Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of benthic macrofauna. Oceanogr Mar Biol Annu Rev 33: 245–303 Diaz RJ, Cutter GR, Rhoads DC (1994) The importance of bioturbation to continental slope sediment structure and benthic processes off Cape Hatteras, North Carolina. Deep-Sea Res II 41: 719–734 Dick MH, Buss LW (1994) A PCR-based survey of homeobox genes in Ctenodrilus serratus (Annelida: Polychaeta). Mol Phylogen Evol 3: 146–158 Diederich J, Fortuner R, Milton J (2000) Genisys and computer-assisted identification of nematodes. Nematology 2: 17–30 Dietrich D (1999) Struktur und Funktion benthischer mikrobieller Nahrungsgewebe unter besonderer Berücksichtigung der heterotrophen Flagellaten. Ph.D. thesis, Univ. Köln, Cologne Dinet A, Grassle F, Tunnicliffe V (1988) Premières observations sur la méiofaune des sites hydrothermaux de la dorsale Est-Pacifique (Guaymas 21°N) et de l’Explorer Ridge. Oceanologica Acta Spec Iss. 8: 7–14 Dinet A, Sornin J-M, Sablière A, Delmas D, Feuillet-Girard M (1990) Influence de la biodéposition de bivalves filtreurs sur les peuplements méiobenthiques d’un marais maritime. Cah Biol Mar 31: 307–322 DiPinto LM (1996) Trophic transfer of a sediment-associated organophosphate pesticide from meiobenthos to bottom-feeding fish. Arch Environ Contam Toxicol 30: 459–466 Dittmann S (1987) Die Bedeutung der Biodeposite für die Benthosgemeinschaft der Wattsedimente. Unter besonderer Berücksichtigung der Miesmuschel Mytilus edulis L. Ph.D. thesis, Univ of Hamburg, Hamburg Dittmann S (1996) Effects of macrobenthic burrows on infaunal communities in tropical tidal flats. Mar Ecol Prog Ser 134: 119–130
References
441
Dixon DR (1979) A differential volumetric micro-respirometer for use with small aquatic organisms. J Exp Biol 82: 379–384 Dobbs FC, Guckert JB (1988) Callianassa trilobata (Crustacea: Thalassinidae) influences abundance of meiofauna and biomass, composition, and physiological state of microbial communities within its burrow. Mar Ecol Prog Ser 45: 69–79 Dole Olivier MJ, Galassi DMP, Marmonier P, Des Chatelliers MC (2000) The biology and ecology of lotic microcrustaceans. Freshw Biol 44: 63–91 Dole-Olivier MJ, Marmonier P (1992) Patch distribution of interstitial communities: prevailing factors. Freshw Biol 27: 177–191 Dörjes J (1968) Zur Ökologie der Acoela (Turbellaria) in der Deutschen Bucht. Helgol wiss Meeresunters 18: 78–115 Dornbos SQ, Bottjer DJ, Chen J-Y (2005) Paleoecology of benthic metazoans in the Early Cambrian Maotianshan shale biota and the Middle Cambrian Burgess Shale biota: evidence for the Cambrian substrate revolution. Paleogeogr Paleoclimatol Paleoecol 220: 47–67 Doty MS (1971) Measurement of water movement in reference to benthic algal growth. Bot Mar 14: 32–35 Douglas AE (1988) Alga-invertebrate symbiosis. In: Rogers LJ, Gallon JR (eds) Biochemistry of the Algae and Cyanobacteria. Clarendon, Oxford Dragesco J (1960) Ciliés mésopsammiques littoraux. Systématique, morphologie, écologie. Trav Stat Biol Roscoff 12: 1–356 Dragesco J (1965) Etude cytologique de quelques Flagellés mésopsammiques. Cah Biol Mar 6: 83–115 Drgas A, Radziejewska T, Warzocha J (1998) Biomass size spectra of near shore shallow water benthic communities in the Gulf of Gdansk (southern Baltic Sea). PSZN Mar Ecol 19 (3): 209–228 Dubilier N, Blazejak A, Ruehland C (2006) Symbioses between bacteria and gutless marine oligochaetes. In: Overmann J (ed) Molecular basis of symbiosis. Springer, Berlin, pp. 251–275 Duffy JE, Tyler S (1984) Quantitative differences in mitochondrial ultrastructure of a thiobiotic and oxybiotic turbellarian. Mar Biol 83: 95–102 Dujardin F (1851) Sur un petit animal marin, l’Echinodère, formant un type intermédiaire entre les Crustacés et les Vers. Ann Sci Nat Zoologie, Sér 3 15: 158–160 Duplisea DE, Drgas A (1999) Sensitivity of a benthic, metazoan, biomass size spectrum to differences in sediment granulometry. Mar Ecol Prog Ser 177: 73–82 Duplisea DE, Hargrave BT (1996) Response of meiobenthic size-structure, biomass and respiration to sediment organic enrichment. Hydrobiologia 339: 161–170 Dye AH (1978) An ecophysiological study of the meiofauna of the Swartskops estuary. 1.The sampling sites: physical and chemical features. Zool Afr 13: 1–18 Dye AH (1983) Composition and seasonal fluctuations of meiofauna in a Southern African mangrove estuary. Mar Biol 73: 165–170 Dye AH, Furstenberg JP (1981) Estuarine meiofauna. In: Day JH (ed) Estuarine ecology. Balkema, Rotterdam, pp. 179–186 Dye AH, Lasiak TA (1986) Microbenthos, meiobenthos and fiddler crabs: trophic interactions in a tropical mangrove sediment. Mar Ecol Prog Ser 32: 259–264 Eckman JE (1979) Small-scale patterns and processes in a soft-substratum intertidal community. J Mar Res 37: 437–456 Eckman JE (1983) Hydrodynamic processes affecting benthic recruitment. Limnol Oceanogr 28: 241–257 Eckman JE (1985) Flow disruption by an animal-tube mimic affects sediment bacterial colonization. J Mar Res 43: 419–435 Eckman JE (1990) A model of passive settlement by planktonic larvae onto bottoms of differing roughness. Limnol Oceanogr 35: 887–901 Eckman JE, Thistle D (1991) Effects of flow about a biologically produced structure on harpacticoid copepods in San Diego trough. Deep-Sea Research A 38: 1397–1416 Eckman JE, Nowell ARM, Jumars PA (1981) Sediment destabilization by animal tubes. J Mar Res 39: 361–374
442
References
Edgar GJ (1999) Experimental analysis of structural versus trophic importance of seagrass beds. I. Effects on macrofaunal and meiofaunal invertebrates. Vie Milieu 49: 239–248 Edmonds SJ (1982) A sipunculan reproted to be “interstitial” from the Netherland Antilles. Bijdr Dierk 52: 228–230 Ehlers U (1985) Phylogenetic relationships within the Platyhelminthes. In: Conway Morris S, George JD, Gibson R, Platt HM (eds) The origins and relationships of lower invertebrates. Clarendon, Oxford., pp. 43–158 Eibye-Jacobsen D, Kristensen RM (1994) A new genus and species of Dorvilleidae (Annelida, Polychaeta) from Bermuda, with a phylogenetic analysis of Dorvilleidae, Iphitimidae and Dinophilidae. Zool Scr 23: 107–131 Eibye-Jacobsen J (1997) New observations on the embryology of the Tardigrada. Zool Anz 235: 201–216 Ekman S (1953) Zoogeography of the sea. Sidgwick and Jackson, London, p. 417 Eleftheriou A, McIntyre A (eds) (2005) Methods for the study of marine benthos. 3rd ed. Blackwell, Oxford Eleftheriou A, Moore CG (2005) Macrofauna techniques. In: Eleftheriou A, McIntyre A (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 160–228 Eleftheriou A, Nicholson MD (1975) The effects of exposure on beach fauna. Cah Biol Mar 16: 695–710 Ellison RL (1984) Foraminifera and meiofauna on an intertidal mudflat, Cornwall England: Populations, respiration and secondary production, and energy budget. Hydrobiologia 109: 131–147 Elmgren R (1978) Structure and dynamics of Baltic benthos communities, with particular reference to the relationship between macro- and meiofauna. Kiel Meeresforsch Sh. 4: 1–22 Elmgren R, Radziejewska T (1989) Recommendations for quantitiative benthic meiofauna studies in the Baltic. Balt Mar Biol Publ 12: 1–20 Elmgren R, Hansson S, Larsson U, Sundelin B, Boehm PD (1983) The “Tsesis” oil spill: acute and long-term impact on the benthos. Mar Biol 73: 51–65 Elner RW, Beninger PG, Jackson DL, Potter TM (2005) Evidence of a new feeding mode in western sandpiper (Calidris mauri) and dunlin (Calidris alpina) based on bill and tongue morphology and ultrastructure. Mar Biol 146: 1223–1234 Elofson O (1941) Zur Kenntnis der marinen Ostracoden Schwedens. Zool Bidr Uppsala 19: 217–534 Epstein SS (1995) Simultaneous enumeration of protozoa and micrometazoa from marine sandy sediments. Aquat Microb Ecol 9: 219–227 Epstein SS (1997a) Microbial food webs in marine sediments. I. Trophic interactions and grazing rates in two tidal flat communities. Microb Ecol 34: 188–198 Epstein SS (1997b) Microbial food webs in marine sediments. II. Seasonal changes in trophic interactions in a sandy tidal flat community. Microb Ecol 34: 199–209 Epstein SS, Gallagher ED (1992) Evidence for facilitation and inhibition of ciliate population growth by meiofauna and macrofauna on a temperate zone sandflat. J Exp Mar Biol Ecol 155: 27–39 Epstein SS, Rossel J (1995a) Enumeration of sandy sediment bacteria: search for optimal protocol. Mar Ecol Prog Ser 117: 289–298 Epstein SS, Rossel J (1995b) Methodology of in situ grazing experiments: evaluation of a new vital dye for preparation of fluorescently labeled bacteria. Mar Ecol Prog Ser 128: 143–150 Epstein SS, Shiaris MP (1992) Rates of microbenthic and meiobenthic bacterivory in a temperature muddy tidal flat community. Appl Environ Microbiol 58: 2426–2431 Epstein SS, Burkovsky IV, Shiaris MP (1992) Ciliate grazing on bacteria, flagellates, and microalgae in a temperate zone sandy tidal flat: ingestion rates and food niche partitioning. J Exp Mar Biol Ecol 165: 103–123 Epstein SS, Alexander D, Cosman K, Dompé A, Gallagher S, Jarsobski J, Laning E, Martinez R, Panasik G, Peluso C, Runde R, Timmer E (1997) Enumeration of sandy sediment bacteria: are the counts quantitative or relative? Mar Ecol Prog Ser 151: 11–16
References
443
Erséus C (1980) Specific and generic criteria in marine Oligochaeta, with special emphasis on Tubificidae. In: Brinkhurst RO, Cook DG (eds) Aquatic oligochaete biology. Plenum, London, pp. 9–24 Erséus C (1984) Taxonomy and phylogeny of the gutless Phallodrilinae (Oligochaeta, Tubificidae), with descriptions of one new genus and twenty-two new species. Zool Scr 13: 239–272 Erséus C (1987) Phylogenetic analysis of the aquatic Oligochaeta under the principle of parsimony. Hydrobiologia 155: 75–89 Erséus C (1990a) Cladistic analysis of the subfamilies within the Tubificidae (Oligochaeta). Zool Scr 19: 57–63 Erséus C (1990b) The marine Tubificidae (Oligochaeta) of the barrier reef ecosystems at Carrie Bow Cay, Belize, and other parts of the Caribbean Sea, with descriptions of twenty-seven new species and revision of Heterodrilus, Thalassodrilides and Smithsonidrilus. Zool Scr 19: 243–303 Erséus C (2005) Phylogeny of oligochaetous Clitellata. Hydrobiologia 535/536: 357–372 Erséus C, Gustavsson L (2002) A proposal to regard the former family Naididae as a subfamily within Tubificidae (Annelida, Clitellata). Hydrobiologia 485: 253–256 Erséus C, Källersjö M (2004) 18S rDNA phylogeny of Clitellata (Annelida). Zool Scr 33: 187–196 Erséus C, Wetzel M, Gustavsson L (2008) ICZN rules - a farewell to Tubificidae (Annelida, Clitellata). Zootaxa 1744: 66–68 Escaravage V, Garcia ME, Castel J (1989) The distribution of meiofauna and its contribution to detritic pathways in tidal flats (Arcachon Bay, France). Sci Mar 53: 551–559 Eskin RA, Coull BC (1984) A priori determination of valid control sites: an example using marine meiobenthic nematodes. Mar Environ Res 12: 161–172. Essink K, Romeyn K (1994) Estuarine nematodes as indicators of organic pollution: an example from the Ems estuary (The Netherlands). Neth J Aquat Ecol 28: 213–219 Esteves AM, Absalao RS, Da Silva VMAP (1997) The importance of cost-effectiveness sampling in the study of intertidal sandy beach meiofauna. Trop Ecol 38: 47–53 Esteves AM, Da Silva VMAP (1998) The behavior of sugar flotation technique in meiofauna extraction form different samples. Trop Ecol 39: 283–284 Eyualem-Abebe E, Andrássy I, Traunspurger W (2006) Freshwater nematodes. Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, U.K., p. 752 Eyualem-Abebe E, Decraemer W, De Ley P (2008) Global diversity of nematodes (Nematoda) in freshwater. Hydrobiologia 595: 67–78 Farke H, Riemann F (1980) Dissolved organic carbon in littoral sediments: concentrations and available amounts demonstrated by the percolation method. Veröff Inst Meeresforsch Bremerhaven 18: 235–244 Faubel A (1976) Populationsdynamik und Lebenszyklen interstitieller Acoela und Macrostomida (Turbellaria). Mikrofauna Meeresboden 56: 1–107 Faubel A (1982) Determination of individual meiofauna dry weight values in relation to definite size classes. Cah Biol Mar 23: 339–345 Faubel A (1984) On the abundance and activity pattern of zoobenthos inhabiting a tropical reef area, Cebu, Philippines. Coral Reefs 3: 205–213 Fauchald K (1974) Polychaete phylogeny: a problem in protostome evolution. Syst Zool 23: 493–506 Fauré-Fremiet E (1950) Écologie des Ciliés psammophiles littoraux. Bull Biol Fr Belg 84: 35–75 Faust MA, Gulledge RA (1996) Associations of microalgae and meiofauna in floating detritus at a mangrove island, Twin Cays, Belize. J Exp Mar Biol Ecol 197: 159–175 Fegley SR (1987) Experimental variation of near-bottom current speeds and its effects on depth distribution of sand-living meiofauna. Mar Biol 95: 183–192 Fegley SR (1988) Comparison of meiofaunal settlement onto the sediment surface and recolonization of defaunated sandy sediment. J Exp Mar Biol Ecol 123: 97–113 Feller RJ (1982) Antigenic similarities among estuarine soft-bottom benthic taxa. Oecologia (Berlin) 52: 305–310
444
References
Feller RJ (2006) Weak meiofaunal trophic linkages in Crangon crangon and Carcinus maenas. J Exp Mar Biol Ecol 330: 274–283 Feller RJ, Coull BC (1995) Non-selective ingestion of meiobenthos by juvenile spot (Leiostomus xanthurus) (Pisces) and their daily ration. Vie Milieu 45: 49–59 Feller RJ, Warwick RM (1988) Energetics. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 181–196 Feller RJ, Taghon GL, Gallagher ED, Kenny GE, Jumars PA (1979) Immunological methods for food web analysis in a soft-bottom benthic community. Mar Biol 54: 61–74 Feller RJ, Hentschel BT, Ferguson RB (1990) Immunoelectrophoretic assay of mixed species meals: an example using penaeid shrimp. Trophic relat mar environ 588–596 Fenchel T (1967) The ecology of marine microbenthos. I. The quantitative importance of ciliates as compared with Metazoans in various types of sediments. Ophelia 4: 121–137 Fenchel T (1968a) The ecology of marine microbenthos. II. The food of marine benthic ciliates. Ophelia 5: 73–121 Fenchel T (1968b) The ecology of marine microbenthos. III. The reproductive potential of ciliates. Ophelia 5: 123–136 Fenchel T (1969) The ecology of marine microbenthos. IV. Structure and function of the benthic ecosystem, its chemical and physical factors and the microfauna communities with special reference to the ciliate Protozoa. Ophelia 6: 1–182 Fenchel T (1970) Studies on the decomposition of organic detritus derived from the turtle grass Thalassia testudinum. Limnol Oceanogr 15: 14–20 Fenchel T (1978) The ecology of micro- and meiobenthos. Ann Rev Ecol Syst 9: 99–121 Fenchel T (1993) There are more small than large species? Oikos 68: 375–378 Fenchel T (1996) Worm burrows and oxic microniches in marine sediments. 1. Spatial and temporal scales. Mar Biol 127: 289–295 Fenchel T, Finlay BJ (1989) Kentrophoros: a mouthless ciliate with a symbiotic kitchen garden. Ophelia 30: 75–93 Fenchel T, Finlay BJ (1991) The biology of free-living anaerobic ciliates. Europ J Protistol 26: 201–215 Fenchel T, Finlay BJ (1995) Ecology and evolution in anoxic worlds. Oxford University Press, Oxford, p. 276 Fenchel T, Finlay BJ (2004) The ubiquity of small species: patterns of local and global diversity. Bioscience 54: 777–784 Fenchel T, Riedl RJ (1970) The sulfide system: a new biotic community underneath the oxidized layer of marine sand bottoms. Mar Biol 7: 255–268 Fenchel T, Straarup BJ (1971) Vertical distribution of photosynthetic pigments and the penetration of light in marine sediments. Oikos 22: 172–182 Fenchel T, Perry T, Thane A (1977) Anaerobiosis and symbiosis with bacteria in free-living ciliates. J Protozool 24: 154–163 Ferguson JC (1982) A comparative study of the net metabolic benefits derived from the uptake and release of free amino acids by marine invertebrates. Biol Bull 162: 1–17 Ferrari FD, Dahms H-U (2007) Post-embryonic development of the Copepoda. Crustaceana Monographs 8: p. 256 Ferris VR, Ferris JM (1979) Thread Worms (Nematoda). In: Hart CW, Fuller SL (ed) Pollution ecology of estuarine invertebrates. Academic, New York, pp. 1–33 Fichet D, Boucher G, Radenac G, Miramand P (1999) Concentration and mobilisation of Cd, Cu, Pb and Zn by meiofauna populations living in harbour sediment: their role in the heavy metal flux from sediment to food web. Sci Total Environ 243: 263–272 Field JG, Clarke KR, Warwick RM (1982) A practical strategy for analysing multispecies distribution patterns. Mar Ecol Prog Ser 8: 37–52 Findlay RH, King GM, Watling L (1989) Efficacy of phospholipid analysis in determining microbial biomass in sediments. Appl Environ Microbiol 55: 2888–2893 Findlay SEG (1981) Small-scale spatial distribution of meiofauna on a mud and sandflat. Estuar Coast Shelf Sci 12: 471–484
References
445
Findlay SEG (1982) Influence of sampling scale on apparent distribution of meiofauna on a sandflat. Estuaries 5: 322–324 Findlay SEG, Tenore KR (1982) Effect of a free-living marine nematode (Diplolaimella chitwoodi) on detrital carbon mineralization. Mar Ecol Prog Ser 8: 161–166 Finlay BJ, Corliss JO, Esteban G, Fenchel T (1996) Biodiversity at the microbial level: the number of free-living ciliates in the biosphere. Quart Rev Biol 71: 221–237 Finlay BJ, Esteban GF, Fenchel T (1996b) Global diversity and body size. Nature 383: 132–133 Finlay BJ, Span ASW, Harman JMP (1983) Nitrate respiration in primitive eukaryotes. Nature 303: 333–336 Fisher CR (1990) Chemoautotrophic and methanotrophic symbioses in marine invertebrates. Rev Aquat Sci 2: 399–436 Flach P, Lenz A, Ozorio C (2007) Spatial distribution of tardigrades in the littoral zone of Pedreira Beach, Guaiba Lake, Brazil. In: Santos PJP dos (ed) Abstr 13th Int Meiof Conf, Recife Brazil, p. 39 Fleeger J, Yund PO, Sun B (1995) Active and passive processes associated with initial settlement and post-settlement dispersal of suspended meiobenthic copepods. J Mar Res 53: 609–645 Fleeger JW (1979) Population dynamics of three estuarine meiobenthic harpacticoids (Copepoda) from South Carolina. Mar Biol 52: 147–156 Fleeger JW, Chandler GT (1983) Meiofauna responses to an experimental oil spill in a Louisiana salt marsh. Mar Ecol Prog Ser 11: 257–264. Fleeger JW, Decho AW (1987) Spatial variability of interstitial meiofauna. Stygologia 3: 35–54 Fleeger JW, Palmer MA (1982) Secondary production of the estuarine, meiobenthic copepod Microarthridion littorale. Mar Ecol Prog Ser 7: 157–162 Fleeger JW, Palmer MA, Moser EB (1990) On the scale of aggregation of meiobenthic copepods on a tidal mudflat. Mar Ecol PSZN 11: 227–237 Fleeger JW, Shirley TC, Ziemann DA (1989) Meiofaunal responses to sedimentation from an Alaskan spring bloom. I. Major taxa. Mar Ecol Prog Ser 57: 137–145 Fleeger JW, Shirley TC, Carls MG, Todaro MA (1996) Meiofaunal recolonization experiment with oiled sediments. Amer Fisher Soc Symp 18: 271–285 Fleeger JW, Gust KA, Marlborough SJ, Tita G (2007) Mixtures of metals and polynuclear aromatic hydrocarbons elicit complex, nonadditive toxicological interactions in meiobenthic copepods. Environ Toxicol Chem 26: 1677–1685 Flemming BW, Delafontaine MT (2000) Mass physical properties of muddy intertidal sediments: some applications, misapplications and non-applications. Contin Shelf Res 20: 1179–1197 Flint HC, Copley JTP, Ferrero TJ, VanDover CL (2006) Patterns of nematode diversity at hydrothermal vents on the East Pacific Rise. Cah Biol Mar 47: 365–370 Fonseca G, Soltwedel T (2007) Deep-sea meiobenthic communities underneath the marginal ice zone off Eastern Greenland. Polar Biol 30: 607–618 Fontaneto D, De Smet WH, Melone G (2008) Identification key to the genera of marine rotifers worldwide. Meiofauna Mar 16: 75–99 Forster S (1996) Spatial and temporal distribution of oxidation events occurring below the sediment-water interface. Mar Ecol PSZN 17: 309–319 Forster S, Graf G (1992) Continuously measured changes in redox potential influenced by oxygen penetrating from burrows of Callianassa subterranea. Hydrobiologia 235236: 527–532 Forster S, Huettel M, Ziebis W (1996) Impact of boundary layer flow velocity on oxygen utilisation in coastal sediments. Mar Ecol Prog Ser 143: 173–185 Forster SJ (1998) Osmotic stress tolerance and osmoregulation of intertidal and subtidal nematodes. J Exp Mar Biol Ecol 224: 109–125 Fortey RA, Briggs DEG, Wills MA (1996) The Cambrian evolutionary “explosion”: Decoupling cladogenesis from morphological disparity. Biol J Linn Soc 57: 13–33 Fossing H, Jørgensen BB (1990) Oxidation and reduction of radiolabeled inorganic sulfur compounds in an estuarine sediment, Kysing Fjord, Denmark. Geochim Cosmochim Acta 54: 2731–2742 Fox CA, Powell EN (1986) Meiofauna and the sulfide system: the effects of oxygen and sulfide on the adenylate pool of three turbellarians and a gastrotrich. Comp Biochem Physiol A 85: 37–44
446
References
Fox CA, Powell EN (1987) The effect of oxygen and sulfide on CO2 production by three acoel turbellarians. Are thiobiotic meiofauna aerobic? Comp Biochem Physiol A 86: 509–514 Foy MS, Thistle D (1991) On the vertical distribution of a benthic harpacticoid copepod: field, laboratory, and flume results. J Exp Mar Biol Ecol 153: 153–164 Frame K, Hunt G, Roy K (2007) Intertidal meiofaunal biodiversity with respect to different algal habitats: a test using phytal ostracodes from Southern California. Hydrobiologia 586: 331–342 Franco MA, Soetaert K, Oevelen D, Gansbeke D, Costa M, Vincx M, Vanaverbeke J (2008) Density, vertical distribution and trophic responses of metazoan meiobenthos to phytoplankton deposition in contrasting sediment types. Mar Ecol Prog Ser 358: 51–62 Franzén Å (1960) Monobryozoon limicola n.sp.: a Ctenostomatous bryozoan from the detritus layer on soft sediments. Zool Bidr Uppsala 33: 135–148 Frey DG (1987) The taxonomy and biogeography of the Cladocera. In: Forró L, Frey DG (eds) Proceedings of Cladocera Symposium, Budapest 1985, pp. 5–17 Fricke H, Giere O, Stetter K, Alfredsson GA, Kristjansson JK, Stoffers P, Svavarsson J (1989) Hydrothermal vent communities at the shallow subpolar Mid-Atlantic Ridge. Mar Biol 102: 425–429 Friedrich C (1997) Ecological investigations on the fauna of the Arctic sea-ice. Ber Polarforsch/ Rep Polar Res 246: 1–211 Friedrich C, Gradinger R, Spindler M (1996) The sea-ice biota of the Greenland Sea pack ice. In: Wadhams P, Wilkinson JP, Wells SCS (eds) European subpolar ocean programme, Brussels, pp. 520–534 Frithsen JB, Rudnick DT, Elmgren R (1983) A new, flow-through corer for quantitative sampling of surface sediments. Hydrobiologia 99: 75–79 Fujiwara M, Highsmith RC (1997) Harpacticoid copepods: potential link between inbound salmon and outbound juvenile salmon. Mar Ecol Prog Ser 158: 205–216 Fukui M, Takii S (1990) Survival of sulfate-reducing bacteria in oxic surface sediment of a seawater lake. FEMS Microbiol Ecol 73 (4): 317–322 Funch P, Kristensen RM (2002) The Micrognathozoa—a new class or phylum of freshwater meiofauna? In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys, Leiden, The Netherlands, pp. 337–348 Furstenberg JP, Wet AG (1982) A comparison of two extractors for separating meiobenthic nematodes from fine-grained sediments. S Afr J Zool 17: 41–43 Gabel B (1971) Die Foraminiferen der Nordsee. Helgol wiss Meeresunters 22: 1–65 Gabrich A, Jaros PP, Brockmeyer V (1991) Application of immunological methods for the taxonomic study of two selected animal taxa: Tisbe (Crustacea, Copepoda) and Enchytraeus (Annelida, Oligochaeta). Z zool Syst Evolutforsch 29: 381–392 Gad G (2004) A new genus of Nanaloricidae (Loricifera) from deep-sea sediments of volcanic orgin in the Kilinailau Trench north of Papua New Guinea. Helgol Mar Res 58: 50–53 Gad G (2005) Giant Higgins-larvae with paedogenetic reproduction from the deep sea of the Angola Basin—evidence fo a new life cycle and for abyssal gigantism in Loricifera?. Organ Divers Evolut 5: 59–75 Gage JD, Tyler PA (1991) Deep-sea biology: a natural history of organisms at the deep-sea floor. Cambridge University Press, Cambridge, UK, p. 504 Galassi D, Marmonier P, Dole-Olivier M-J, Rundle S (2002) Microcrustacea. In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys, Leiden, The Netherlands, pp. 135–175 Galhano MH (1970) Contribuição para o conhecimentoda fauna intersticial em Portugal. Publ Inst Zool «Dr Augusto Nobre», Porto 110: 206 Gal’tsova VV (1991) Meiobenthos in marine ecosystems (with special reference to freeliving nematodes). Proc Zool Inst, Leningrad/Trudy Zool Inst Leningrad 224: 1–241 Gal’tsova VV, Platonova TA (1980) Distribution of meiofauna from a muddy-sand beach of Dalnezelenetsky Bay, Barents Sea (in Russian). Biol Mora 2: 15–20 Gambi C et al. (2007) Unravelling patterns in deep-sea Nematoda communities. In: Santos PJP dos (ed) Abstr 13th Int Meiof Conf, Recife Brazil, p O 20 Gamenick I, Giere O (1994) The microdistribution of coral sand meiofauna affected by water currents. Vie Milieu 44: 93–100
References
447
Gardner WS, Hanson RB (1979) Dissolved free amino acids in interstitial waters of Georgia salt marsh soils. Estuaries 2: 113–118 Garey JR (2001) Ecdysozoa: the relationship between Cycloneuralia and Panarthropoda. Zool Anz 240: 321–330 Garey JR (2002) The lesser-known protostome taxa: an introduction and a tribute to Robert P. Higgins. Integr Comp Biol 42: 611–618 Gasol JM (1993) Benthic flagellates and ciliates in fine freshwater sediments: calibration of a live counting procedure and estimation of their abundances. Microb Ecol 25: 247–262 Gaston GR (1992) Green-winged teal ingest epibenthic meiofauna. Estuaries 15: 227–229 Gaston KJ (2000) Global patterns in biodiversity. Nature 405: 220–227 Gaudy R, Guérin JP (1977) Dynamique des populations de Tisbe holothuriae (Crustacea: Copepoda) en élevage sur trois régimes artificiels différents. Mar Biol 39: 137–145 Gee JM (1987) Impact of epibenthic predation on estuarine intertidal harpacticoid copepod populations. Mar Biol 96 (4): 497–510 Gee JM (1989) An ecological and economic review of meiofauna as food for fish. Zool J Linn Soc 93 (3): 243–261 Gee JM, Warwick RM (1994) Body-size distribution in a marine metazoan community and the fractal dimensions of macroalgae. J Exp Mar Biol Ecol 178 (2): 247–259 George KH, Schminke HK (2002) Harpacticoida (Crustacea, Copepoda) of the Great Meteor Seamount, with first conclusions as to the origin of the plateau fauna. Mar Biol 141: 887–895 Gerbersdorf S, Black HJ, Meyercordt J, Meyer-Reil L-A, Rieling T, Stodian I (1999) Significance of microphytobenthic primary production in the “Bodden” of the southern Baltic Sea. In: Flemming BW, Delafontaine MT, Liebezeit G (eds) Muddy coasts—processes and products. Elsevier, Amsterdam, pp. 127–136 Gerlach SA (1954) Das Supralitoral der sandigen Meeresküsten als Lebensraum einer Mikrofauna. Kiel Meeresforsch 10: 121–129 Gerlach SA (1971) On the importance of marine meiofauna for benthos communities. Oecologia (Berl) 6: 176–190 Gerlach SA (1977a) Attraction to decaying organisms as a possible cause for patchy distribution of nematodes in a Bermuda beach. Ophelia 16: 151–165 Gerlach SA (1977b) Means of meiofauna dispersal. Mikrofauna Meeresboden 61: 89–103 Gerlach SA (1978) Food-chain relationships in subtidal silty-sand marine sediments and the role of meiofauna in stimulating bacterial growth. Oecologia (Berl) 33: 55–69 Gerlach SA, Schrage M (1971) Life cycles in marine meiobenthos. Experiments at various temperatures with Monhystera disjuncta and Theristus pertenuis (Nematoda). Mar Biol 9: 272–280 Gerlach SA, Hahn AE, Schrage M (1985) Size spectra of benthic biomass and metabolism. Mar Ecol Prog Ser 26: 161–173 Gerner L (1969) Nemertinen der Gattungen Cephalothrix und Ototyphlonemertes aus dem marinen Mesopsammal. Helgol wiss Meeresunters 19: 68–110 Gheller PF, Corbisier TN (2007) Antarctic nematodes as indicators of anthropogenic impact near the Brazilian station. In: Santos PJP dos (ed) Abstr 13th Int Meiof Conf, Recife Brazil, p. O 27 Gheskiere T, Vincx M, Urban-Malinga B, Rossano C, Scapini F, Degraer S (2005) Nematodes from wave-dominated sandy beaches: diversity, zonation patterns and testing of the isocommunities concept. Estuar Coast Shelf Sci 62: 365–375 Giard A (1904) Sur une faunule charactéristique des sables à diatomées d’Ambleteuse. C R Séanc Soc Biol Paris 56: 107–165 Gibbons MJ (1991) Rocky shore meiofauna: a brief overview. Trans R Soc S Afr 47: 595–603 Gibbons MJ, Griffiths CL (1988) An improved quantitative method for estimating intertidal meiofaunal standing stock on an exposed rocky shore. S Afr J Mar Sci 6: 55–58 Gibbs PE (1985) On the genus Phascolion (Sipuncula) with particular reference to the North-East Atlantic species. J Mar Biol Ass UK 65: 311–323 Gibert J, Deharveng L (2002) Subterranean ecosystems: a truncated functional biodiversity. Bioscience 52: 473–481 Gibert J, Danielopol DL, Stanford JA (1994) Groundwater ecology. In: Thorp JH (ed) Aquatic biology series. Academic Press, San Diego, CA, p. 571
448
References
Gibson GR, Parkes RJ, Herbert RA (1989) Biological availability and turnover of acetate in marine and estuarine sediments in relation to dissimilitory sulphate reduction. FEMS Microb Ecol 62: 303–306 Giere O (1973) Oxygen in the marine hygropsammal and the vertical microdistribution of oligochaetes. Mar Biol 21: 180–189 Giere O (1975) Population structure, food relations and ecological role of marine oligochaetes. With special reference to meiobenthic species. Mar Biol 31: 139–156 Giere O (1980) Tolerance and preference reactions of marine Oligochaeta in relation to their distribution. In: Brinkhurst RO, Cook DG (eds) Aquatic oligochaete biology Plenum, London, pp. 385–409 Giere O (1981) The gutless marine oligochaete Phallodrilus leukodermatus. Structural studies on an aberrant tubificid associated with bacteria. Mar Ecol Prog Ser 5: 353–357 Giere O (1992) Benthic life in sulfidic zones of the sea—ecological and structural adaptations to a toxic environment. Verh Dtsch Zool Ges 85, 2: 77–93 Giere O (1996) Bacterial endosymbioses in marine littoral worms. In: Uiblein F, Ott J, Stachowitsch M (eds) Deep sea and extreme shallow-water habitats: affinities and adaptations. 1, Austrian Academy of Sciences Press, Vienna, pp. 353–367 Giere O (2006) Ecology and biology of marine oligochaeta—an inventory rather than another review. Hydrobiologia 564: 103–116 Giere O, Hauschildt D (1979) Experimental studies on the life cycle and production of the littoral oligochaete Lumbricillus lineatus, and its response to oil pollution. In: Naylor E, Hartnoll RG (eds) Cyclic phenomena in marine plants and animals Pergamon, Oxford, pp. 113–122 Giere O, Langheld C (1987) Structural organization, transfer and biological fate of endosymbiotic bacteria in gutless oligochaetes. Mar Biol 93: 641–650 Giere O, Pfannkuche O (1982) Biology and ecology of marine Oligochaeta, a review. Oceanogr Mar Biol Ann Rev 20: 173–308. Giere O, Pfannkuche O (1978) An ecophysiological approach to the microdistribution of meiobenthic Oligochaeta. II. Phallodrilus monospermathecus (Tubificidae) from boreal brackish-water shores in comparison to populations from subtropical beaches. Kieler Meeresforsch Sdh. 4: 289–301 Giere O, Welberts H (1985) An artificial “sand system” and its application for studies on interstitial fauna. J Exp Mar Biol Ecol 88: 83–89 Giere O, Liebezeit G, Dawson R (1982) Habitat conditions and distribution pattern of the gutless oligochaete Phallodrilus leukodermatus. Mar Ecol Prog Ser 8: 291–299 Giere O, Eleftheriou A, Murison DJ (1988) Abiotic factors. In: Higgins RPT, Thiel H (eds) Introduction to the study of meiofauna Smithsonian Institution Press, Washington, DC, pp. 61–78 Giere O, Wirsen CO, Schmidt C, Jannasch HW (1988a) Contrasting effects of sulfide and thiosulfide on symbiotic CO2-assimilation of Phallodrilus leukodermatus (Annelida). Mar Biol 97: 413–419 Giere O, Rhode B, Dubilier N (1988b) Structural peculiarities of the body wall of Tubificoides benedii (Oligochaeta) and possible relations to its life in sulphidic sediments. Zoomorphology 108: 29–39 Giere O, Conway NM, Gastrock G, Schmidt C (1991) ‘Regulation’ of gutless annelid ecology by endosymbiotic bacteria. Mar Ecol Prog Ser 68: 287–299 Giere O, Preusse J-H, Dubilier N (1999) Tubificoides benedii (Tubificidae, Oligochaeta)—a pioneer in hypoxic and sulfidic environments. An overview of adaptive pathways. Hydrobiologia 406: 235–241 Gilboa-Garber N (1971) Direct spectrophotometric determination of inorganic sulfide in biological materials and in other complex mixtures. Analyt Biochem 43: 129–133 Gilmour THJ (1989) A method for studying the hydrodynamics of microscopic animals. J Exp Mar Biol Ecol 133: 189–193 Giribet G (2003) Molecules, development and fossils in the study of metazoan evolution: articulata versus ecdysozoa revisited. Zoology 106: 303–326
References
449
Giribet G, Richter S, Gd E, Wheeler WC (2005) The position of crustaceans within Arthropoda— evidence from nine molecular loci and morphology. Crustac Issues 16: 307–352 Glatzel T, Königshoff D (2005) Cross-breeding experiments among different populations of the cosmopolitan species Phyllognathopus viguieri (Copepoda: Harpacticoida). Hydrobiologia 534: 141–149 Glud RN, Ramsing NB, Gundersen JK, Klimant I (1996) Planar optrodes: a new tool for fine scale measurements of two-dimensional O2 distribution in benthic communities. Mar Ecol Prog Ser 140: 217–226 Glud RN, Tengberg A, Kühl M, Hall POJ, Klimant I (2001) An in situ instrument for planar O2 optode measurements at benthic interfaces. Limnol Oceanogr 46: 2973–2080 Gnaiger E (1983) The Twin-Flow micro respirometer and simultaneous calorimetry. In: Gnaiger E, Forstner H. (eds) Polarographic oxygen sensors. Aquatic and physiological applications Springer, Berlin Heidelberg New York, pp. 134–166 Gnaiger E (1991) Animal energetics at very low oxygen: information from calorimetry and respirometry. In: Woakes AJ, Grieshaber MK, Bridges CR (eds) Physiological strategies for gas exchange and metabolism. Cambridge University Press, Cambridge, pp. 149–171 Gobin JF, Warwick RM (2006) Geographical variation in species diversity: a comparison of marine polychaetes and nematodes. J Exp Mar Biol Ecol 330: 234–244 Goerke H, Ernst W (1975) ATP content of estuarine nematodes: a contribution to the determination of meiofauna biomass by ATP measurements. Proceedings of the 9th European Marine Biology Symposium, Oban, Scotland, 1974. Aberdeen University Press, Aberdeen, pp. 683–691 Goldstein ST, Corliss BH (1994) Deposit feeding in selected deep-sea and shallow-water benthic Foraminifera. Deep-Sea Res I 41: 229–241 Golemansky V (1976) Contribution to the study of the Rhizopoda and Heliozoa of the supralittoral psammon of the Mediterranean. Acta Protozoologica 15: 35–45 Golemansky V (1978) Adaptations morphologiques de thécamoebiens psammobiontes du psammal supralittoral des mers. Acta Protozool 17: 141–152 Golemansky V (1986) Rhizopoda: Testacea. In: Botosaneanu L (ed) Stygofauna mundi. A faunistic, distributional, and ecological synthesis of the world fauna inhabiting subterranean waters (including the marine interstitial). Brill/Backhuys, Leiden, The Netherlands, pp. 5–9 Golemansky V (1994) On some ecological preferences of marine interstitial testate amoebas. Arch Protistenkunde 144: 424–432 Golemansky V, Todorov M (2004) Shell morphology, biometry and distribution of some marine interstitial testate amoebae (Sarcodina: Rhizopoda). Acta Protozoologica 43: 147–162 Gollner S, Zekely J, Van Dover CL, Govenar B, Le Bris N, Nemeschkal H, Bright M (2006) Benthic copepod communities associated with tubeworm and mussel aggregations on the East Pacific Rise. Cah Biol Mar 47: 397–402 Gollner S, Zekely J, Govenar B, Le Bris N, Nemeschkal HL, Fisher CR, Bright M (2007) Tubeworm-associated permanent meiobenthic communities from two chemically different hydrothermal vent sites on the East Pacific Rise. Mar Ecol Prog Ser 337: 39–49 Gomme J (1982) Epidermal nutrient absorption in marine invertebrates: a comparative analysis. Am Zool 22: 691–708 Gooday AJ (1986) Meiofaunal foraminiferans from the bathyal Porcupine Seabight (northern Atlantic): size structure, standing stock, taxonomic composition, species diversity and vertical distribution in the sediment. Deep-Sea Res 33: 1345–1373. Gooday AJ (2002) Biological responses to seasonally varying fluxes of organic matter to the ocean floor: a review. J Oceanogr 58: 305–332 Gooday AJ, Turley CM (1990) Responses by benthic organisms to inputs of organic material to the ocean floor: a review. Phil Trans R Soc London A 331: 119–138 Gooday AJ, Levin LA, Linke P, Heeger T (1992) The role of benthic Foraminifera in deep-sea food webs and carbon cycling. In: Rowe GT, Pariente V (eds) Deep-sea food chains and the global carbon cycle. Kluwer, Dordrecht, pp. 63–91 Gooday AJ, Bowser SS, Bernhard JM (1994) The Foraminifera of explorer’s cove, Antarctica: a deep-sea assemblage in shallow water?. Antarct J US 29: 149–151
450
References
Gooday AJ, Bernhard JM, Levin LA, Suhr SB (2000) Foraminifera in the Arabian Sea oxygen minimum zone and other oxygen-deficient settings: taxonomic composition, diversity, and relation to metazoan faunas. Deep-Sea Res II 47: 25–54 Gooday AJ, Cedhagen T, Kamenskaya OE, Cornelius N (2007) The biodiversity and biogeography of komokiaceans and other enigmatic foraminiferan-like protists in the deep Southern Ocean. Deep-Sea Res II 54: 1691–1719 Goodman D, Parrish WB (1971) Ultrastructure of the epidermis in the ice worm, Mesenchytraeus solifugus. J Morph 135: 71–86 Goulden CE (1971) Environmental control of the abundance and distribution of the chydorid Cladocera. Limnol Oceanogr 16: 320–331 Gourbault N (1987) Long-term monitoring of marine nematode assemblages in the Morlaix estuary (France) following the Amoco Cadiz oil spill. Estuar Coast Shelf Sci 24: 657–670 Gourbault N, De Craemer W (1996) Marine nematodes of the family Epsilonematidae: a synthesis with phylogenetic relationships. Nematologica 42: 133–158 Gourbault N, Renaud-Mornant J (1990) Micro-meiofaunal community structure and nematode diversity in a lagoonal ecosystem (Fangataufa, eastern Tuamotu Archipelago. Mar Ecol PSZN 11: 173–189 Gourbault N, Warwick RM (1994) Is the determination of meiobenthic diversity affected by the sampling method in sandy beaches? Mar Ecol PSZN 15: 267–279 Gourbault N, Warwick RM, Helleouet MN (1998) Spatial and temporal variability in the composition and structure of meiobenthic assemblages (especially nematodes) in tropical beaches (Gouadeloupe, FWI). Cah Biol Mar 39: 29–39 Gowing MM, Silver MW (1983) Origins and microenvironments of bacteria mediating fecal pellets decomposition in the sea. Mar Biol 73: 7–16 Gradinger R (1999) Integrated abundance and biomass of sympagic meiofauna in Arctic and Antarctic pack ice. Polar Biol 22: 169–177 Gradinger R (2001) Adaptation of Arctic and Antarctic ice metazoa to their habitat. Zoology 104: 339–345 Gradinger R, Spindler M, Weissenberger J (1992) On the structure and development of Arctic pack ice communities in Fram Strait: a multivariate approach. Polar Biol 12: 727–733 Gradinger R, Friedrich C, Spindler M (1999) Abundance, biomass and composition of the sea ice biota of the Greenland Sea pack ice. Deep-Sea Res II 46: 1457–1472 Gradinger R, Meiners K, Plumley G, Zhang Q, Bluhm BA (2005) Abundance and composition of the sea-ice meiofauna in off-shore pack ice of the Beaufort Gyre in summer 2002 and 2003. Polar Biol 28: 171–181 Graf G, Bengtson W, Diesner U, Schulz R, Theede H (1982) Benthic response to sedimentation of a spring phytoplankton bloom: process and budget. Mar Biol 67: 201–208 Graf G, Rosenberg R (1997) Bioresuspension and biodeposition: a review. J Mar Systems 11: 269–278 Grainger EH (1991) Exploitation of arctic sea ice by epibenthic copepods. Mar Ecol Prog Ser 77: 119–124 Grainger EH, Mohammed A, Lovrity J (1985) The sea ice fauna of Frobisher Bay, Arctic Canada. Arctic 38: 23–30 Grassle JF, Elmgren R, Grassle JP (1980–81) Response of benthic communities in MERL experimental ecosystems to low level, chronic additions of No. 2 fuel oil. Mar Environ Res 4: 279–297 Gray JS (1965) Selection of sands by Protodrilus symbioticus Giard. Veröff Inst Meeresforsch Bremerh Sbd. 2: 105–116 Gray JS (1966a) The attractive factors of intertidal sands to Protodrilus symbioticus Giard. J mar Biol Ass U K 46: 627–645 Gray JS (1966b) Factors controlling the localizations of populations of Protodrilus symbioticus Giard. J Anim Ecol 35: 435–442 Gray JS (1971) Factors controlling population localization in Polychaete worms. Vie Milieu 22: 707–722
References
451
Gray JS (1974) Animal–sediment relationships. Oceanogr Mar Biol Ann Rev 12: 223–261 Gray JS (1978) The structure of meiofauna communities. Sarsia 64: 265–272 Gray JS (1981) The ecology of marine sediments. An introduction to the structure and function of benthic communities. Cambridge studies in modern biology: 2. Cambridge University Press, Cambridge, p. 185 Gray JS (2000) The measurement of marine species diversity, with an application to the benthic fauna of the Norwegian continental shelf. J Exp Mar Biol Ecol 250: 23–49 Gray JS, Clarke KR, Warwick RM, Hobbs G (1990) Detection of initial effects of pollution on marine benthos: an example from the Ekofisk and Eldfisk oilfields, North Sea. Mar Ecol Prog Ser 66: 285–299 Green AS, Chandler GT (1994) Meiofaunal bioturbation effects on the partitioning of sedimentassociated cadmium. J Exp Mar Biol Ecol 180: 59–70 Green AS, Chandler GT (1996) Life-table evaluation of sediment-associated Chlopyrifos chronic toxicity to the benthic copepod, Amphiascus tenuiremis. Arch Environ Contam Toxicol 31: 77–83 Green AS, Chandler GT, Blood ER (1993) Aqueous-, pore-water-, and sediment-phase cadmium: toxicity relationships for a meiobenthic copepod. Environ Toxicol Chem 12: 1497–1506 Green AS, Chandler GT, Piegorsch WW (1996) Life-stage-specific toxicity of sediment-associated Chlorpyrifos to a marine, infaunal copepod. Environ Toxicol Chem 15: 1182–1188 Green J, MacQuitty M (1987) Halacarid mites. Brill/Backhuys, Leiden, p. 178 (Synopses of the British Fauna, vol 36) Green MA, Aller RC, Aller JY (1998) Influence of carbonate dissolution on survival of shell-bearing meiobenthos in nearshore sediments. Limnol Oceanogr 43: 18–28 Gregg CS, Fleeger JW (1998) Grass shrimp Palaemonetes pugio predation on sediment- and stemdwelling meiofauna: field and laboratory experiments. Mar Ecol Prog Ser 175: 77–86 Grelet Y (1985) Vertical distribution of meiobenthos and estimation of nematode biomass from sediments of the Gulf of Aqaba (Jordan, Red Sea). Proc 5th Int Coral Reef Congr, Tahiti, pp. 251–256 Grelet Y, Falconetti C, Thomassin BA, Vitiello P, Abu Hilal AH (1987) Distribution of the macroand meiobenthic assemblages in the littoral soft-bottoms of the Gulf of Aqaba (Jordan). Atoll Res Bull 308: 1–14 Grieshaber MK, Hardewig I, Kreutzer U, Pörtner H-O (1994) Physiological and metabolic responses to hypoxia in invertebrates. Rev Physiol Biochem Pharmacol 125: 43–147 Grieshaber MK, Völkel S (1998) Animal adaptations for tolerance and exploitation of poisonous sulfide. Annu Rev Physiol 60: 33–53 Griffiths HI, Butlin RK (1994) Darwinula stevensoni: a brief review of the biology of a persistent parthenogen. In: Horne DJ, Martens K (eds) The evolutionary ecology of reproductive modes in non-marine Ostracoda. Greenwich University Press:, London, pp. 27–36 Grimaldi de Zio S, D’Addabbo Gallo M (1975) Reproductive cycle of Batillipes pennaki Marcus (Heterotardigrada) and observations on the morphology of the female genital apparatus. Pubbl Staz Napoli 39, Suppl.: 212–225 Grimaldi de Zio S, D’Addabo Gallo M, De Lucia RM (1987) Adaptive radiation and phylogenesis in marine Tardigrada and the establishment of Neostygarctidae, a new family of Heterotardigrada. Boll Zool 54: 27–33 Grimaldi de Zio S, De Lucia RM, D’Addabo Gallo M (1983) Marine tardigrades ecology. Oebalia 9: 15–31 Grossmann S, Reichardt W (1991) Impact of Arenicola marina on bacteria in intertidal sediments. Mar Ecol Prog Ser 77: 85–94 Grove SL, Probert PK, Berkenbusch K, Nodder SD (2006) Distribution of bathyal meiofauna in the region of the Subtropical Front, Chatham Rise, south-west Pacific. J Exp Mar Biol Ecol 330: 342–355 Gschwentner R, Baric S, Rieger R (2002) New model for the formation and function of sagittocysts: Symsagittifera corsicae n.sp. (Acoela). Invert Biol 121: 95–103 Guarini JM, Blanchard GF, Bacher C, Gros P, Riera P, Richard P, Gouleau D, Galois R, Prou J, Sauriau PG (1998) Dynamics of spatial patterns of microphytobenthos biomass: inferences from a geostatistical analysis of two comprehensive surveys in Marennes-Oléron Bay (France). Mar Ecol Prog Ser 166: 131–141
452
References
Guerrini A, Colangelo MA, Ceccherelli VU (1998) Recolonization patterns of meiobenthic communities in brackish vegetated and unvegetated habitats after induced hypoxia/anoxia. Hydrobiologia 375/376: 73–87 Guidetti R, Bertolani R (2005) Tardigrade taxonomy: an updated check list of the taxa and a list of characters for their identification. Zootaxa 845: 1–46 Guzmán HM, Obando VL, Cortés J (1987) Meiofauna associated with a Pacific coral reef in Costa Rica. Coral Reefs 6: 107–112 Gwyther J, Fairweather PG (2002) Colonisation by epibionts and meiofauna of real and mimic pneumatophores in a cool temperate mangrove habitat. Mar Ecol Prog Ser 229: 137–149 Gwyther J, Fairweather PG (2005) Meiofaunal recruitment to mimic pneumatophores in a cool-temperate mangrove forest: spatial context and biofilm effects. J Exp Mar Biol Ecol 317: 69–85 Haarløv N, Weis-Fogh T (1953) A microscopical technique for studying the undisturbed texture of soils. Oikos 4: 44–57 Hackstein JHP, Akhmanova A, Boxma B, Harhangi HR, Voncken FGJ (1999) Hydrogenosomes: eukaryotic adaptations to anaerobic environments. Trends Microbiol 7: 441–447 Hackstein JHP et al. (2001) Hydrogenosomes: convergent adaptations to anaerobic environments. Zoology 104: 290–302 Hadl G, Kothbauer H, Peter R, Wawra E (1970) Substratwahlversuche mit Microhedyle milaschewitchii Kowalevsky (Gastropoda, Opisthobranchia: Acochlidiacea). Oecologia (Berl) 4: 74–82 Hadzi J (1956) Das Kleinsein und Kleinwerden im Tierreiche. Ein weiterer Beitrag zu meiner Turbellarientheorie der Knidarien. Proceedings of the 14th International Congress of Zoology Kopenhagen,1953: 154–158 Hagerman GM, Rieger RM (1981) Dispersal of benthic meiofauna by wave and current action in Bogue Sound, North Carolina, USA. Mar Ecol PSZN 2: 245–270 Hahn HJ (2003) Eignen sich Fallen zur repräsentativen Erfassung aquatischer Meiofauna im hyporheischen Interstitial und im Grundwasser? Are traps suitable for representative sampling of aquatic meiofauna in both the hyporheic zone and in groundwater?. Limnologica 33: 138–146 Hairston NG, Kearns CM (2002) Temporal dispersal: ecological and evolutionary aspects of zooplankton egg banks and the role of sediment mixing. Integr Comp Biol 42: 481–491 Haitzer M, Höss S, Traunspurger W, Steinberg C (1998) Effects of dissolved organic matter (DOM) on the bioconcentration of organic chemicals in aquatic organisms—a review. Chemosphere 37: 1335–1362 Håkanson L (1973) Sampling of recent sedimentary deposits: a new sampler. Naturvårdsverk Limnol Undersökn Rapp 65: 1–20 Hakenkamp CC, Morin A (2000) The importance of meiofauna to lotic ecosystem functioning. Freshw Biol 44: 165–175 Hakenkamp CC, Palmer MA (2000) The ecology of hyporheic meiofauna. In: Jones JB, Mulholland PJ (eds) Streams and ground water. Academic Press, San Diego, CA, pp. 307–336 Hakenkamp CC, Morin A, Strayer DL (2002) The functional importance of freshwater meiofauna. In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater Meiofauna: biology and ecology. Backhuys, Leiden, The Netherlands, pp. 321–355 Halanych KM, Dahlgren TG, McHugh D (2002) Unsegmented annelids? Possible origin of four lophotrochozoan worm taxa. Integ Comp Biol 42: 678–684 Hall MO, Bell SS (1988) Response of small motile epifauna to complexity of epiphytic algae on seagrass blades. J Mar Res 46: 613–630. Hall MO, Bell SS (1993) Meiofauna on the seagrass Thalassia testudinum—Population characteristics of harpactocoid copepods and associations with algal epiphytes. Mar Biol 116: 137–146 Hall SJ (1994) Physical disturbance and marine benthic communities: life in unconsolidated sediments. Oceanogr Mar Biol Annu Rev 32: 179–239 Hall SJ, Basford DJ, Robertson MR, Raffaeli DG, Tuck I (1991) Patterns of recolonization and the importance of pit-digging by the crab Cancer pagurus in a subtidal sand habitat. Mar Ecol Prog Ser 72: 93–102
References
453
Hamburger K (1981) A gradient diver for measurement of respiration in individual organisms from the microfauna and meiofauna. Mar Biol 61: 179–183 Hamels I, Moens T, Muylaert K, Vyverman W (2002) Trophic interactions between ciliates and nematodes from an intertidal flat. Aquat Microb Ecol 26: 61–72 Hamilton AL (1969) A method of separating invertebrates from sediments using longwave ultraviolet light and fluorescent dyes. J Fish Res Bd Can 26: 1667–1672 Hancock PJ (2006) The response of hyporheic invertebrate communities to a large flood in the Hunter River, New South Wales. Hydrobiologia 568: 255–262 Hancock PJ, Boulton AJ, Humphreys WF (2005) Aquifers and hyporheic zones: towards an ecological understanding of groundwater. Hydrogeol J 13: 98–111 Hannah F, Rogerson A (1997) The temporal and spatial distribution of Foraminiferans in marine benthic sediments of the Clyde Sea area, Scotland. Estuar Coast Shelf Sci 44: 377–383 Hansen JG, Jørgensen A, Kristensen RM (2001) Preliminary studies of the tardigrade fauna of the Faroe Bank. Zool Anz 240: 385–393 Hargrave BT (1972) Oxidation-reduction potentials, oxygen concentrations and oxygen uptake of profundal sediments in an eutrophic lake. Oikos 23: 167–177 Harris RP (1972) Seasonal changes in the meiofauna population of an intertidal sand beach. J Mar Biol Ass UK 52: 389–404 Hartmann G (1966–1975) Ostracoda. Akad. Verlagsges. Geest & Portig, Leipzig, and VEB Gustav Fischer, Jena, p. 786 Hartmann G (1973) Zum gegenwaertigen Stand der Erforsachung der Ostracoden interstitieller Systeme. Ann Spéléol 28: 417–426 Hartmann G (1986) Biogeographie und Plattentektonik. Gondwana und die rezente Verteilung der Organismen. Naturwissensch 73: 471–480 Hartmann G (1990) Antarktische benthische Ostracoden VI. Auswertung der Reise der “Polarstern” Ant. VI-2 (1. Teil, Meiofauna und Zehnerserien) sowie Versuch einer vorläufigen Auswertung aller bislang vorliegenden Daten. Mitt Hamb Zool Mus Inst 87: 191–245 Hartmann G, Puri HS (1974) Summary of neontological and palaeontological classification of Ostracoda. Mitt Hamb Zool Mus Inst 70: 7–73 Hartwig E (1973a) Die Ciliaten des Gezeiten-Sandstrandes der Nordseeinsel Sylt. I. Systematik. Mikrofauna Meeresboden 18: 387–453 Hartwig E (1973b) Die Ciliaten des Gezeiten-Sandstrandes der Nordseeinsel Sylt. II. Ökologie. Mikrofauna Meeresboden 21: 1–171 Hasemann C, Soltwedel T (2006) Small-scale heterogeneity in the Arctic deep sea: impact of small coldwater-sponges on the diversity of benthic nematode communities. Diss. Univ. Bremen, Rep Polar Mar Res 527: 1–294 Haszprunar G, Schäfer K (1997) Anatomy and phylogenetic significance of Micropilina arntzi (Mollusca, Monoplacophora, Micropilinidae fam. nov.). Acta Zool 77: 315–334 Haszprunar G, Salvini-Plawen LV, Rieger RM (1995) Larval planktotrophy—a primitive trait in the Bilateria?. Acta Zool 76: 141–154 Hausdorf B, Helmkampf M, Meyer A, Witek A, Herlyn H, Bruchhaus I, Hankeln T, Struck TH, Lieb B (2007) Spiralian phylogenomics supports the resurrection of Bryozoa comprising Ectoprocta and Entoprocta. Mol Biol Evol 24: 2723–2729 Healy B (1994) New species of Marionina (Annelida: Oligochaeta: Enchytraeidae) from Spartina salt marshes on Sapelo Island, Georgia, USA. Proc Biol Soc Wash 107: 164–173 Healy B, Walters K (1994) Oligochaeta in Spartina stems: the microdistribution of Enchytraeidae and Tubificidae in a salt marsh, Sapelo Island, USA. In: Reynoldson TB, Coates KA (eds) Aquatic oligochaete biology. Kluwer, Dordrecht, pp. 111–123 Hedfi A, Mahmoudi E, Boufahja F, Beyrem H, Aissa P (2007) Effects of increasing levels of nickel contamination on structure of offshore nematode communities in experimental microcosms. Bull Environ Contam Toxicol 79: 345–349 Heiner I, Neuhaus B (2007) Loricifera from the deep sea at the Galápagos spreading center, with a description of Spinoloricus turbatio gen. et sp. nov. (Nanaloricidae). Helgol Mar Res 61: 167–182
454
References
Heininger P, Höss S, Claus E, Pelzer J, Traunspurger W (2007) Nematode communities in contaminated river sediments. Environ Pollut 146: 64–76 Heip C (1976) The calculation of eliminated biomass. Biol Jb Dodonea 44: 217–225 Heip C (1980a) The influence of competition and predation on production of meiobenthic copepods. In: Tenore KR, Coull BC (eds) Marine benthic dynamics. University of South Carolina Press, Columbia, SC, pp. 166–177 (Belle W Baruch Libr Mar Sci, vol 11) Heip C (1980b) Meiobenthos as a tool in the assessment of marine environmental quality. Rapp P-V Réun Cons Int Explor Mer 179: 182–187 Heip C, De Craemer W (1974) The diversity of nematode communities in the southern North Sea. J mar biol Ass UK 54: 251–255 Heip C, Smol N, Hautekiet W (1974) A rapid method of extracting meiobenthic nematodes and copepods from mud and detritus. Mar Biol 28: 79–81 Heip C, Herman PMJ, Coomans A (1982a) The productivity of marine meiobenthos. Acad Analecta, Kl Wetensch 44: 1–20 Heip C, Vincx M, Smol N, Vranken G (1982b) The systematics and ecology of free-living marine nematodes. Helminth Abstr Ser B 51: 1–31 Heip C, Vincx M, Vranken G (1985a) The ecology of marine nematodes. Oceanogr Mar Biol Ann Rev 23: 399–489 Heip C, Herman PMJ, Smol N, van Brussel D, Vranken G (1985b) Energy flow through the meiobenthos. In: Heip C, Polk P (eds) Benthic studies of the Southern Bight of the North Sea and its adjacent continental estuaries. Vol 3, pp. 11–40 Heip C, Warwick RM, Carr MR, Herman PMJ, Huys R, Smol N, van Holsbeke K (1988) Analysis of community attributes of the benthic meiofauna of Frierfjord/ Langesundfjord. Mar Ecol Prog Ser 46: 171–180 Heip CHR, Goosen NK, Herman PMJ, Kromkamp J, Middelburg JJ, Soetaert K (1995) Production and consumption of biological particles in temperate tidal estuaries. Oceanogr Mar Biol Ann Rev 33: 1–149 Heissenberger A, Herndl GJ (1994) Formation of high molecular weight material by free-living marine bacteria. Mar Ecol Prog Ser 111: 129–135 Heissenberger A, Leppard GG, Herndl GJ (1996) Ultrastructure of marine snow. II. Microbial considerations. Mar Ecol Prog Ser 135: 299–308 Hejnol A, Schnabel R (2006) What a couple of dimensions can do for you: comparative developmental studies using 4D microscopy—examples from tardigrade development. Integr Comp Biol 46: 151–161 Helder W, Bakker JF (1985) Shipboard comparison of micro- and minielectrodes for measuring oxygen distribution in marine sediments. With Addition: technical description of manufactoring and application of needle and microoxygen electrodes, in sediments. Limnol Oceanogr 30: 1106–1108 Hellwig-Armonies M (1988) High abundance of Plathelminthes in a North Sea salt marsh creek. Progr Zool 36: 499–504. Helmkampf M, Bruchhaus I, Hausdorf B (2008) Multigene analysis of lophophorate and chaetognath phylogenetic relationships. Mol Phylogen Evol 46: 206–214 Hendelberg M, Jensen P (1993) Vertical distribution of the nematode fauna in a coastal sediment influenced by seasonal hypoxia in the bottom water. Ophelia 37: 83–94 Henderson PA (1990) Freshwater ostracods. Keys and notes for the identification of the species. Universal Book Series/Dr W. Backhuys, Oegstgeest, p. 228 (Synopses of the British fauna, vol 42) Hennig H-F, Eagle GA, Fielder L, Fricke A, Gledhill WJ, Greenwood PJ, Orren MJ (1983) Ratio and population density of psammolitoral meiofauna as a perturbation indicator of sandy beaches in South Africa. Environ Monit Assess 3: 45–60 Hentschel BT, Jumars PA (1994) In situ chemical inhibition of benthic diatom growth affects recruitment of competing, permanent and temporary meiofauna. Limnol Oceanogr 39: 816–838 Hentschel U, Berger EC, Bright M, Felbeck H, Ott JA (1999) Metabolism of nitrogen and sulfur in ectosymbiotic bacteria of marine nematodes (Nematoda, Stilbonematinae). Mar Ecol Prog Ser 183: 149–158
References
455
Heptner MV, Ivanenko VN (2002a) Copepoda (Crustacea) of hydrothermal ecosystems of the World Ocean. Arthropoda Selecta 11: 117–134 Heptner MV, Ivanenko VN (2002b) Hydrothermal vent fauna: composition, biology and adaptation. Copepoda. In: Gebruk AV (ed) Biology of hydrothermal systems. Scientific Press, Moscow, pp. 159–176 Herman PMJ, Heip C (1988) On the use of meiofauna in ecological monitoring: who needs taxonomy? Mar Pollut Bull 19: 665–668 Herman PMJ, Vranken G (1988) Studies of the life-history and energetics of marine and brackish-water nematodes. II. Production, respiration and food uptake by Monhystera disjuncta. Oecologia 77: 457–463 Herman R, Vincx M, Heip C (1985) Meiofauna of the Belgian coastal waters: spatial and temporal variability and productivity. In: Heip C, Polk P (eds) Biological processes and translocations. Concerted actions in oceanography. Ministry of Scientific Policy, Brussels, Belgium, pp. 41–63 Herman RL, Dahms HU (1992) Meiofauna communities along a depth transect off Halley Bay (Weddell Sea—Antarctica). Polar Biol 12: 313–320 Hermans CO (1983) The duo-gland adhesive system. Oceanogr Mar Biol Ann Rev 21: 283–339 Hicks GRF (1985) Biomass and production estimates for an estuarine meiobenthic copepod, with an instantaneous assessment of exploitation by flatfish predators. N Zeal J Ecol 8: 125–127 Hicks GRF (1985) Meiofauna associated with rocky shore algae. In: Moore PG, Seed R (eds) The ecology of rocky coasts. Hodder and Stoughton, London, pp. 36–64 Hicks GRF (1986) Distribution and behaviour of meiofaunal copepods inside and outside seagrass beds. Mar Ecol Prog Ser 31: 159–170 Hicks GRF (1988a) Evolutionary implications of swimming behaviour in meiobenthic copepods. Hydrobiologia: 497–505. Hicks GRF (1988b) Sediment rafting: a novel mechanism for the small-scale dispersal of intertidal estuarine meiofauna. Mar Ecol Prog Ser 48: 69–80 Hicks GRF (1989) Does epibenthic structure negatively affect meiofauna? J Exp Mar Biol Ecol 133: 39–55 Hicks GRF (1991) Monitoring with meiofauna: a compelling option for evaluating environmental stress in tidal inlets. Wat Qual Cent Publ 21: 387–391 Hicks GRF (1992) Tidal and diel fluctuations in abundance of meiobenthic copepods on an intertidal estuarine sandbank. Mar Ecol Prog Ser 87: 15–21 Hicks GRF, Coull BC (1983) The ecology of marine meiobenthic harpacticoid copepods. Oceanogr Mar Biol Ann Rev 21: 67–175 Higgins RP (1964) A method for meiobenthic invertebrate collection. Am Zool 4: 291 Higgins RP (1988) Kinorhyncha. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 328–331 Higgins RP, Storch V (1989) Ultrastructural observations of the larva of Tubiluchus corallicola (Priapulida. Helgol Meeresunters 43: 1–11 Higgins RP, Storch V (1991) Evidence for direct development in Meiopriapulus fijiensis (Priapulida). Trans Am Microsc Soc 110: 37–46 Higgins RP, Thiel H (1988) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC Hillebrand H et al. (2007) Consumer versus resource control of producer diversity depends on ecosystem type and producer community structure. Proc Natl Acad Sci 104: 10904–10909 Höckelmann C, Moens T, Jüttner F (2004) Odor compounds from cyanobacterial biofilms acting as attractants and repellents for free-living nematodes. Limnol Oceanogr 49: 1809–1819 Hockin DC (1982a) Experimental insular zoogeography: some tests of the equilibrium theory using meiobenthic harpacticoid copepods. J Biogeogr 9: 487–498. Hockin DC (1982b) The effects of sediment particle diameter upon meiobenthic copepod community of an intertidal beach: a field and a laboratory experiment. J Anim Ecol 51: 555–572 Hodda M (2006) Nematodes in lotic systems. In: Eyualem-Abebe E, Andrássy I, Traunspurger W (eds). CABI Publishing, Wallingford, Oxfordshire, pp. 163–178
456
References
Hofker J (1977) The Foraminifera of Dutch tidal flats and salt marshes. Neth J Sea Res 11: 223–296 Hogue EW, Miller CB (1981) Effects of sediment microtopography on small-scale spatial distributions of meiobenthic nematodes. J Exp Mar Biol Ecol 53: 181–191 Hohberg K, Traunspurger W (2005) Predator-prey interaction in soil food web: functional response, size-dependent foraging efficiency, and the influence of soil texture. Biol Fertil Soil 41: 419–427 Holme NA, McIntyre AD (1984) Methods for the study of marine benthos. Blackwell, Oxford, p. 387 Holopainen IJ, Paasivirta L (1977) Abundance and biomass of the meiobenthos in the oligotrophic and mesohumic lake Pääjärvi, Southern Finland. Ann Zool Fenn 14: 124–134 Holopainen IJ, Sarvala J (1975) Efficiencies of two corers in sampling softbottom invertebrates. Ann Zool Fenn 12: 280–284 Hondeveld BJM, Bak RPM, Van Duyl FC (1992) Bacterivory by heterotrophic nanoflagellates in marine sediments measured by uptake of fluorescently labeled bacteria. Mar Ecol Prog Ser 89: 63–71 Hooge MD, Tyler PA (2006) Concordance of molecular and morphological data: the example of the Acoela. Integr Comp Biol 46: 118–124 Hoppenrath M (2000) Taxonomische und ökologische Untersuchungen von Flagellaten mariner Sande. Ph.D. thesis, University of Hamburg, Hamburg Hopper BE, Meyers SP (1966) Aspects of the life cycles of marine nematodes. Helgol wiss Meeresunters 13: 444–449 Hopper GJ, Davenport J (2006) Epifaunal composition and fractal dimensions of intertidal marine macroalgae in relation to emersion. J Mar Biol Ass UK 86: 1297–1304 Horne DJ, Martens K (1994) The evolutionary ecology of reproductive modes in non-marine Ostracoda. Greenwich University Press, London, p. 73 Horne DJ, Schön I, Smith RJ, Martens K (2005) What are Ostracoda? A cladistic analysis of the extant super-families of the subclasses Myodocopa and Podocopa (Crustacea: Ostracoda). Crustac Issues 16: 249–273 Höss S, Henschel T, Haitzer M, Traunspurger W, Steinberg CEW (2001) Toxicity of cadmium to Caenorhabditis elegans (Nematoda) in whole sediment and pore water—the ambiguous role of organic matter. Envrion Toxicol Chem 20: 2794–2801 Höss S, Traunspurger W, Severin GF, Jüttner I, Pfister G, Schramm K-W (2004) Influence of 4nonylphenol on the structure of nematode communities in freshwater microcosms. Environ Toxicol Chem 23: 1268–1275 Höss S, Traunspurger W, Zullini A (2006) Freshwater nematodes in environmental science. In: Eyualem-Abebe E, Andrássy I, Traunspurger W (eds) Freshwater nematodes. Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, UK, pp. 144–162 Höss S, Weltje L (2007) Endocrine disruption in nematodes: effects and mechanisms. Ecotoxicology 16: 15–28 Hoste E, Vanhove S, Schewe I, Soltwedel TAV (2007) Spatial and temporal variations in deep-sea meiofauna assemblages in the marginal ice zone of the Arctic Ocean. Deep-Sea Res I 54: 109–129 Houle A (1999) The origin of platyrrhines: an evaluation of the Antarctic scenario and the floating island model. Amer J Physic Anthropol 109: 541–559 Howell R (1983) Heavy metals in marine nematodes: uptake, tissue distribution and loss of copper and zinc. Mar Pollut Bull 14: 263–268 Huettel M, Gust G (1992) Impact of bioroughness on interfacial solute exchange in permeable sediments. Mar Ecol Prog Ser 89: 253–267 Huettel M, Rusch A (2000) Transport and degradation of phytoplankton in permeable sediment. Limnol Oceanogr 45: 534–549 Huettel M, Ziebis W, Forster S (1996) Flow-induced uptake of particulate matter in permeable sediments. Limnol Oceanogr 41: 309–322 Hulings NC, Gray JS (1976) Physical factors controlling abundance of meiofauna on tidal and atidal beaches. Mar Biol 34: 77–83 Hull SL (1997) Seasonal changes in diversity and abundance of ostracods on four species of intertidal algae with differing structural complexity. Mar Ecol Prog Ser 161: 71–82
References
457
Hummon WD (1971) The marine and brackish-water Gastrotricha in perspective. In: Hulings NC (ed) Proceedings of the 1st International Conference on Meiofauna, Tunesia. Smiths Contr Zool 76: 21–23 Hummon WD (1972) Dispersion of Gastrotricha in a marine beach of the San Juan Archipelago, Washington. Mar Biol 16: 349–355 Hummon WD (1974) Respiratory and osmoregulatory physiology of a meiobenthic marine gastrotrich, Turbanella ocellata Hummon 1974. Cah Biol Mar 16: 255–268 Hummon WD (1989) The fetch-energy index: an a priori estimator of coastal exposure, applied to littoral marine Gastrotricha of the British Isles. In: Ryland JS, Tyler PA (eds) Reproduction, genetics and distributions of marine organisms Olsen & Olsen, Fredensborg, Denmark, pp. 387–393 Hummon WD (2004) Global database for marine Gastrotricha. Server at http://132.235.243.28 Humphreys WF (1979) Production and respiration in animal populations. J Anim Ecol 48: 427–453 Hurlbert SH (1984) Pseudoreplication and the design of ecological field experiments. Ecol Monogr 54: 187–211 Huston M (1979) A general hypothesis of species diversity. Amer Nat 113: 81–101 Huys R, Boxshall GA (1991) Copepod evolution. Ray Society, London, p. 468 Huys R, Herman PMJ, Heip CHR, Soetaert K (1992) The meiobenthos of the North Sea: density, biomass trends and distribution of copepod communities. ICES J Mar Sci 49: 23–44 Huys R, Gee JM, Moore CG, Hamond R (1996) Marine and brackish water harpacticoid copepods: Part I. Keys and notes for identification of the species. Field Studies Council, Shrewsbury, UK, p. 352 (Synopses of the British fauna, vol 51) Hylbom R (1991) Arenonemertes arenicolus sp.n. (Nemertea) from the Swedish west coast, with notes on the genus Arenonemertes. Zool Scr 20: 1–6 Hylleberg J, Henriksen K (1980) The central role of bioturbation in sediment mineralization and element re-cycling. Ophelia, Suppl 1: 1–16 Iharos G (1975) Summary of the results of forty years of research on Tardigrada. Mem Ist Ital Idrobiol Suppl. 32: 159–169 Ikeya N, Tsukagoshi A, Horne DJ (eds) (2005) Evolution and diversity of Ostracoda. Hydrobiologia 538: 1–265 Iliffe TM (1990) Crevicular dispersal of marine cave faunas. Mém Biospéol 17: 93–96 Iliffe TM (2005) Search for new life: critically endangered biodiversity hotspots in anchialine caves. Abstr 40th Eur Mar Biol Symp Wien, 21 Iliffe TM, Wilkens H, Parzefall J, Williams D (1984) Marine lava cave fauna: composition, biogeography and origins. Science 225: 309–311 Ingels J, Vanreusel A (2006) Biodiversity of meiofauna on margins of European seas. In: Tyler PA (ed) Abst 11th Int Deep-Sea Biol Symp, Southampton, 122 Ingole BS (1994) Influence of salinity on the reproductive potential of a laboratory cultured harpacticoid copepod, Amphiascoides subdebilis, Willey, 1935. Zool Anz 232: 31–40 Ingole BS, Parulekar AH (1998) Role of salinity in structuring the intertidal meiofauna of a tropical estuarine beach. Field evidence. Ind J Mar Sci 27: 356–361 Ingole BS, Goltekar R, Gonsalves S, Ansari ZA (2005) Recovery of deep-sea meiofauna after artificial disturbance in the Central Indian Basin. Mar Georesour Geotechnol 23: 253–266 Ingolfsson A (1995) Floating clumps of seaweed around iceland: natural microcosms and a means of dispersal for shore fauna. Mar Biol 122: 13–21 Ivanenko VN (1998) Deep-sea hydrothermal vent copepoda (Siphonostomatoida, Dirivultidae) in plankton over the Mid-Atlantic Ridge (29°N), morphology of their first copepodite stage. Zool Zh 77: 1–7 Iverson SJ, Field C, Bowen WD, Blanchard W (2004) Quantitative fatty acid signature analysis: a new method of estimating predator diets. Ecol Monogr 74: 211–235 Ivester MS, Coull BC (1977) Niche fractionation studies of two sympatric species of Enhydrosoma. Mikrofauna Meeresboden 61: 131–145 Jackson D (1986) A manually operated core-sampler suitable for use on fine-particulate sediments. Estuar Coast Shelf Sci 23: 419–422 Jahn A, Gamenick I, Theede H (1996) Physiological adaptation of Cyprideis torosa (Crustacea, Ostracoda) to hydrogen sulphide. Mar Ecol Prog Ser 142: 215–223
458
References
Jahnke RA (1988) A simple, reliable, and inexpensive pore-water sampler. Limnol Oceanogr 33: 483–487 Jahnke RA, Knight LH (1997) A gravity-driven hydraulically-damped multiple piston corer for sampling fine-grained sediments. Deep-Sea Res I 44: 713–718 Janssen M, Hust M, Rhiel E, Krumbein W (1999) Vertical migration behavior of diatom assemblages of Wadden Sea sediments (Dangast, Germany): a study using cryo-scanning electron microscopy. Int Microbiol 2: 103–110 Jansson B-O (1966a) Microdistribution of factors and fauna in marine sandy beaches. Veröff Inst Meeresforsch Bremerhaven Sbd 2: 77–86 Jansson B-O (1966b) On the ecology of Derocheilocaris remanei Delamare and Chappuis (Crustacea, Mystacocarida). Vie Milieu 17: 143–186 Jansson B-O (1967) Diurnal and annual variations of temperature and salinity of interstitial water in sandy beaches. Ophelia 4: 173–201 Jarvis SC, Seed R (1996) The meiofauna of Ascophyllum nodosum (L) Le Jolis: characterization of the assemblages associated with two common epiphytes. J Exp Mar Biol Ecol 199: 249–267 Jenkins RJF (1991) The early environment. In: Bryant C (ed) Metazoan life without oxygen. Chapman & Hall, London, pp. 38–64 Jenner RA (2004) Towards a phylogeny of the Metazoa: evaluating alternative phylogenetic positions of Plathyhelminthes, Nemertea, and Gnathostomulida, with a critical reappraisal of cladistic characteristics. Contr Zool 73: 3–163 Jenner RA, Scholtz G (2005) Playing another round of metazoan phylogenetics: historical epistemology, sensitivity analysis, and the position of Arthropoda within the Metazoa on the basis of morphology. Crustac Issues 16: 355–385 Jennings JB, Hick AJ (1990) Differences in the distribution, mitochondrial content and probable roles of haemoglobin-containing parenchymal cells in four species of entosymbiotic turbellarians (Rhabdocoela: Umagillidae and Pterastericolidae). Ophelia 31: 163–175 Jensen P (1981) Species distribution and a microhabitat theorie for marine mud dwelling Comesomatidae (Nematoda) in European waters. Cah Biol Mar 22: 231–241 Jensen P (1982) A new meiofauna sample splitter. Ann Zool Fenn 19: 233–236 Jensen P (1983) Life history of the free-living marine nematode Chromadorita tenuis (Nematoda: Chromadorida). Nematologica 29: 335–345 Jensen P (1983) Meiofaunal abundance and vertical zonation in a sublittoral soft bottom, with a test of the Haps corer. Mar Biol 74: 319–326 Jensen P (1984a) Ecology of benthic and epiphytic nematodes in brackish waters. Hydrobiologia 108: 201–217 Jensen P (1984b) Measuring carbon content in nematodes. Helgol Meeresunters 38: 83–86 Jensen P (1986) Nematode fauna in the sulphide-rich brine seep and adjacent bottoms of the East Flower Garden, NW Gulf of Mexico. IV. Ecological apects. Mar Biol 92: 489–503 Jensen P (1987a) Feeding ecology of free-living aquatic nematodes. Mar Ecol Prog Ser 35: 187–196 Jensen P (1987b) Differences in microhabitat, abundance, biomass and body size between oxybiotic and thiobiotic free-living marine nematodes. Oecologia (Berlin) 71: 564–567 Jensen P (1995) Life history of the nematode Theristus anoxybioticus from sublittoral muddy sediment at methane seepages in the northern Kattegatt, Denmark. Mar Biol 123: 131–136 Jensen P (1996) Burrows of marine nematodes as centres for microbial growth. Nematologica 42: 320–329 Jensen P, Rumohr J, Graf G (1992a) Sedimentological and biological differences across a deep-sea ridge exposed to advection and accumulation of fine-grained particles. Oceanol Acta 15: 287–296 Jensen P, Emrich R, Weber K (1992b) Brominated metabolites and reduced numbers of meiofauna organisms in the burrow wall lining of the deep-sea enteropneust Stereobalanus canadensis. Deep-Sea Res 39A: 1247–1253 Joergensen RG, Mueller T (1995) Estimation of the microbial biomass in tidal sediments by fumigation-extraction. Helgol Meeresunters 49: 213–221
References
459
Joint IR, Gee JM, Warwick RM (1982) Determination of fine-scale vertical distribution of microbes and meiofauna in an intertidal sediment. Mar Biol 72: 157–164 Jondelius U, Ruiz-Trillo I, Baguñà J, Riutort M (2002) The Nemertodermatida are basal bilaterians and not members of the Platyhelminthes. Zool Scr 31: 201–215 Jones RW, Charnock MA (1985) “Morphogroups” of agglutinating Foraminifera. Their life positions and feeding habits and potential applicability in (palaeo)ecological studies. Rev Paléobiol 4: 311–320 Jørgensen BB (1977a) Bacterial sulfate reduction within reduced microniches of oxidized marine sediments. Mar Biol 41: 7–17 Jørgensen BB (1977b) The sulfur cycle of a coastal marine sediment (Limfjorden, Denmark). Limnol Oceanogr 22: 814–832 Jørgensen BB (1982) Ecology of the bacteria of the sulphur cycle with special reference to anoxicoxic interface environments. Phil Trans R Soc Lond B 298: 543–561 Jørgensen BB (1990) Thiosulfate shunt in the sulfur cycle of marine sediments. Science 249: 152–154 Jørgensen BB, Bak F (1991) Pathways and microbiology of thiosulfate transformations and sulfate reduction in a marine sediment (Kattegat, Denmark). Appl Environ Microbiol 57: 847–856 Jørgensen BB, Revsbech NP (1985) Diffuse boundary layers and the oxygen uptake of sediments and detritus. Limnol Oceanogr 30: 111–122 Jørgensen CB (1976) August Pütter, August Krogh, and the modern ideas on the use of dissolved organic matter in aquatic environments. Biol Rev 51: 291–328 Jørgensen NOG (1979) Annual variation of dissolved free primary amines in estuarine water and sediment. Oecologia (Berl) 40: 207–217 Jørgensen NOG, Mopper K, Lindroth P (1980) Occurence, origin, and assimilation of free amino acids in an estuarine environment. Ophelia Suppl. 1: 179–192 Jørgensen NOG, Lindroth P, Mopper K (1981) Extraction and distribution of free amino acids and ammonium in sediment interstial waters from the Limfjord, Denmark. Oceanol Acta 4: 465–474 Josefson AB, Hansen JLS (2003) Quantifying plant pigments and live diatoms in aphotic sediments of Scandinavian coastal waters confirms a major route in the pelagic-benthic coupling. Mar Biol 142: 649–658 Josefson AB, Widbom B (1988) Differential response of benthic macrofauna and meiofauna to hypoxia in the Gullmar Fjord basin. Mar Biol 100: 31–40 Jouin C (1992) The ultrastructure of a gutless annelid, Parenterodrilus gen.nov. taenioides (=Astomus taenioides) (Polychaeta, Protodrilidae). Can J Zool 70 (9): 1833–1848 Jouk PEH, Martens PM, Schockaert ER (1988) Horizontal distribution of the Plathelminthes in a sandy beach of the Belgian coast. Progr Zool 36: 481–487 Kaariainen JI, Bett BJ (2006) Evidence for benthic body size miniaturization in the deep sea. J Mar Biol Ass UK 86: 1339–1345 Kamenev GM, Vi F, Selin NI, Tarasov VG (1993) Composition and distribution of macro- and meiobenthos around sublittoral hydrothermal vents in the Bay of Plenty, New Zealand. NZ J Mar Freshw Res 27: 407–418 Kammenga JE, Herman MA, Ouborg NJ, Johnson L, Breitling R (2007) ) Microarray challenges in ecology. Trends Ecol Evol 22: 273–279 Kampfer S, Sturmbauer C, Ott J (1998) Phylogenetic analysis of rDNA sequences from adenophorean nematodes and implications for the Adenophorea—Secernentea controversy. Invert Biol 117: 29–36 Kanneworff E, Nicolaisen W (1973) The Haps: a frame supported bottom corer. Ophelia 10: 119–129 Kanneworff E, Nicolaisen W (1983) A simple, hand-operated quantitative bottom sampler. Ophelia 22: 253–255 Kapp H, Giere O (2005) Spadella interstitialis, a meiobenthic chaetognath from Mediterranean calcareous sands. Meiofauna Mar 14: 109–114 Karaman S (1935) Die Fauna der unterirdischen Gewässer Jugoslawiens. Verh int Ver Limnol 7: 46–73
460
References
Karl DM, La Rock PA (1975) Adenosine triphosphate measurements in soil and marine sediments. J Fish Res Bd Can 32: 599–607 Katayama T, Yamamoto M, Wada H, Satoh N (1993) Phylogenetic position of acoel turbellarians inferred from partial 18S rDNA sequences. Zool Sci 10: 529–536 Katz S, Cavanaugh CM, Bright M (2006) Symbiosis of epi- and endocuticular bacteria with Helicoradomenia spp. (Mollusca, Aplacophora, Solenogastres) from deep-sea hydrothermal vents. Mar Ecol Prog Ser 320: 89–99 Kaye BH (1993) Chaos & complexity: discovering the surprising patterns of science and technology. Sections 3.11- 3.14: Using dot counting to estimate irregular areas. VCH, Verlagsges, pp. 140–155 Kelly JM (1993) Ballast water and sediments as mechanisms for unwanted species introductions into Washington State. J Shelf Res 12: 405–410 Kemp PF (1988) Bacterivory by benthic ciliates: significance as a carbon source and impact on sediment bacteria. Mar Ecol Prog Ser 49: 163–169 Kemp PF (1990) The fate of benthic bacterial production. Rev Aquat Sci 2: 109–124 Kemp PF (1994) Benthic microbial ecology. Marine science book series. CRC Press, Boca Raton, FL Kendall MA, Warwick RM, Somerfield PJ (1997) Species size distributions in Arctic benthic communities. Polar Biol 17: 389–392 Kennedy AD (1993) Minimal predation upon meiofauna by endobenthic macrofauna in the Exe Estuary, south west England. Mar Biol 117: 311–319 Kennedy AD (1994a) Carbon partitioning within meiobenthic nematode communities in the Exe Estuary, UK. Mar Ecol Prog Ser 105: 71–78 Kennedy AD (1994b) Predation within meiofaunal communities: description and results of a rapid-freezing method of investigation. Mar Ecol Prog Ser 114: 71–79 Kennedy AD, Jacoby CA (1999) Biological indicators of marine environmental health: meiofauna—a neglected benthic component? Environ Monit Assess 54: 47–68 Kepkay PE (1994) Particle aggregation and the biological reactivity of colloids. Mar Ecol Prog Ser 109: 293–304 Kern JC, Carey AG (1983) The faunal assemblage inhabiting seasonal sea ice in the nearshore Arctic Ocean with emphasis on copepods. Mar Ecol Prog Ser 10: 159–167 Kerr RA (1998) Pushing back the origins of animals. Science 279: 803–804 King GM (1986) Inhibition of microbial activity in marine sediments by a bromphenol from a hemichordate. Nature 323: 257–259 Kinne O (1964) Non-genetic adaptation to temperature and salinity. Helgol wiss Meeresunters 9: 433–458 Kirsteuer E (1976) Notes on adult morphology and larval development of Tubiluchus corallicola (Priapulida), based on in vivo and scanning electron microscopic examinations of specimens from Bermuda. Zool Scr 5: 239–255 Kirsteuer E (1977) Remarks on taxonomy and geographic distribution of the genus Ototyphlonemertes Diesing (Nemertina, Monostilifera). Mikrofauna Meeresboden 61: 167–181 Kisielewski J (1987) Two new interesting genera of Gastrotricha (Macrodasyoida and Chaetonotida) from the Brazilian freshwater psammon. Hydrobiologia 153: 23–30 Kisielewski J (1990) Origin and phylogenetic significance of freshwater psammic Gastrotricha. Stygologia 5: 87–92 Klimant I, Meyer V, Kühl M (1995) Fiber-optic microsensors, a new tool in aquatic biology. Limnol Oceanogr 40: 1159–1165 Kogure K, Do HK, Kim DS, Shirayama Y (1996) High concentration of neurotoxin in free-living marine nematodes. Mar Ecol Prog Ser 136: 147–151 Kolasa J (2000) The biology and ecology of lotic microturbellarians. Freshw Biol 44: 5–14 Kolesnikova YA, Povchun AS, Serenko IV (1995) Migration of meiobenthos in the inshore zone of the Black Sea. Hydrobiol J 31: 45–54 Koller H, Dworschak PC, Abed-Navandi D (2006) Burrows of Pestarella tyrrhena (Decapoda: Thalassinidea): hot spots for Nematoda, Foraminifera and bacterial densities. J Mar Biol Ass UK 86: 1113–1122
References
461
Koop K, Griffiths CL (1982) The relative significance of bacteria, meio- and macrofauna on an exposed sandy beach. Mar Biol 66: 295–300 Koski M, Kiorboe T, Takahashi K (2005) Benthic life in the pelagic: aggregate encounter and degradation rates by pelagic harpacticoid copepods. Limnol Oceanogr 50: 1254–1263 Köster M, Meyer-Reil L-A (2001) Characterization of carbon and microbial biomass pools in shallow water coastal sediments of the southern Baltic Sea (Nordrügensche Bodden). Mar Ecol Prog Ser 214: 25–41 Kotwicki L, De Troch M, Urban-Malinga B, Gheskiere T, Weslawski JM (2005a) Horizontal and vertical distribution of meiofauna on sandy beaches of the North Sea (The Netherlands, Belgium, France). Helgol Mar Res 59: 255–264 Kotwicki L, Szymelfenig M, De Troch M, Urban-Malinga B, Weslawski JM (2005b) Latitudinal biodiversity pattern of meiofauna from sandy littoral beaches. Biodivers Conservat 14: 461–474 Kovalevsky A (1901) Les Hédylides, études anatomiques. Mém Acad Sci St Petersburg (Sci Math Phys Nat) 12: 1–32 Kovatch CE, Chandler GT, Coull BC (1999) Utility of a full life-cycle copepod bioassay approach for assessment of sediment-associated contaminant mixtures. Mar Pollut Bull 38: 692–701 Kowarc VA (1990) Production of a harpacticoid copepod from the meiofaunal community of a second order mountain stream. Stygologia 5: 25–3 Kraus MG, Found BW (1975) Preliminary observation on the salinity and temperature tolerances and salinity preferences of Derocheilocaris typica Pennak and Zinn 1943. Cah Biol Mar 16: 751–762 Krembs C, Gradinger R, Spindler M (2000) Implications of brine channel geometry and surface area for the interaction of sympagic organisms in Arctic sea ice. J Exp Mar Biol Ecol 243: 55–80 Kristensen E (1984) Life cycle, growth and production in estuarine populations of the polychaetes Nereis virens and N. diversicolor. Holarct Ecol 7: 249–256 Kristensen E (1988) Benthic fauna and biogeochemical processes in marine sediments: microbial activities and fluxes. In: Blackburn TH, Sørensen J (eds) Nitrogen cycling in coastal marine environments. Wiley, Chichester, pp. 275–299 Kristensen E, Ahmed SI, Devol AH (1995) Aerobic and anaerobic decomposition olf organic matter in marine sediment: Which is faster? Limnol Oceanogr 40: 1430–1437 Kristensen RM (1982) The first record of cyclomorphosis in Tardigrada based on a new genus and species from Arctic meiobenthos. Z zool Syst Evolutionsforsch 20: 249–270 Kristensen RM (1983) Loricifera, a new phylum with Aschelminthes characters from the meiobenthos. Z zool Syst Evolutforsch 21: 161–180 Kristensen RM (1991a) Loricifera. 4: Aschelminthes. Wiley-Liss, Chichester, pp. 351–375 Kristensen RM (1991b) Loricifera—A general biological and phylogenetic overview. Verh Dtsch Zool Ges Gießen 84 B, 231–246 Kristensen RM (2002) An introduction to Loricifera, Cycliophora, and Micrognathozoa. Integr Comp Biol 42: 641–651 Kristensen RM, Funch P (2000) Micrognathozoa: A new class with complicated jaws like those of Rotifera and Gnathostomulida. J Morphol 246: 1–49 Kristensen RM, Heiner I, Higgins RP (2007) Morphology and life cycle of a new loriciferan from the Atlantic coast of Florida with an emended diagnosis and life cycle of Nanaloricidae (Loricifera). Invert Biol 126: 120–137 Kristensen RM, Higgins RP (1984) A new family of Arthrotardigrada (Tardigrada: Heterotardigrada) from the Atlantic coast of Florida, U.S.A. Trans Am Microsc Soc 103: 295–311 Kristensen RM, Higgins RP (1991) Kinorhyncha. Microscopic anatomy of invertebrates, vol 4: Aschelminthes. Wiley-Liss, Chichester, pp. 377–404 Krogh A, Spärck R (1936) On a new bottom sampler for investigation of the microfauna of the sea bottom. K dansk Vidensk Selsk Skr 13: 1–12 Krumbein WC (1939) Graphic presentation and statistical analysis of sedimentary data. In: Trask PD (ed) Recent marine sediments. A symposium. American Association Of Petroleum Geology, Tulsa, Oklahoma. Murby, London, pp. 558–591
462
References
Krumbein WE, Paterson DM, Stal LJ (1994) Biostabilization of sediments. BIS Univ Oldenburg, Oldenburg, p. 526 Kuehn KA, O’Neil RM, Koehn RD (1992) Viable photosynthetic microalgal isolates from aphotic environments of the Edwards aquifer (Central Texas). Stygologia 7: 129–142 Kühl M, Lassen C, Jørgensen BB (1994) Light penetration and light intensity in sandy marine sediments measured with irradiance and scalar irradiance fiber-optic microprobes. Mar Ecol Prog Ser 105: 139–148 Kühl M, Steuckart C, Eickert G, Jeroschewski P (1998) A H2S microsensor for profiling biofilms and sediments: application in an acidic lake sediment. Aquat Microb Ecol 15: 201–209 Kuipers BR, Wilde PAWJD, Creutzberg F (1981) Energy flow in a tidal flat ecosystem. Mar Ecol Prog Ser 5: 215–221 Kurashov EA (2002) The role of meiobenthos in lake ecosystems. Aquat Ecol 36: 447–463 Kurdziel JP, Bell SS (1992) Emergence and dispersal of phytal-dwelling meiobenthic copepods. J Exp Mar Biol Ecol 163: 43–64 LaBarbera M, Vogel S (1976) An inexpensive thermistor flowmeter for aquatic biology. Limnol Oceanogr 21: 750–756 Lackey JB (1936) Occurrence and distribution of the marine protozoan species in the Woods Hole area. Biol Bull 70: 264–278 Lafont M, Durbec A (1990) Essai de description biologique des interaction entre eau de surface et eau souterraine: vulnérabilité d’un aquifère à la pollution d’un fleuve. Annls Limnol 26: 119–129 Lafont M, Durbec A, Ille C (1992) Oligochaete worms as biological describers of the interactions between surface and groundwaters: a first synthesis. Regul Rivers Res Manag 7: 65–73 Lambshead PJD, Packer M (eds) (2006) Marine benthic nematode molecular protocol handbook. ISA Publications, Tucson, AZ Lambshead PJD (1993) Recent developments in marine benthic biodiversity research. Océanis 19: 5–24 Lambshead PJD, Boucher G (2003) Marine nematode deep-sea biodiversity—hyperdiversity or hype?. J Biogeogr 30: 475–485 Lambshead PJD, Gooday AJ (1990) The impact of seasonally deposited phytodetritus on epifaunal and shallow infaunal benthic foraminiferal populations in the bathyal northeast Atlantic: the assemblage response. Deep-Sea Res 37: 1263–1283 Lambshead PJD, Hodda M (1994) The impact of disturbance on measurements of variability in marine nematode populations. Vie Milieu 44: 21–27 Lambshead PJD, Platt HM, Shaw KM (1983) The detection of differences among assemblages of marine benthic species based on an assessment of dominance and diversity. J Nat Hist 17: 859–874 Lambshead PJD, Tietjen J, Ferrero T, Jensen P (2000) Latitudinal diversity gradients in the deep sea with special reference to North Atlantic nematodes. Mar Ecol Prog Ser 194: 159–167 Lambshead PJD, Tietjen J, Glover A, Ferrero T, Thistle D, Gooday A (2001) The impact of largescale natural physical disturbance on the diversity of deep-sea North Atlantic nematodes. Mar Ecol Prog Ser 214: 121–126 Lambshead JPD, Brown CJ, Ferrero TJ, Mitchell NJ, Smith CR, Hawkins LE, Tietjen J (2002) Latitudinal diversity patterns of deep-sea marine nematodes and organic fluxes: a test from the central equatorial Pacific. Mar Ecol Prog Ser 236: 129–136 Lampadariou N, Hatziyanni E, Tselepides A (2003) Community structure of meiofauna and macrofauna in Mediterranean Deep-Hyper-saline Anoxic Basins. In: Briand F (ed) Mare Incognitum? Exploring Mediterranean deep-sea biology CIESM Monaco, Heraklion, pp. 55–60 Landén A, Hall POJ (1998) Seasonal variation of dissolved and adsorbed amino acids and ammonium in a near-shore marine sediment. Mar Ecol Prog Ser 170: 67–84 Lang K (1988) Monographie der Harpacticiden I und II. Reprint. Otto Koeltz Science Publisher, Koenigstein, p. 1682 Larkin KE, Gooday AJ, Pond DW, Bett BJ (2006) Fatty acid analysis reveals the importance of foraminifera in benthic organic matter cycling within an oxygen minimum zone. In: Tyler PA (ed) Abst 11th Int Deep-Sea Biol Symp , Southampton, p. 59
References
463
Larson F, Petersen DG, Dahlloef I, Sundbaeck K (2007) Combined effects of an antifouling biocide and nutrient status on a shallow-water microbenthic community. Aquat Microb Ecol 48: 277–294 Lassen C, Ploug H, Jørgensen BB (1992) A fibre-optic scalar irradiance microsensor: application for spectral light measurements in sediments. FEMS Microbiol Ecol 86 (3): 247–254 Lassèrre P (1976) Metabolic activities of benthic microfauna and meiofauna. Recent advances and review of suitable methods of analysis. In: McCave IN (ed) The Benthic boundary layer. Plenum, New York, pp. 95–142 Lassèrre P, Tournié T (1984) Use of microcalorimetry for the characterization of marine metabolic activity at the water-sediment interface. J Exp Mar Biol Ecol 74: 123–139 Leduc D (2008) Description of Oncholaimus moanae sp. nov. (Nematoda: Oncholaimidae), with notes on feeding ecology based on isotopic and fatty acid composition. J Mar Biol Ass UK (in press) Lee DL (2002) The biology of nematodes. Taylor and Francis, London, p. 635 Lee JJ (1980a) A conceptual model of marine detrital decomposition and the organisms associated with the process. Adv Aquat Microbiol 2: 257–291 Lee JJ (1980b) Nutrition and physiology of the Foraminifera. In: Levandowsky M, Hutner SH (eds) Biochemistry and physiology of Protozoa III. Academic, New York, pp. 43–66 Lee MR, Correa JA, Castilla JC (2001) An assessment of the potential use of the nematode to copepod ratio in the monitoring of metal pollution. The Chanaral Case. Mar Pollut Bull 42: 696–701 Lee MR, Correa JA (2006) An assessment of the impact of copper mine tailings disposal on meiofaunal assemblages using microcosm bioassays. Mar Environ Res 64: 1–20 Lee RF, Page DS (1997) Petroleum hydrocarbons and their effects in subtidal regions after major oil spills. Mar Pollut Bull 34: 928–940 Lee RW (2003) Physiological adaptations of the invasive cordgrass Spartina anglica to reducing sediments: rhizome metabolic gas fluxes and enhanced O2 and H2S transport. Mar Biol 143: 9–15 Lee RW, Kraus DW, Doeller JE (1999) Oxidation of sulfide by Spartina alterniflora roots. Limnol Oceanogr 44: 1155–1159 Levin L (1991) Interactions between metazoans and large agglutinating protozoans: implications for the community structure of deep-sea bottoms. Am Zool 31: 886–900. Levin L (2003) Oxygen minimum zone benthos: adaptation and community response to hypoxia. Oceanogr Mar Biol Ann Rev 41: 1–45 Levin L (2005) Ecology of cold seep sediments: interactions of fauna with flow, chemistry, and microbes. Oceanogr Mar Biol Ann Rev 43: 1–46 Levin L, Blair N, DeMaster D, Plaia G, Fornes W, Martin C, Thomas C (1997) Rapid subduction of organic matter by maldanid polychaetes on the North Carolina slope. J Mar Res 55: 595–611 Levin LA, Huggett CL, Wishner KF (1991) Control of deep-sea benthic community structure by oxygen and organic-matter gradients in the eastern Pacific Ocean. J Mar Res 49: 763–800 Levinton JS (1983) The latitudinal compensation hypothesis: Growth data and a model of latitudinal growth differentiation based upon energy budgets. I. Interspecific comparison of Ophryotrocha (Polychaeta: Dorvilleidae. Biol Bull 165: 686–698 Levinton JS, Monahan RK (1983) The latitudinal compensation hypothesis: growth data and a model of latitudinal growth differentiation based upon energy budgets. II. Interspecific comparisons between subspecies of Ophryotrocha puerilis (Polychaeta: Dorvilleidae). Biol Bull 165: 699–707 Li J, Vincx M (1993) The temporal variation of intertidal nematodes in the Westerschelde. I. The importance of an estuarine gradient. Neth J Aquat Ecol 27: 319–326 Li J, Vincx M, Herman PMJ, Heip C (1997) Monitoring meiobenthos using cm-, m- and km-scales in the southern Bight of the North Sea. Mar Env Res 43: 265–278 Liebezeit G, Felbeck H, Dawson R, Giere O (1983) Transepidermal uptake of dissolved carbohydrates by the gutless marine oligochaete Phallodrilus leukodermatus (Annelida). Océanis 9: 205–211
464
References
Limén H, Levesque C, Juniper SK (2007) POM in macro-/meiofaunal food webs associated with three flow regimes at deep-sea hydrothermal vents on Axial Volcano, Juan de Fuca Ridge. Mar Biol 153: 129–139 Lindegarth M, Jonsson PR, André C (1991) Fluorescent microparticles: a new way of visualizing sedimentation and larval settlement. Limnol Oceanogr 36: 1471–1476 Litvaitis MK, Bates JW, Hope WD, Moens T (2000) Inferring a classification of the Adenophorea (Nematoda) from nucleotide sequences of the D3 expansion segment (26/28S rDNA). Can J Zool 78: 911–922 Litvaitis MK, Nunn G, Thomas WK, Kocher TD (1994) A molecular approach for the identification of meiofaunal turbellarians (Plathyhelminthes, Turbellaria). Mar Biol 120: 437–442. Locke JM (2000) Ultrastructure of the statocyst of the marine enchytraeid Grania americana (Annelida: Clitellata). Invert Biol 119: 83–93 Logan BE (1993) Theoretical analysis of size distributions determined with screens and filters. Limnol Oceanogr 38: 372–381. Logan BE, Passow U, Alldredge AL, Grossart HP, Simon M (1995) Rapid formation and sedimentation of large aggregates is predictabe from coagulation rates (halflives) of transparent exopolymer particles (TEP). Deep-Sea Res 42: 230–214 Lohrer AM, Thrush SF, Gibbs MM (2004) Bioturbators enhance ecosystem function through complex biogeochemical interactions. Nature 431: 1092–1095 Lombardi J, Ruppert EE (1982) Functional morphology of locomotion in Derocheilocaris typica (Crustacea, Mystacocarida). Zoomorphology 100: 1–10 Long BG, Wang YG (1994) Method for comparing the capture efficiency of benthic sampling devices. Mar Biol 121: 397–399 Long ER, Chapman PM (1985) A sediment quality triad: measures of sediment contamination, toxicity and infaunal community composition in Puget Sound. Mar Pollut Bull 16: 405–415 Lorenzen S (1986) Nematoda: interstitial nematodes from marine, brackish and hypersaline environments. In: Botosaneanu (ed) Stygofauna Mundi. Brill, Leiden, pp. 133–142 Lorenzen S (1994) The phylogenetic systematics of freeliving nematodes. The Ray Society, London Lorenzen S (2000) The role of the biogenetic convergence rule in polarizing transformation series—arguments from nematology, chaos science, and phylogenetic systematics. Ann Zool (Warszawa) 50: 267–275 Lorenzen S, Prein M, Valentin C (1987) Mass aggregations of the free-living marine nematode Pontonema vulgare (Oncholaimidae) in organically polluted fjords. Mar Ecol Prog Ser 37: 27–34 Lovén S (1844) Chaetoderma, ett nytt masksläkte n.g. Öfvers Kungl Vetenskaps-Akad Förh 1: 116+pl.112 Lucchesi P, Santangelo G (1997) The interstitial ciliate microcommunity of a Mediterranean sandy shore under differing hydrodynamic disturbances. Ital J Zool 64: 253–259 Luckenbach MW (1986) Sediment stability around animal tubes: the roles of hydrodynamic processes and biotic activity. Limnol Oceanogr 31: 779–787. Luth C, Luth U, Gebruk AV, Thiel H (1999) Methane gas seeps along the oxic/anoxic gradient in the Black Sea: manifestations, biogenic sediment compounds and preliminary results on benthic ecology. Mar Ecol PSZN 20: 221–249 Maas A, Huang D, Chen J, Waloszek D, Braun A (2007) Maotianshan-Shale nemathelminths— morphology, biology, and the phylogeny of Nemathelminthes. Palaeogeogr Palaeoclimatol Palaeoecol 254: 288–306 MacIntyre HL, Geider RJ, Miller DC (1996) Microphytobenthos: the ecological role of the “secret garden” of unvegetated, shallow-water marine habitats, I. Distribution, abundance and primary production. Estuaries 19: 186–201 Maddocks RF (1992) Crustacea. Ostracoda. In: Harrison FW, Humes AG (eds) Microscopic anatomy of invertebrates. Wiley-Liss, New York, pp. 415–441 Madsen KN, Nilsson P, Sundbäck K (1993) The influence of benthic microalgae on the stability of a subtidal sediment. J Exp Mar Biol Ecol 170: 159–177 Mahmoudi E, Essid N, Beyrem H, Hedfi A, Boufahja F, Aissa P (2007) Réponse d’une communautée de nématodes libre marins à une contamination par un métal lourd (le cobalt): étude microcosmique. Bull Soc Zool France 132: 111–123
References
465
Malakhov VV (1994) Nematodes. Structure, development, classification, and phylogeny. English translation by GV Bentz. Smithsonian Institution Press, Washington, DC, p. 286 Malan DE, McLachlan A (1991) In situ benthic oxygen fluxes in a nearshore coastal marine system: a new approach to quantify the effect of wave action. Mar Ecol Prog Ser 73: 69–81 Malard F, Gibert J, Laurent R, Reygrobellet J-L (1994) A new method for sampling the fauna of deep karstic aquifers (Une nouvelle méthode d’échantillonnage de la faune des aquifères karstiques profonds). C R Acad Sci Paris Sci, Vie, Sér 3 317: 955–966 Mallat J, Giribet G (2006) Further use of nearly complete 28S and 18S rRNA genes to classify Ecdysozoa: 37 more arthropods and a kinorhynch. Mol Phylogen Evolut 40: 772–794 Mangubhai S, Greenwood JG (2004) A simple practical method for bulk and rapid extraction of free-living nematodes from marine and estuarine sediments. Hydrobiologia 522: 343–347 Mangum C (1991) Precambrian oxygen levels, the sulfide biosystem, and the origin of the Metazoa. J exp Zool 260: 33–42 Manylov OG, Vladychenskaya NS (2004) Analysis of 18S rRNA gene sequences suggests significant molecular differences between Macrodasyida and Chaetonotida (Gastrotricha). Mol Phylogen Evolut 30: 850–854 Marcotte BM (1983) The imperatives of copepod diversity: perception, cognition, competition and predation. In: Schram FR (ed) Crustacean phylogeny. Balkema, Rotterdam, pp. 47–72 Marcotte BM (1984) Behaviourally defined ecological resources and speciation in Tisbe (Copepoda: Harpacticoida). J Crust Biol 4: 404–416 Marcotte BM (1986a) Sedimentary particle sizes and the ecological grain of food resources for meiobenthic copepods. Estuar Coast Shelf Sci 23: 423–427 Marcotte BM (1986b) Phylogeny of the Copepoda Harpacticoida. Syllogeus 58: 186–190 Mare MF (1942) A study of a marine benthic community with special reference to the microorganisms. J Mar Biol Ass UK 25: 517–554 Mari X, Dam HG (2004) Production, concentration, and isolation of transparent exopolymeric particles using paramagnetic functionalized microspheres. Limnol Oceanogr Methods 2: 13–24 Markmann M, Tautz D (2005) Reverse taxonomy: an approach towards determining the diversity of meiobenthic organisms based on ribosomal RNA signature sequences. Phil Trans R Soc B Biol Sci 360: 1917–1924 Marmonier P, Vervier P, Gibert J, Dole-Olivier M-J (1993) Biodiversity in groundwaters: a research field in progress. Trends Ecol Evol 8: 392–395 Marotta R, Guidi L, Pierboni L, Ferraguti M, Todaro MA, Balsamo M (2005) Sperm ultrastructure of Macrodasys caudatus (Gastrotricha: Macrodasyida) and a sperm-based phylogenetic analysis of Gastrotricha. Meiofauna Mar 14: 9–21 Marshall BA (2006) Four new species of Monoplacophora (Mollusca) from the New Zealand region. Moll Res 26: 61–68 Martens K (1994) A European network on the evolutionary ecology of reproductive modes in non-marine Ostracoda: background and objectives. In: Horne DJ, Martens K (eds) The evolutionary ecology of reproductive modes in non-marine Ostracoda. Greenwich University Press, London, pp. 3–7 Martens K, Rossetti G, Horne DJ (2003) How ancient are ancient asexuals?. Proc R Soc B 270: 723–729 Martens PM, Schockaert ER (1986) The importance of turbellarians in the marine meiobenthos: a review. Hydrobiologia 132: 295–303. Martin MW, Grazhdankin DV, Bowring SA, Evans DAD, Fedonkin MA, Kirschvink JL (2000) Age of Neoproterozoic bilatarian body and trace fossils, White Sea, Russia: implications for metazoan evolution. Science 288: 841–845 Martinez Arbizu P, Ivanenko VN, G Gad G (2006) Hydrothermal vent fauna in the Midatlantic Ridge. Is the Romanche Fracture Zone a biogeographic barrier? Lessons from meiofauna. In: Tyler PA (ed) Abstr 11th Int Deep-Sea Biol Symp, Southampton, p. 61 Mason WT, Yevich PP (1967) The use of phloxine B and Rose bengal stains to faciliate sorting benthic samples. Trans Am Microsc Soc 86: 221–223 Mathieu J, Marmonier P, Laurent R, Martin D (1991) Récolte du matériel biologique aquatique souterrain et stratégie d’échantillonnage. Hydrogéologie 3: 217–223
466
References
Mayer LM, Rossi PM (1982) Specific surface areas in coastal sediments: relationships with other textural factors. Mar Geol 45: 241–252 Mayr E (1982) Processes of speciation in animals. In: Barigozzi C (ed) Mechanisms of speciation. Liss, New York, pp. 1–19 McCall JN, Fleeger JW (1995) Predation by juvenile fish on hyperbenthic meiofauna. A review with data on post-larval Leiostomus xanthurus. Vie Milieu 45: 61–73. McCall PL, Tevesz MJS (eds) (1982) Animal–sediment relations. The biogenic alteration of sediments. In: Stehli FG (ed) Topics in Geology 2. Plenum, New York, p. 336 McCann KS (2000) The diversity–stability debate. Nature 405: 228–233 McDowell EM (1978) Fixation and processing. In: Trump BFJ, Jones RT (eds) Diagnostic electron microscopy Wiley, New York, pp. 113–139 McGinty MM, Higgins RP (1968) Ontogenetic variation of taxonomic characters of two marine tardigrades with the description of Batillipes bullacaudatus n.sp. Trans Am Microsc Soc 87: 252–262 McHugh D (2000) Molecular phylogeny of the Annelida. Can J Zool 78: 1873–1884 McHugh D (2005) Molecular systematics of polychaetes (Annelida). Hydrobiologia 535/536: 309–318 McIntyre AD (1968) The meiofauna and macrofauna of some tropical beaches. J Zool (Lond) 156: 377–392. McIntyre AD (1969) Ecology of marine meiobenthos. Biol Rev 44: 245–290 McIntyre AD, Warwick RM (1984) Meiofauna techniques. In: Holme NA, McIntyre AD. (eds) Methods for the study of marine benthos Blackwell Scientific, Oxford, pp. 217–244 McLachlan A (1977) Studies on the psammolittoral meiofauna of Algoa Bay, South Africa. II. The distribution, composition and biomass of the meiofauna and macrofauna. Zool Afr 12: 33–60 McLachlan A (1980) The definition of sandy beaches in relation to exposure: a simple rating system. S Afr J Sci 76: 137–138 McLachlan A (1985) The biomass of macro- and interstitial fauna on clean and wrack-covered beaches in Western Australia. Est Coast Shelf Sci 21: 587–599 McLachlan A (1989) Water filtration by dissipative beaches. Limnol Oceanogr 34: 774–779 McLachlan A, Harty B (1982) Effects of crude oil on the supralittoral meiofauna of a sandy beach. Mar Environ Res 7: 71–79 McLachlan A, Romer G (1990) Trophic relations in a high energy beach and surf-zone ecosystem. In: Barnes M, Gibson RN (eds) Trophic relations in the marine environment. Aberdeen University Press, pp. 356–371 McLachlan A, Turner I (1994) The interstitial environment of sandy beaches. Mar Ecol PSZN 14 (3–4): 177–211 McLachlan A, Erasmus T, Furstenberg JP (1977) Migration of sandy beach meiofauna. Zool Afr 12: 257–277 McLachlan A, Dye AH, van der Ryst P (1979) Vertical gradients in the fauna and oxidation of two exposed sandy beaches. S Afr J Zool 14: 43–49 McLachlan A, Woolridge T, Dye AH (1981) The ecology of sandy beaches in southern Africa. S Afr J Zool 16: 219–231 McNair JN, Newbold JD, Hart DD (1997) Turbulent transport of suspended particles and dispersing benthic organisms: how long to hit bottom? J Theor Biol 188: 29–52 McNeill S, Lawton JH (1970) Annual production and respiration in animal populations. Nature 225: 472–474 Meadows A, Meadows PS (1994) Bioturbation in deep sea Pacific sediments. J Geol Soc Lond 151: 361–375 Meadows PS (1986) Biological activity and seabed sediment structure. Nature 323: 207 Meadows PS, Anderson FG (1966) Micro-organisms attached to marine and freshwater sand grains. Nature 212: 1059–1060 Meadows PS, Anderson FG (1968) Micro-organisms attached to marine sand grains. J Mar Biol Ass UK 48: 161–175
References
467
Meadows PS, Tait J (1985) Bioturbation, geotechnics and microbiology at the sediment-water interface in deep-sea sediments. In: Gibbs PE (ed) Proceedings of the 19th European Marine Biology Symposiium. Cambridge University Press, Cambridge, pp. 191–199 Meadows PS, Tait J (1989) Modification of sediment permeability and shear strength by two burrowing invertebrates. Mar Biol 101: 75–82 Meadows PS, Tait J, Hussain SA (1990) Effects of estuarine infauna on sediment stability and particle sedimentation. Hydrobiologia 190: 263–266 Meineke T, Westheide W (1979) Gezeitenabhängige Wanderungen der Interstitialfauna in einem Gezeitenstrand der Insel Sylt (Nordsee). Mikrofauna Meeresboden 75: 203–236 Meisch C (2000) Freshwater Ostracoda of Western and Central Europe. In: Schwoerbel J, Zwick P (eds) Süßwasserfauna von Mitteleuropa. Spektrum Akademie Verlag, Heidelberg, p. 522 Meldal BHM, Debenham NJ, De Ley P, De Ley IT, Vanfleteren JR, Vierstraete AR, Bert W, Borgonie G, Moens T, Tyler PA, Austen MC, Blaxter ML, Rogers AD, Lambshead JPD (2007) An improved molecular phylogeny of the Nematoda with special emphasis on marine taxa. Mol Phylogen Evolut 42: 622–636 Menden-Deuer S, Lessard EJ, Satterberg J (2001) Effects of preservation on dinoflagellate and diatom cell volume and consequences for carbon biomass predictions. Mar Ecol Prog Ser 222: 41–50 Menker D, Ax P (1970) Zur Morphologie von Arenadiplosoma migrans n.g.n.sp. einer vagilen Ascidien-Kolonie aus dem Mesopsammal der Nordsee (Tunicata, Ascidiacea). Z Morph Tiere 66: 323–336 Menn I (2002a) Beach morphology and food web structure: comparison of an eroding and an accreting sandy shore in the North Sea. Helgol Mar Res 56: 177–189 Menn I (2002b) Ecological comparison of two sandy shores with different wave energy and morphodynamics in the North Sea. Ber Polarforsch Meeresforsch 417: 1–170 Menn I, Armonies W (1999) Predatory Promesostoma species (Plathelminthes, Rhabdocoela) in the Wadden Sea. J Sea Res 41: 309–320 Meyer HA, Bell SS (1989) Response of harpacticoid copepods to detrital accumulation on seagrass blades: A field experiment with Metis holothuriae (Edwards). J Exp Mar Biol Ecol 132: 141–149 Meyer R, Bartolomaeus T (1997) Ultrastruktur und Morphogenese der Hakenborsten bei Psammodrilus balanoglossoides—Bedeutung für die Stellung der Psammodrilida (Annelida). Microfauna Mar 11: 87–113 Meyer-Reil L-A (1994) Microbial life in sedimentary biofilms—the challenge to microbial ecologists. Mar Ecol Prog Ser 112: 303–311 Meyer-Reil L-A, Faubel A (1980) Uptake of organic matter by meiofauna organisms and interrelationships with bacteria. Mar Ecol Prog Ser 3: 251–256 Meyer-Reil L-A, Dawson R, Liebezeit G, Tiedge H (1978) Fluctuations and interactions of bacterial activity in sandy beach sediments and overlying waters. Mar Biol 48: 161–171 Meyers MB, Fossing H, Powell EN (1987) Microdistribution of interstitial meiofauna, oxygen and sulfide gradients, and the tubes of macro-infauna. Mar Ecol Prog Ser 35: 223–241 Meyers MB, Powell EN, Fossing H (1988) Movement of oxybiotic and thiobiotic meiofauna in response to changes in pore-water oxygen and sulfide gradients around macro-infaunal tubes. Mar Biol 98: 395–414 Meysman FJR, Middelburg JJ, Heip CHR (2006a) Bioturbation: a fresh look at Darwin’s last idea. Trends Ecol Evolut 21: 688–695 Meysman FJR, Galaktionov OS, Gribsholt B, Middelburg JJ (2006b) Bio-irrigation in permeable sediments: advective pore water transport induced by burrow ventilation. Limnol Oceanogr 51: 142–156 Michelson AR, Jacobson ME, Scranton MI, Mackin JE (1989) Modeling the distribution of acetate in anoxic estuarine sediments. Limnol Oceanogr 34: 747–757 Michiels IC, Traunspurger W (2005) Impact of resource availability on species composition and diversity in freshwater nematodes. Oecologia 142: 98–103
468
References
Middelburg JJ, Barranguet C, Boschker HTS, Herman PMJ, Moens T, Heip CH (2000) The fate of intertidal microphytobenthos carbon: an in situ 13C-labeling study. Limnol Oceanogr 45: 1224–1234 Mikryukov KA (1994) Marine and brackish-water centrohelid Heliozoa (Centroheliozoa; Sarcodina) of Kandalaksha Bay of the White Sea (in Russian, English summary). Zool Zhurn 73: 5–17 Mikryukov KA (2001) Heliozoa as a component of marine microbenthos: a study of Heliozoa of the White Sea Ophelia 54: 51–73 Miljutin DM, Tchesunov AV, Hope DW (2006) Rhaptothyreus typicus Hope & Murphy, 1969 (Nematoda: Rhaptothyreidae): an anatomical study of an unusual deep-sea nematode. Nematology 8: 1–20 Miljutina M, Miljutin D, Mahatma R, Martinez Arbizu P (2006) Structure of nematode communities from a deep-sea polymetallic nodule site. In: Tyler PA (ed) Abstr 11th Int Deep-Sea Biol Symp Southampton, p. 63 Miller DC, Geider RJ, MacIntyre HL (1996) Microphytobenthos: the ecological role f the “Secret Garden” of unvegetated, shallow-water marine habitats. II. Role in sediment stability and shallow-water food webs. Estuaries 19: 202–212 Millward RN, Grant A (1995) Assessing the impact of copper on nematode communities from a chronically metal-enriched estuary using pollution induced community tolerance. Mar Pollut Bull 30: 701–706 Mirto S, Danovaro R (2004) Meiofaunal colonization on artificial substrates: a tool for biomonitoring the environmental quality of coastal marine systems. Mar Pollut Bull 48: 919–926 Modig H, Olafsson E (1998) Responses of Balthic benthic invertebrates to hypoxic events. J Exp Mar Biol Ecol 229: 133–148 Modig H, Van den Bund WJ, Ólafsson E (2000) Uptake of phytodetritus by three ostracod species from the Baltic Sea: effects of amphipod disturbance and ostracod density. Mar Ecol Prog Ser 202: 125–134 Moens T, Vincx M (1997a) Observations on the feeding ecology of estuarine nematodes. J Mar Biol Ass UK 77: 211–227 Moens T, Vincx M (1997b) A state of the art on meiofaunal respiration and production. In: Baeyens J, Dehairs F, Goeyens L (eds) Second network meeting of the European network for integrated marine system analysis, FWO-Vlaanderen. VUB-University Press, Brussels, pp. 347–361 Moens T, Vincx M (1998) On the cultivation of free-living marine and estuarine nematodes. Helgol Meeresunters 52: 115–139 Moens T, Vincx M (2000a) Temperature and salinity constraints on the life cycle of two brackishwater nematode species. J Exp Mar Biol Ecol 243: 115–135 Moens T, Vincx M (2000b) Temperature, salinity and food thresholds in two brackish-water bacterivorous nematode species: assessing niches from food absorption and respiration experiments. J Exp Mar Biol Ecol 243: 137–154 Moens T, Vierstraete A, Vanhove S, Verbeke M, Vincx M (1996) A handy method for measuring meiobenthic respiration. J Exp Mar Biol Ecol 197: 177–190 Moens T, Gansbeke DV, Vincx M (1999a) Linking estuarine nematodes to their suspected food. A case study form the Westerschelde estuary (south-west Netherlands). J Mar Biol Ass UK 79: 1017–1027 Moens T, Verbeeck L, Maeyer AD, Swings J, Vincx M (1999b) Selective attraction of marine bacterivorous nematodes to their bacterial food. Mar Ecol Prog Ser 176: 165–178 Moens T, Verbeeck L, Vincx M (1999c) Feeding biology of a predatory and a facultatively predatory nematode (Enoploides longispiculosus and Adoncholaimus fuscus). Mar Biol 134: 585–593 Moens T, Dos Santos GAP, Thompson F, Swings J, Fonseca-Genevois V, Vincx M, De Mesel I (2005) Do nematode mucus secretions affect bacterial growth?. Aquat Microb Ecol 40: 77–83 Moens T, Traunspurger W, Bergtold M (2006) Feeding ecology of free-living benthic nematodes. In: Eyualem-Abebe E, Andrássy I, Traunspurger W (eds) Freshwater nematodes. Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, UK., pp. 105–131
References
469
Mokievskiy VO, Udalov AA, Azovskii AL (2007) Quantitative distribution of meiobenthos in deep-water zones of the world ocean Oceanology 47: 787–813 Mokievskiy V, Azovsky A (2002) Re-evaluation of species diversity patterns of free-living marine nematodes. Mar Ecol Prog Ser 238: 101–108 Mokievskiy VO, Kamenskaya OE (2002) Meiofauna of hydrothermal zones and other reducing biotopes. In: Gebruk AV (ed) Biology of hydrothermal systems. KMK Press, Moscow, pp. 237–253 Monaghan E, Giblin AE (1994) The effects of coupling between the oxic and anoxic layers of sediment on nutrient release to overlying water. Biol Bull 187: 288–289. Monniot C, Monniot F (1990) Revision of the class Sorberacea (benthic tunicates) with descriptions of 7 new species. Zool J Linn Soc 99: 239–290 Monniot C, Monniot F, Gaill F (1975) Les Sorberacea: une nouvelle classe de Tuniciers. Arch Zool Expér Gén 116: 77–122 Monniot F (1965) Ascidies interstitielles des côtes d’Europe. Mém Mus Nat Hist Nat Paris, Sér A 35: 1–154+110 tab Monniot F (1966) Un Palpigrade interstitiel Leptokoenenia scurra n.sp. Rev Ecol Biol Sol 3: 41–64 Monniot F, Monniot C (1988) Tunicata. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 461–464 Monod T (1940) Thermosbaenacea. Bronns Klassen und Ordnungen des Tierreichs, Bd. 5, Abt. 1, Buch 4, T. 4. Akad Verlagsges, Leipzig, pp. 1–24 Montagna PA (1983) Live controls for radioisotope food chain experiments using meiofauna. Mar Ecol Prog Ser 12: 43–46 Montagna PA (1984) Competition for dissolved glucose between meiobenthos and sediment microbes. J Exp Mar Biol Ecol 76: 177–190 Montagna PA (1995) Rates of metazoan meiofaunal microbivory: A review. Vie Milieu 45: 1–9 Montagna PA, Bauer JE (1988) Partitioning radiolabeled thymidine uptake by bacteria and meiofauna using metabolic blocks and poisons in benthic feeding studies. Mar Biol 98: 101–110. Montagna PA, Coull BC, Herring TL, Dudley BW (1983) The relationship between abundances of meiofauna and their suspected microbial food (diatoms and bacteria). Estuar Coast Shelf Sci 17: 381–394 Montagna PA, Bauer JE, Hardin D, Spies RB (1989) Vertical distribution of microbial and meiofauna populations in sediments of a natural coastal hydrocarbon seep. J Mar Res 47: 657–680 Montagna P, Bauer JE, Hardin D, Spies RB (1995) Meiofaunal and microbial trophic interactions in a natural submarine hydrocarbon seep. Vie Milieu 45: 17–25 Moodley L, Zwaan GJ van der, Herman PMJ, Kempers L, Breugel P van (1997) Differential response of benthic meiofauna to anoxia with special reference to Foraminifera (Protista: Sarcodina). Mar Ecol Prog Ser 158: 151–163 Moodley L, Schaub BEM, Zwaan GJ van der, Herman PMJ (1998a) Tolerance of benthic Foraminifera (Protista: Sarcodina) to hydrogen sulphide. Mar Ecol Prog Ser 169: 77–86 Moodley L, Zwaan GJ van der, Rutten GMW, Boom RCE, Kempers AJ (1998b) Subsurface activity of benthic Foraminifera in relation to porewater oxygen content: laboratory experiments. Mar Micropalaeontol 34: 91–106 Moodley L, Boschker HTS, Middelburg JJ, Herman PMJ, De Deckere E, Heip CHR (2000) The ecological significance of benthic Foraminifera: 13C labelling experiments. Mar Ecol Prog Ser 202: 289–295 Moodley L, Middelburg JJ, Boschker HTS, Duineveld GCA, Pel R, Herman PMJ, Heip CHR (2002) Bacteria and Foraminifera: keyplayers in a short-term deep-sea benthic response to phytodetritus. Mar Ecol Prog Ser 236: 23–29 Moodley L, Steyaert M, Epping E, Middelburg JJ, Vincx M, Avesaath P van, Moens T, Soetaert K (2008) Biomass-specific respriation rates of benthic meiofauna: demonstrating a novel oxygen micro-respiration system. J Exp Mar Biol Ecol 357: 41–47 Moore CG, Stevenson JM (1997) A possible new meiofaunal tool for rapid assessment of the environmental impact of marine oil pollution. Cah Biol Mar 38: 277–282
470
References
Moore HB (1931) The muds of the Clyde Sea area. III. Chemical and physiological conditions, rate and nature of sedimentation, and fauna. J Mar Biol Ass UK 17: 325–358 Moore HB, Neill RG (1930) An instrument for sampling marine muds. J Mar Biol Ass UK 16: 589–594 Moreno M, Ferrero TJ, Granelli V, Marin V, Albertelli G, Fabiano M (2006) Across-shore variabilty and trophodynamic features of meiofauna in a microtidal beach of the NW Mediterranean. Estuar Coast Shelf Sci 66: 357–367 Morgan NC et al. (1980) Secondary production. In: Le Cren ED, Lowe-McConnell RH (eds) The functioning of freshwater ecosystems. Cambridge University Press, Cambridge, pp. 247–340 Moriarty DJW (1980) Measurement of bacterial biomass in sandy sediments. Biogeochemistry of ancient and modern environments. Proceedings of International Symposium on Environmental Biogeochemistry (ISEB), Berlin 4: pp. 131–138 Moriarty DJW, Pollard PC, Hunt WG, Moriarty CM, Wassenberg TJ (1985) Productivity of bacteria and microalgae and the effect of grazing by holothurians in sediments on a coral reef flat. Mar Biol 85: 293–300 Morill AC, Powell EN, Bidigare RR, Shick JM (1988) Adaptations to life in the sulfide system: a comparison of oxygen detoxifying enzymes in thiobiotic and detoxifying enzymes in thiobiotic and oxybiotic meiofauna (and freshwater planarians). J Comp Phys B 158: 335–344 Morris JT, Coull BC (1992) Population dynamics, numerical production, and potential predation impact on a meiobenthic copepod. Can J Fish Aquat Sci 49: 609–616 Morse MP (1981) Meiopriapulus fijiensis n.gen., n.sp.: an interstitial priapulid from coarse sand in Fiji. Trans Amer Micros Soc 100: 239–252 Morse MP (1987) Distribution and ecological adaptations of interstitial molluscs in Fiji. Amer Malacol Bull 5: 281–286 Mortensen T (1925) An apparatus for catching the micro-fauna of the sea bottom. Vidensk Medd Dansk Naturh Foren København 80: 445–451 Müller KJ, Walossek D (1985) Skaracarida, a new order of Crustacea from the Upper Cambrian of Västergötland, Sweden. Fossil Strata 17: 1–65 Müller KJ, Walossek D (1991) Ein Blick durch das
-Fenster in die Arthropodenwelt vor 500 Millionen Jahren. Verh Dtsch Zool Ges 84 B: 281–294 Müller KJ, Walossek D, Zakharov A (1995) ‘Orsten’ type phosphatized soft-integument preservation and a new record from the Middle Cambrian Kuonamka formation in Siberia. N Jb Geol Paläont Abh 197: 101–118 Müller MCM (2006) Polychaete nervous systems: ground pattern and variations—cLS microscopy and the importance of novel characteristics in phylogenetic analysis. Integr Compar Biol 46: 125–133 Müller U, Ax P (1971) Gnathostomulida von der Nordseeinsel Sylt mit Beobachtungen zur Lebensweise und Entwicklung von Gnathostomula paradoxa Ax. Mikrofauna Meeresboden 9: 1–41 Mullineaux LS (1987) Organisms living on manganese nodules and crusts: distribution and abundance at three North Pacific sites. Deep-Sea Res 34: 165–184 Munro ALS, Wells JBJ, Mc Intyre AD (1978) Energy flow in the flora and meiofauna of sandy beaches. Proc R Soc Edinb 76 B: 297–315 Muralikrishnamurty PV (1993) Intertidal phytal fauna off Gangavaram, east coast of India. Indian J mar Sci 12: 85–89 Murray JMH, Meadows A, Meadows PS (2002) Biogeomorphological implications of microscale interactions between sediment geotechnics and marine benthos: a review. Geomorphology 47: 15–30 Murray JW (1973) Distribution and ecology of living benthic foraminiferids. Heinemann Educational Books, London Murray JW (1979) British nearshore foraminiferids. Keys and notes for the identification of the species. In: Kermack DM, Barnes RSK (eds) Synopses of the British fauna. Academic, London, pp. 68 Murrell MC, Fleeger JW (1989) Meiofauna abundance on the Gulf of Mexico continental shelf affected by hypoxia. Continent Shelf Res 9: 1049–1062
References
471
Musat N, Giere O, Gieseke A, Thiermann F, Amann R, Dubilier N (2007) Molecular and morphological characterization of the association between bacterial endosymbionts and the marine nematode Astomonema sp. from the Bahamas. Environ Microbiol 9: 1345–1353 Muschiol D, Traunspurger W (2007) Life cycle and calculation of the intrinsic rate of natural increase of two bacterivorous nematodes, Panagrolaimus sp. and Poikilolaimus sp. from chemoautotrophic Movile Cave, Romania. Nematology 9: 271–284 Muschiol D, Traunspurger W (2008) Life at the extreme: meiofauna from three unexplored lakes in the caldera of volcano Cerro Azul, Galápagos Islands, Ecuador. Aquatic Ecology (accepted) Muschiol D, Markovic M, Threis I, Traunspurger W (2008a) Predatory copepods can control nematode populations: A functional-response experiment with Eucyclops subterraneus and bacterivorous nematodes Fundam Appl Limnol (in press) Muus B (1964) A new quantitative sampler for the meiobenthos. Ophelia 1: 209–216 Muus B (1968) A field method for measuring “exposure” by means of plaster balls. A preliminary account. Sarsia 34: 61–68 Myers RF (1967) Osmoregulation in Pangrellus redivivus and Aphelenchus avanae. Nematologica 12: 579–586 Neel JK (1948) A limnological investigation of the psammon in Douglas Lake, Michigan, with especial reference to shoal and shoreline dynamics. Trans Amer Micros Soc 67: 1–53 Neher DA, Darby BJ (2006) Computation and application of nematode community indices: general guidelines. In: Eyualem-Abebe E, Andrássy I, Traunspurger W (eds) Freshwater nematodes: Ecology and taxonomy. CABI Publishing, Wallingford, Oxfordshire, UK, pp. 211–222 Nehring S, Jensen P, Lorenzen S (1990) Tube-dwelling nematodes: tube construction and possible ecological effects on sediment-water interfaces. Mar Ecol Prog Ser 64: 123–128 Neira C, Rackemann M (1996) Black spots produced by buried macroalgae in intertidal sandy sediments of the Wadden Sea: effects on the meiobenthos. J Sea Res 36: 153–170 Neira C, Gad G, Arroyo NL, Decraemer W (2001a) Glochinema bathyperuvensis sp.n. (Nematoda, Epsilonematidae): a new species from Peruvian bathyal sediments, SE Pacific Ocean. Contr Zool 70: 147–159 Neira C, Sellanes J, Levin LA, Arntz WE (2001b) Meiofaunal distribution on the Peru margin: relationship to oxygen and organic matter availability. Deep-Sea Res I 48: 2453–2472 Nelson AL, Coull BC (1989) Selection of meiobenthic prey by juvenile spot (Pisces): an experimental study. Mar Ecol Prog Ser 53: 51–57 Nelson DR (2002) Current status of the Tardigrada: evolution and ecology. Integr Comp Biol 42: 652–659 Nelson DR, Higgins RP (1990) Tardigrada. In: Dindal DL (ed) Soil biology guide. Wiley, New York, pp. 393–419 Netto SA, Warwick RM, Attrill MJ (1999) Meiobenthic and macrobenthic community structure in carbonate sediments of Rocas Atoll (north-east Brazil). Estuar Coast Shelf Sci 48: 39–50 Neu TR, Kuhlicke U, Lawrence JR (2002) Assessment of fluorochromes for Two-Photon Laser Scanning Microscopy of Biofilms. Appl Environ Microbiol 68: 901–909 Neuhaus B (1994) Ultrastructure of alimentary canal and body cavity, ground pattern, and phylogenetic relationships of the Kinorhyncha. Microfauna Mar 9: 61–156. Neuhaus B, Higgins RP (2002) Ultrastructure, biology, and phylogenetic relationships of Kinorhyncha. Integr Comp Biol 42: 619–632 Neusser TP, Heß M, Haszprunar G, Schrödl M (2006) Computer-based three-dimensional reconstruction of the anatomy of Microhedyle remanei (Marcus, 1953), an interstitial acochlidian gastropod from Bermuda. J Morphol 267: 231–247 Nicholas WL (1984) The biology of free-living nematodes. Clarendon, Oxford, p. 219 Nicholas WL, Elek JA, Stewart AC, Marples TG (1991) The nematode fauna of a temperate Australian mangrove mudflat—its population density, diversity and distribution. Hydrobiologia 209: 13–28
472
References
Nicholas WL, Goodchild DJ, Stewart A (1987) The mineral composition of intracellular inclusions in nematodes from thiobiotic mangrove mud-flats. Nematologica 33: 167–179 Nicholls AG (1935) Copepods from the interstitial fauna of a sandy beach. J Mar Biol Ass UK 20: 379–405 Nichols JA (1979) A simple flotation technique for separating meiobenthic nematodes from finegraded sediments. Trans Am Microsc Soc 98: 127–130 Nielsen C (1994) Larval and adult characters in animal phylogeny. Am Zool 34: 492–501 Nielsen C (2001) Animal evolution: interrelationships of the living phyla. 2nd ed, Oxford University Press, Oxford, p. 563 Nilsson P, Jönsson B, Lindström-Swanberg I, Sundbäck K (1991) Response of a marine shallowwater sediment system to an increased load of inorganic nutrients. Mar Ecol Prog Ser 71: 275–290 Nilsson P, Sundbäck K, Jönsson B (1993) Effect of the brown shrimp Crangon crangon L. on endobenthic macrofauna, meiofauna and meiofaunal grazing rates. Neth J Sea Res 31: 95–106. Nixon SW (1995) Coastal marine eutrophication: a definition, social causes, and future concerns. Ophelia 41: 199–219 Nodot C (1978) Cycle biologiques de quelques espèces de copépodes harpacticoides psammiques. Téthys 8: 241–248 Noldt U, Wehrenberg C (1984) Quantitative extraction of living Plathelminthes from marine sands. Mar Ecol Prog Ser 20: 193–201 Noodt W (1965) Natürliches System und Biogeographie der Syncarida (Crustacea, Malacostraca). Gewäss Abwäss 37/38: 77–186 Noodt W (1971) Ecology of the Copepoda. Smiths Contr Zool 76: 97–102 Nordheim H von (1984) Life histories of subtidal interstitial polychaetes of the family Polygordidae, Protodrilidae, Nerillidae, Dinophilidae and Diurodrilidae from Helgoland (North Sea). Helgol Meeresunters 38: 1–20 Nordheim H von (1989) Six new species of Protodrilus (Annelida, Polychaeta) from Europe and New Zealand, with a concise presentation of the genus. Zool Scr 18: 245–268 Norenburg JL (1988) Remarks on marine interstitial nemertines and key to the species. Hydrobiologia 156: 87–92 Norenburg JL, Morse MP (1983) Systematic implications of Euphysa ruthae n.sp. (Athecata: Corymorphidae), a psammophilic solitary hydroid with unusual morphogenesis. Trans Am Microsc Soc 102: 1–17 Novak R (1989) Ecology of nematodes in the Mediterranean seagrass Posidonia oceanica (L.) Delile. 1. General part and faunistics of the nematode community. Mar Ecol PSZN 10: 335–363 Novak R (1992) Dynamic aspects of seagrass-nematode community structure. In: Colombo G, Ferrari I, Ceccherelli VU, Rossi R (eds) Marine eutrophication and population dynamics, with a special section on the Adriatic Sea. Olsen & Olsen, Fredensborg, Denmark, pp. 277–284 Novitsky JA (1983) Heterotrophic activity throughout a vertical profile of seawater and sediment in Halifax Harbor, Canada. Appl Environ Microbiol 45: 1753–1760 Nozais C, Gosselin M, Michel C, Tita G (2001) Abundance, biomass, composition and grazing impact of the sea-ice meiofauna in the North Water, northern Baffin Bay. Mar Ecol Prog Ser 217: 235–250 Nozais C, Perissiniotto R, Tita G (2005) Seasonal dynamics of meiofauna in a South African temporarily open/closed estuary (Mdloti Estuary, Indian Ocean). Estuar Coast Shelf Sci 62: 325–338 Nuß B, Trimkowski V (1984) Physikalische Mikroanalysen an kristalloiden Einschlüssen bei Tobrilus gracilis (Nematoda, Enoplida). Veröff Inst Meeresforsch Bremerh 20: 17–27 Ockelmann KW (1964) An improved detritus-sledge for collecting meiobenthos. Ophelia 1: 217–222 Oden BJ (1979) The freshwater littoral meiofauna in a South Carolina reservoir receiving thermal effluents. Freshw Biol 9: 291–304 Oguri K, Kitazato H, Glud RN (2006) Platinum octaetylporphyrin based planar optodes combined with an UV-LED excitation light source: an ideal tool for high-resolution O2 imaging in O2depleted envrinments. Mar Chem 100: 95–107
References
473
Ólafsson E (1991) Intertidal meiofauna of four sandy beaches in Iceland. Ophelia 33: 55–65 Ólafsson E (1992) Small-scale spatial distribution of marine meiobenthos: the effects of decaying macrofauna. Oecologica 90: 37–42 Ólafsson E (1995) Meiobenthos in mangrove areas in estern Africa with emphasis on assemblage structure of free-living marine nematodes. Hydrobiologia 312: 47–57 Ólafsson E (2003) Do macrofauna structure meiofauna assemblages in marine soft-bottoms? A review of experimental studies. Vie Milieu 53: 249–265 Ólafsson E, Elmgren R (1997) Seasonal dynamics of sublittoral meiobenthos in relation to phytoplankton sedimentation in the Baltic Sea. Estuar Coast Shelf Sci 45: 149–164 Ólafsson E, Moore CG (1992) Effects of macroepifauna on developing nematode and harpacticoid assemblages in a subtidal muddy habitat. Mar Ecol Prog Ser 84: 161–171 Ólafsson E, Ndaro SGM (1997) Impact of the mangrove crabs Uca annulipes and Dotilla fenestrata on meiobenthos. Mar Ecol Prog Ser 158: 225–231 Ólafsson E, Modig H, Bund WJ van den (1999) Species specific uptake of radio-labelled phytodetritus by benthic meiofauna from the Baltic Sea. Mar Ecol Prog Ser 177: 63–72 Ólafsson E, Ingólfsson A, Steinarsdottir MB (2001) Harpacticoid copepod communities of floating seaweed: controlling factors and implications for dispersal. Hydrobiologia 453/454: 189–200 Ólafsson E, Ullberg J, Arroyo NL (2005) The clam Macoma balthica prevents in situ growth of microalgal mats: implications for meiofaunal assemblages. Mar Ecol Prog Ser 298: 179–188 Olsen DA, Townsend CR (2005) Flood effects on invertebrates, sediments and particulate organic matter in the hyporheic zone of a gravel-bed stream. Freshw Biol 50: 839–853 Olu K, Lance S, Sibuet M, Henry P, Fiala-Médioni A, Dinet A (1997) Cold seep communities as indicators of fluid expulsion patterns through mud volcanoes seaward of the Barbados accretionary prism. Deep-Sea Res I 44: 811–841 Orghidan T (1955) Ein neuer Lebensraum des unterirdischen Wassers: Der hyporheische Biotop. Arch Hydrobiol 55: 392–414 Osenga GA, Coull BC (1983) Spartina alterniflora Loisel root structure and meiofaunal abundance. J Exp Mar Biol Ecol 67: 221–225 Ott J (1972) Determination of fauna boundaries of nematodes in an intertidal sand flat. Int Rev gesamt Hydrobiol 57: 645–663 Ott J (1993) Sulphide symbioses in shallow sands. In: Eleftheriou A, Ansell AD, Smith CJ (eds) Biology and ecology of shallow coastal waters. 28th European Marine Biology Symposium. Olsen and Olsen, Fredensborg, Denmark, pp. 143–148 Ott J, Novak R (1989) Living at an interface: meiofauna at the oxygten/sulfide boundary of marine sediments. In: Ryland JS, Tyler PA (eds) 23rd European Marine Biology Symposium. Olsen and Olsen, Fredenborg, Denmark, pp. 415–422 Ott J, Novak R, Schiemer F, Hentschel U, Nebelsick M, Polz M (1991) Tackling the sulfide gradient: a novel strategy involving marine nematodes and chemoautotrophic ectosymbionts. Mar Ecol PSZN 12: 261–279 Ott J, Bright M, Bulgheresi S (2003) Symbioses between marine nematodes and sulfur-oxidizing chemoautotrophic bacteria. Symbiosis 36: 103–126 Ott J, Bright M, Bulgheresi S (2004) Marine microbial thiotrophic ectosymbioses. Oceanogr Mar Biol Ann Rev 42: 95–118 Pace MC, Carman KR (1996) Interspecific differences among meiobenthic copepods in the use of microalgal food resources. Mar Ecol Prog Ser 143: 77–86 Packer M, Blenkin S, Floyd R, Abebe E, Angel P, Baldwin J, Cook A, Creer S, Lunt DH, Rogers AD, Thomas K, Lambshead PJD (2006) Development and uses of a deep-sea nematode morphological and DNA barcode database. In: Tyler PA (ed) Abstr 11th Int Deep-Sea Biol Symp, Southampton, p. 145 Page HG (1955) Phi-millimeter conversion table. J Sed Petrol 25: 285–292 Pallo P, Widbom B, Ólafsson E (1998) A quantitative survey of the benthic meiofauna in the Gulf of Riga (Eastern Baltic Sea), with special reference to the structure of nematode assemblages. Ophelia 49: 117–139
474
References
Palmer MA (1988) Dispersal of marine meiofauna: a review and conceptual model explaining passive transport and active emergence with implications for recruitment. Mar Ecol Prog Ser 48: 81–91 Palmer MA (1990a) Temporal and spatial dynamics of meiofauna within the hyporheic zone of Goose Creek, Virginia. J N Am Benthol Soc 9: 17–25 Palmer MA (1990b) Understanding the movement dynamics of a stream-dwelling meiofauna community using marine analogs. Stygologia 5: 67–74 Palmer MA (1992) Incorporating lotic meiofauna into our understanding of faunal transport processes. Limnol Oceanogr 37: 329–341 Palmer MA (1993) Experimentation in the hyporheic zone: challenges and prospectus. J N Am Benthol Soc 12: 84–93 Palmer MA, Gust G (1985) Dispersal of meiofauna in a turbulent tidal creek. J Mar Res 43: 179–210 Palmer MA, Strayer DL (1996) Meiofauna. In: Hauer RR, Lamberti GA (eds) Methods in stream ecology. Elsevier, Amsterdam, pp. 315–337 Palmer MA, Montagna PA, Spies RB, Harding D (1988) Meiofauna dispersal near natural petroleum seeps in the Santa Barbara Channel: a recolonization experiment. Oil Chem Pollut 4: 179–189 Palmer MA, Bely AE, Berg KE (1992) Response of invertebrates to lotic disturbance: a test of the hyporheic refuge hypothesis. Oecologia 89: 182–194 Palmer MA, Arensburger P, Botts PS, Hakenkamp CC, Reid JW (1995) Disturbance and the community structure of stream invertebrates: patch-specific effects and the role of refugia. Freshw Biol 34: 343–356 Palmer MA, Allan JD, Butman CA (1996) Dispersal as a regional process affecting the local dynamics of marine and stream benthic invertebrates. Trends Ecol Evol 11: 322–326 Palmer MA, Strayer DL, Rundle SD (2006) Meiofauna. In: Hauer RR, Lamberti GA (eds) Methods in stream ecology. Academic, Burlington, pp. 415–434 Park JK, Rho HS, Kristensen RM, Kim W, Giribet G (2006) First molecular data on the phylum Loricifera—An investigation into the phylogeny of Ecdysozoa with emphasis on the positions of Loricifera and Priapulida. Zool Sci 23: 943–954 Parkinson J et al. (2004) A transcriptomic analysis of the phylum Nematoda. Nat Genet 36: 1259–1267 Parry GD (1981) The meanings of r- and K-selection. Oecologia (Berlin) 48: 260–264 Pascal P-Y, Dupuy C, Richard P, Rzeznik-Orignac J, Niquil N (2008 ) Bacterivory of a mudflat nematode community under different environmental conditions. Mar Biol 154: 671–682 Passow U (2000) Formation of transparent exopolymer particles, TEP, from dissolved precursor material. Mar Ecol Prog Ser 192: 1–11 Patterson DJ (1996) Free-living Freshwater Protozoa. CSIRO Publishing, Collingwood, VIC, Australia, p. 224 Patterson DJ, Larsen J, Corliss JO (1989) The ecology of the heterotrophic flagellates and ciliates living in marine sediments. Progr Protistol 3: 185–277 Pawlowski J (2000) Introduction to the molecular systematics of Foraminifera. Micropalaeont 46, Suppl. 1: 1–12 Pawlowski J, Bolivar I, Guiard-Maffia J, Gouy M (1994) Phylogenetic position of Foraminifera inferred from LSU rRNA gene sequences. Mol Biol Evolut 11: 929–938. Pawlowski J, Fahrni JF, Guiard J, Conlan K, Hardecker J, Habura A, Bowser SS (2005) Allogromiid foraminifera and gromiids from under the Ross Ice Shelf: morphological and molecular diversity. Polar Biol 28: 514–522 Pawlowski J, Fahrni J, Lecroq B, Longet D, Cornelius N, Cedhagen T, Excoffier L, Gooday A (2007) Bipolar gene flow in deep-sea benthic foraminifera. Mol Ecol 16: 4089–4096 Peachey RL, Bell SS (1997) The effects of mucous tubes on the distribution, behavior and recruitment of seagrass meiofauna. J Exp Mar Biol Ecol 209: 279–292 Pearson TH (2001) Functional group ecology in soft-sediment marine benthos: the role of bioturbation. Oceanogr Mar Biol Ann Rev 39: 233–267
References
475
Peck LS, Uglow RF (1990) Two methods for the assessment of the oxygen content of small volumes of seawater. J Exp Mar Biol Ecol 141: 53–62 Pennak RW (1939) The microscopic fauna of the sandy beaches. Problems of lake biology. Publ Amer Ass Adv Sci 10: 94–106 Pennak RW (1940) Ecology of the microscopic Metazoa inhabiting the sandy beaches of some Wisconsin lakes. Ecol Monogr 10: 537–615 Pennak RW (1951) Comparative ecology of the interstitial fauna of fresh-water and marine beaches. Année Biol Sér 3 27: 217–248 Pennak RW (1988) Ecology of the freshwater meiofauna. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp. 39–60 Pennak RW, Ward JV (1985) Bathynellacea (Crustacea: Syncarida) in the United States, and a new species from the phreatic zone of a Colorado mountain stream. Trans Am Microsc Soc 104: 209–215 Pennak RW, Ward JV (1986) Interstitial fauna communities of the hyporheic and adjacent groundwater biotopes of a Colorado mountain stream. Arch Hydrobiol Suppl. 74: 356–396 Pennak RW, Zinn DJ (1943) Mystacocarida, a new order of Crustacea from intertidal beaches in Massachussetts and Connecticut. Smiths Misc Collect 103: 1–11 Perry RN, Wright DJ (1998) The physiology and biochemistry of free-living and plant-parasitic nematodes. CABI Publishing, Wallingford, Oxfordshire, UK, pp. 438 Peter R, Ladurner P, Rieger RM (2001) The role of stem cell strategies in coping with environmental stress and choosing between alternative reproductive modes: Turbellaria rely on a single cell type to maintain individual life and propagate species. Mar Ecol PSZN 22: 35–51 Peters L, Wetzel MA, Traunspurger W, Rothaupt K (2007) Epilithic communities in a lake littoral zone: the role of water-column transport and habitat development for dispersal and colonization of meiofauna. J N Am Benthol Soc 26: 232–243 Peters RH (1983) Ecological implications of body size. Cambridge University Press, Cambridge, p. 329 Petersen CGJ (1913) Havets bonitering. II. Om havbundens dyresamfund og om disses betydning for den marine zoogeografi. Beretn Landbrugsminist Dansk Biolog Station 21, 1–42 plus add 41–68 Petersen S, Arlt G, Faubel A, Carman KR (1998) On the nutritive significance of dissolved free amino acids uptake for the cosmopolitan oligochaete Nais elinguis Müller (Naididae). Estuar Coast Shelf Sci 46: 85–91 Peterson KJ, Eernisse DJ (2001) Animal phylogeny and the ancestry of bilaterians: inferences from morphology and 18SrDNA gene sequences. Evol Develop 3: 170–205 Petrov NB, Vladychenskaya NS (2005) Phylogeny of molting protostomes (Ecdysozoa) as inferred from 18S and 28S rRNA gene sequences. Mol Biol 39: 503–513 Pfannenstiel H-D (1981) Endocrine control of sexual differentiation in the protandric polychaete, Ophryotrocha puerilis. In: Clark WH, Adams TS (eds) Advances in invertebrate reproduction. Elsevier, North Holland, 332 (only) Pfannkuche O (1992) Organic carbon flux through the benthic community in the temperate abyssal Northeast Atlantic. In: Rowe GT, Pariente V. (eds) Deep-Sea food chains and the global carbon cycle Kluwer, Dordrecht, pp. 183–198 Pfannkuche O, Lochte K (1990) Metabolismus und Energiefluß im Benthal. Biotrans Biologischer Vertikaltransport und Energiehaushalt in der bodennahen Wasserschicht der Tiefsee. Ber Zentr Meeres- Klimaforsch Univ Hamb 10: 130–154 Pfannkuche O, Thiel H (1987) Meiobenthic stocks and benthic activity on the NE-Svalbard shelf and in the Nansen-Basin. Polar Biol 7: 253–266 Pfannkuche O, Thiel H (1988) Sample processing. In: Higgins RP, Thiel H (eds) Introduction to the study of meiobenthos.. Smithsonian Institution Press, Washington, DC, pp. 134–145 Pfeifer D, Bäumer HP, Ortleb H, Sach G, Schleier U (1996) Modelling spatial distributional patterns of benthic meiofauna species by Thomas and related processes. Ecol Modelling 87: 285–294 Philippe H, Lartillot N, Brinkmann H (2005) Multigene analyses of bilaterian animals corroborate the monophyly of Ecdysozoa, Lophotrochozoa, and Protostomia. Mol Biol Evol 22: 1246–1253 Pianka ER (1966) Latitudinal grandients in species diversity—a review of concepts. Amer Nat 100: 33–46
476
References
Pianka ER (1989) Latitudinal gradients in species diversity. Trends Ecol Evolut 4: 223 Pihl L (1985) Food selection and consumption of mobile epibenthic fauna in shallow marine areas. Mar Ecol Prog Ser 22: 169–179 Pike JB, Bernhard JM, Moreton SG, Butler IB (2001) Microirrigation of marine sediments in dysoxic environments: implications for early sediment fabric formation and diagenetic processes. Geology 29: 923–926 Pinckney JL, Carman KR, Lumsden SE, Hymel SN (2003) Microalgal-meiofaunal trophic relationships in muddy intertidal estuarine sediments. Aquat Microb Ecol 31: 99–108 Pitcher TJ, Morato T, Hart PJB, Clark MR, Haggan N, Santos RS (eds) (2007) Sea Mounts: ecology, fisheries and conservation. Fish and Aquatic Resources Series 12. Blackwell Publ, Oxford, p. 527 Plante C, Downing JA (1989) Production of freshwater invertebrate populations in lakes. Can J Fish Aquat Sci 46: 1489–1498 Platt HM, Shaw KM, Lambshead PJD (1984) Nematode species abundance patterns and their use in the detection of environmental perturbations. Hydrobiologia 118: 59–66 Platt HM, Warwick RM (1980) The significance of free-living nematodes to the littoral ecosystem. In: Price JH, Irvine DEG, Farnham WF (eds) The Shore environment. 2. Ecosystems. Academic, New York, pp. 729–759 Platt HM, Warwick RM (1983) Freeliving marine nematodes. Pt. 1. British enoplids. Pictorial key to world genera and notes for the identification of British species. Cambridge University Press, Cambridge, p. 307 (Synopses of the British Fauna, vol 28) Platt HM, Warwick RM (1988) Freeliving marine nematodes. Pt. 2. British chromadorids. Pictorial key to world genera and notes for the identification of British species. Brill/Backhuys, Leiden, p. 502 (Synopses of the British Fauna, vol 38) Poff NL, Palmer MA, Angermeier PL, Vadas RLJ, Hakenkamp C, Bely A, Arensburger P, Martin AP (1993) Size structure of metazoan community in a Piedmont stream. Oecologia 95: 202–209 Poff NL, Palmer MA, Angermeier PL, Vadas RLJ, Hakenkamp C, Bely A, Arensburger P, Martin AP (1993) Size structure of metazoan community in a Piedmont stream. Oecologia 95: 202–209 Pogrebov VB, Fokin SI, Gal’tsova VV, Ivanov GI (1997) Benthic communities as influenced by nuclear testing and radioactive waste disposal off Novaya Zemlya in the Russian Arctic. Mar Pollut Bull 35: 333–339 Poizat C (1985) Interstitial opisthobranch gastropods as indicator organisms in sublittoral sandy habitats. Stygologia 1: 26–42 Pollock LW (1970) Distribution and dynamics of interstitial Tardigrada at Woods Hole, Massachusetts, USA. Ophelia 7: 145–165 Por FD, Bromley HJ (1974) Morphology and anatomy of Maccabaeus tentaculatus (Priapulida: Seticoronaria). J Zool (London) 173: 197 Por FD, Masry D (1968) Survival of a nematode and an oligochaete species in the anaerobic benthal of Lake Tiberias. Oikos 19: 388–391 Potel P, Reise K (1987) Gastrotricha Macrodasyida of intertidal and subtidal sandy sediments in the northern Wadden Sea. Microfauna Mar 3: 363–376 Powell EN (1989) Oxygen, sulfide and diffusion: why thiobiotic meiofauna must be sulfide-insensitive first-order respirers. J Mar Res 47: 887–932 Powell EN, Bright TJ (1981) A thiobios does exist—Gnathostomulid domination of the Canyon community at the East Flower Garden brine seep. Int Rev Ges Hydrobiol 66: 675–683 Powell EN, Bright TJ, Woods A, Gittings S (1983) Meiofauna and the thiobios in the East Flower Garden brine seep. Mar Biol 73: 269–283 Powell MA, Arp AJ (1989) Hydrogen sulfide oxidation by abundant nonhemoglobin heme compounds in marine invertebrates from sulfide-rich habitats. J Exp Zool 249: 121–132 Powell MA, Somero GN (1986a) Adaptations to sulfide by hydrothermal vent animals: sites and mechanisms of detoxification and metabolism. Biol Bull 171: 274–290
References
477
Powell MA, Somero GN (1986b) Hydrogen sulfide oxidation is coupled to oxidative phosphorylation in mitochondria of Solemya reidi. Science 233: 563–566 Powell MA, Vetter RD, Somero GN (1987) Sulfide detoxification and energy exploitation by marine animals. In: Dejours P, Bolis L, Taylor CR, Weibel ER (eds) Comparative physiology: Life in water and on land. Liviana press, Padova, pp. 241–250 Powilleit M, Kitlar J, Graf G (1994) Particle and fluid bioturbation caused by the priapulid worm Halicryptus spinulosus (V. Seibold). Sarsia 79: 109–117 Precht E, Huettel M (2003) Advective pore-water exchange driven by surface gravity waves and its ecological implications. Limnol Oceanogr 48: 1674–1684 Precht E, Huettel M (2004) Rapid wave-driven porewater exchange in a permeable coastal sediment. J Sea Res 51: 93–107 Pugh PJA, King PE (1985a) Feeding in intertidal Acari. J Exp Mar Biol Ecol 94: 269–280 Pugh PJA, King PE (1985b) Vertical distribution and substrate association of the British Halacaridae. J Nat Hist 19: 961–968 Purschke G (1988) Anatomy and ultratructure of ventral pharyngeal organs and their phylogenetic importance in Polychaeta (Annelida). V. The pharynges of the Ctenodrilidae and Orbiniidae. Zoomorphology 108: 119–135 Purschke G (2002) On the ground pattern of Annelida. Organ Divers Evolut 2: 181–196 Queirago H, Cunha M, Cunha A, Moreira M, Quintino V, Rodrigues A, Serôdio J, Warwick RE (2006) Marine biodiversity. Contr 38th Europe Biol Symp, Aveiro, Portugal. Developments in Hydrobiology 183, Springer, Heidelberg, New York, p. 356 Quijon PA, Snelgrove PVR (2006) The use of coarser taxonomic resolution in studies of predation on marine sedimentary fauna. J Exp Mar Biol Ecol 330: 159–168 Rachor E (1975) Quantitative Untersuchungen über das Meiobenthos der nordostatlantischen Tiefsee. Meteor Forsch-Erg D 21: 1–10 Radziejewska T, Gruszka P, Rokicka-Praxmajer J (2006) A home away from home: a meiobenthic assemblage in a ship’s ballast water tank sediment. Oceanologia 48: 259–265 Radziejewska T (2002) Responses of deep-sea meiobenthic communities to sediment disturbance simulating effects of polymetallic nodule mining. Int Rev Hydrobiol 87: 457–477 Raes M, Vanreusel A (2006) Microhabitat type determines the composition of nematode communities associated with sediment-clogged cold-water coral framework in the Porcupine Seabight (NE Atlantic). Deep-Sea Res I 53: 1880–1894 Raffaeli D (1987) The behaviour of the nematode/copepod ratio in organic pollution studies. Mar Environ Res 23: 135–152 Raffaeli DG, Mason CF (1981) Pollution monitoring with meiofauna, using the ratio of nematodes to copepods. Mar Pollut Bull 12: 158–163 Ramsing NB, Kühl M, Jørgensen BB (1993) Distribution of sulfate-reducing bacteria, O2, and H2S in photosynthetic biofilms determined by oligonucleotide probes and microelectrodes. Appl Environ Microbiol 59: 3840–3849 Rasheed M, Badran MI, Hüttel M (2003a) Particulate matter filtration and seasonal nutrient dynamics in permeable carbonate and silicate sands of the Gulf of Aqaba, Red Sea. Coral Reefs 22: 167–177 Rasheed M, Badran MI, Huettel M (2003b) Influence of sediment permeability and mineral composition on organic matter degradation in three sediments from the Gulf of Aquaba, Red Sea. Estuar Coast Shelf Sci 57: 369–384 Rasmussen MB, Henriksen K, Jensen A (1983) Possible causes of temporal fluctuations in primary production of the microphytobenthos in the Danish Wadden Sea. Mar Biol 73: 109–114 Reeburgh WS (1967) An improved interstitial water sampler. Limnol Oceanogr 12: 163–165 Reichardt W (1989) Microbiological aspects of bioturbation. Sci Mar 53: 301–306 Reichelt AC (1991) Environmental effects of meiofaunal burrowing. In: Meadows PS, Meadows A (eds) The environmental impact of burrowing animals and animal burrows. Oxford Scientific Publications, Oxford, pp. 33–52 Reid DM (1932–33) Salinity interchange between salt water in sand overflowing fresh water at low tide. J Mar Biol Ass UK 18: 299–306.
478
References
Reid JW (1994) Latitudinal diversity patterns of continental benthic copepod species assemblages in the Americas. Hydrobiologia 292/293: 341–349 Reise K (1979) Moderate predation on meiofauna by the macrobenthos of the Wadden Sea. Helgol Wiss Meeresunters 32: 453–465 Reise K (1981a) High abundance of small zoobenthos around biogenic structures in tidal sediments of the Wadden Sea. Helgol Meeresunters 34: 413–425 Reise K (1981b) Gnathostomulida abundant alongside polychaete burrows. Mar Ecol Prog Ser 6: 329–333 Reise K (1983) Sewage, green algal mats anchored by lugworms, and the effects on Turbellaria and small Polychaeta. Helgol Meeresunters 36: 151–162 Reise K (1984) Free-living Plathelminthes (Turbellaria) of a marine sand flat: an ecological study. Microfauna Mar 1: 1–62 Reise K (1985) Predator control in marine tidal sediments. Cambridge University Press, Cambridge Reise K (1987) Experimental analysis of processes between species on marine tidal flats. Ecol Stud 61: 391–400 Reise K (1988) Plathelminth diversity in littoral sediments around the island of Sylt in the North Sea. Fortschr Zool 36: 469–480 Reise K (2002) Sediment mediated species interactions in coastal waters. J Sea Res 48: 127–141 Reise K, Ax P (1979) A meiofaunal thiobios limited to the anaerobic sulfide system of marine sand does not exist. Mar Biol 54: 225–237. Remane A (1927) Halammohydra, ein eigenartiges Hydrozoon der Nord- und Ostsee. Z Morph Ökol Tiere 7: 643–677 Remane A (1932) Archiannelida. In: Grimpe GW, E. (ed) Die Tierwelt der Nord- und Ostsee. Akademische Verlagsgesellschaft, Leipzig, VIa 1, pp. 1–36 Remane A (1933) Verteilung und Organisation der benthonischen Mikrofauna der Kieler Bucht. Wiss Meeresunters, Abt Kiel, NF 21, pp. 161–221 Remane A (1934) Die Brackwasserfauna. Zool Anz Suppl 7: 34–74 Remane A (1936a) Gastrotricha und Kinorhyncha. Bronns Klassen und Ordnungen des Tierreichs 4. Bd.: Vermes 2. Abt. 1. B, 4d, 2T, Akademische Verlagsgesellschaft Geest & Portig, Leipzig, p. 385 Remane A (1936b) Monobryozoon ambulans n.g. n.sp. ein eigenartiges Bryozoon des Meeressandes. Zool Anz 113: 161–167 Remane A (1940) Einführung in die zoologische Ökologie der Nord- und Ostsee. In: Grimpe G, Wagler E (eds) Die Tierwelt der Nord- und Ostsee. Akademische Verlagsgesellschaft Geest & Portig, Leipzig., 1a, p. 238 Remane A (1949) Die psammobionten Rotatorien der Nord- und Ostsee. Kieler Meeresforsch 6: 59–67 Remane A (1952a) Die Besiedlung des Sandbodens im Meere und die Bedeutung der Lebensformtypen für die Ökologie. Verh Dt Zool Ges Wilhelmshaven 1951 Zool. Anz. Suppl.16: 327–359 Remane A (1952b) Die Grundlagen des natürlichen Systems, der vergleichenden Anatomie und der Phylogenetik. Theoretische Morphologie und Systematik I. Akademische Verlagsgesellschaft Geest & Portig, Leipzig, p. 400 Renaud-Mornant JC (1982) Species diversity in marine Tardigrada. In: Nelson D (ed) Proceedings of 3rd International Symposium on Tardigrada. East Tennessee State University Press, Johnson City, USA, pp. 149–178 Renaud-Mornat J, Serène P (1967) Note sur la microfauna de la côte orientale de la Malaisie. Cah Pacif 11: 51–73 Revsbech NP, Jørgensen BB (1986) Microelectrodes: their use in microbial ecology. In: Marshall KC (ed) Advances in microbial ecology. Plenum, New York, pp. 293–352 Revsbech NP, Ward DM (1983) Oxygen microelectrode that is insensitive to medium chemical composition: use in an acid microbial mat dominated by Cyanidium caldarium. Appl Environ Microbiol 45: 755–759 Revsbech NP, Jørgensen BB, Blackburn TH (1980a) Oxygen in the sea bottom measured with a microelectrode. Science 207: 1355–1356
References
479
Revsbech NP, Sørensen J, Blackburn TH, Lomholt JP (1980b) Distribution of oxygen in marine sediments measured with microelectrodes. Limnol Oceanogr 25: 403–411 Rex MA, Stuart C, Hessler R, Allen J, Sanders H, Wilson G (1993) Global-scale latitudinal patterns of species diversity in the deep-sea benthos. Nature 365: 636–639 Rex MA, Etter RJ, Morris JS, Crouse J, McClain CR, Johnson NA, Stuart CT, Deming JW, Thies R, Avery R (2006) Global bathymetric patterns of standing stock and body size in the deep-sea benthos. Mar Ecol Prog Ser 317: 1–8 Rey JR, Shaffer J, Kain T, Stahl R, Crossman R (1992) Sulfide variation in the pore and surface waters of artificial salt-marsh ditches and a natural tidal creek. Estuaries 15: 257–269 Rhoads DC, Young DK (1970) The influence of deposit-feeding organisms on sediment stability and community trophic structure. J Mar Res 28: 150–178 Rhoads DC, Aller RC, Goldhaber MB (1977) The influence of colonizing benthos on physical properties and chemical diagenesis of the estuarine seafloor. In: Coull BC (ed) Ecology of marine benthos. University of South Carolina Press, Columbia, BC, pp. 113–138 Rhoads DC, McCall PL, Yingst JY (1978) Disturbance and production on the estuarine seafloor. Am Sci 66: 577–586 Rhumbler L (1938) Foraminiferen aus dem Meeressand von Helgoland, gesammelt von A. Remane (Kiel). Kiel Meeresforsch 2: 157–222 Ricci C, Balsamo M (2000) The biology and ecology of lotic rotifers and gastrotrichs. Freshw Biol 44: 15–28 Ricciardi A, Bourget E (1998) Weight-to-weight conversion factors for marine benthic macroinvertebrates. Mar Ecol Prog Ser 163: 245–251 Richmond CE, Wethey DS, Woodin SA (2007) Climate change and increased environmental variability: Demographic responses in an estuarine harpacticoid copepod. Ecol Modell 209: 189–202 Riddle MJ (1989) Bite profiles of some benthic grab samplers. Estuar Coast Shelf Sci 29: 285–292 Rieber RM (1994) The biphasic life cycle—a central theme of metazoan evolution. Am Zool 34: 484–491 Riebesell U (1992) The formation of large marine snow and its sustained residence in surface waters. Limnol Oceanogr 37: 63–76 Riedl RJ (1969) Gnathostomulida from America. Science 163: 445–452 Riedl RJ (1971) Energy exchange at the bottom water interface. Thalass Jugosl 7: 329–339 Riedl RJ, Huang N, Machan R (1972) The subtidal pump: a mechanism of interstial water exchange by wave action. Mar Biol 13: 210–221 Riedl RJ, Machan R (1972) Hydrodynamic patterns in lotic intertidal sands and their bioclimatological implications. Mar Biol 13: 179–209 Riedl RJ, Ott JA (1970) A suction-corer to yield electric potentials in coastal sediment layers. Senckenb Marit 2: 67–84 Rieger R (1976) Monociliated epidermal cells in Gastrotricha: significance for concepts of early metazoan evolution. Z Zool Syst Evolut-forsch 14: 198–226 Rieger RM (1980) A new group of interstitial worms, Lobatocerebridae nov. fam. (Annelida) and its significance for metazoan phylogeny. Zoomorphologie 95: 41–84 Rieger RM (1981) Morphology of the Turbellaria at the ultrastructural level. Hydrobiologia 84: 213–229 Rieger RM (1985) The phylogenetic status of the acoelomate organization within the Bilateria: a histological perspective. In: Conway Morris S, George JD, Gibson R, Platt HM (eds) The origins and relationships of lower invertebrates. Clarendon, Oxford, pp. 101–122 (Syst. Ass, vol 28) Rieger RM (1991) Neue Organisationstypen aus der Sandlückenraumfauna: die Lobatocerebriden und Jennaria pulchra. Verh Dtsch Zool Ges 84: 247–259 Rieger RM (1998) 100 years of research on ‘Turbellaria’. Hydrobiologia 383: 1–27 Rieger RM, Rieger GE (1975) Fine structure of the pharyngeal bulb in Trilobodrilus and its phylogenetic significance within Archiannelida. Tissue Cell 7: 267–279 Rieger RM, Rieger GE (1976) Fine structure of the archiannelid cuticle and remarks on the evolution of the cuticle within the Spiralia. Acta Zool (Stockh) 57: 53–68
480
References
Rieger RM, Rieger GE (1991) Bacterial symbionts of Jennaria pulchra, a new genus of interstitial worms with uncertain systematic position. Am Zool 31: 25 A, 149 Rieger RM, Ruppert E (1978) Resin embedments of quantitative meiofauna samples for ecological and structural studies. Description and application. Mar Biol 46: 223–235 Rieger RM, Sterrer W (1975) New spicular skeletons in Turbellaria, and the occurence of spicules in marine meiofauna. Z zool Syst Evolut- forsch 13: 207-248 Rieger R, Tyler S (1979) The homology theorem in ultrastructure research. Am Zool 19: 655–664 Rieger RM, Tyler S (1995) Sister-group relationship of Gnathostomulida and Rotifera — Acanthocephala. Invert Biol 114: 186–188 Rieger R, Ruppert E, Rieger GE, Schoepfer-Sterrer C (1974) On the fine structure of gastrotrichs with description of Chordodasys antennatus sp.n. Zool Scr 3: 219–237 Riemann F (1966) Die Verbreitung der interstitiellen Fauna im Elbe-Aestuar. Veröff Inst Meeresforsch Bremerhaven Sbd. 2: 117–124 Riemann F (1985) Eisen und Mangan in pazifischen Tiefsee-Rhizopoden und Beziehungen zur Manganknollen-Genese. Int Rev Ges Hydrobiol 70: 165–172 Riemann F (1989) Gelatinous phytoplankton detritus aggregates on the Atlantic deep-sea bed. Structure and mode of formation. Mar Biol 100: 533–539 Riemann F (1995) The deep-sea nematode Thalassomonhystera bathislandica sp.nov. and microhabitats of nematodes in flocculent surface sediments. J mar biol Ass UK 75: 715–724 Riemann F, Schrage M (1978) The mucus-trap hypothesis on feeding of aquatic nematodes and implications for biodegradation and sediment texture. Oecologia (Berl) 34: 75–88 Riemann F, Helmke E (2002) Symbiotic relations of sediment-agglutinating nematodes and bacteria in detrital habitats: The enzyme-sharing concept. Mar Ecol PSZN 23: 93–113 Riemann F, Sime-Ngando T (1997) Note on sea-ice nematodes (Monhysteroidea) from Resolute Passage, Canadian high Arctic. Polar Biol 18: 70–75 Riemann F, Ernst W, Ernst R (1990) Acetate uptake from ambient water by the free-living marine nematode Adoncholaimus thalassophygas. Mar Biol 104: 453–457 Rieper-Kirchner M (1989) Microbial degradation of North Sea macroalgae: field and laboratory studies. Bot Mar 32: 241–252 Riera P, Richard P, Grémare A, Blanchard G (1996) Food source of intertidal nematodes in the Bay of Marennes-Oléron (France), as determined by dual stable isotope analysis. Mar Ecol Prog Ser 142: 303–309 Riess W, Giere O, Kohls O, Sarbu SM (1999) Anoxic thermomineral cave waters and bacterial mats as habitat for freshwater nematodes. Aquat Microb Ecol 18: 157–164 Riser NW (1984) General observations on the intertidal interstitial fauna of New Zealand. Tane 30: 239–250 Riser NW (1985) Epilogues: Nemertinea, a Successful Phylum. Am Zool 25: 145–151 Riser NW (1989) Speciation and time—relationships of the nemertines to the acoelomate metazoan Bilateria. Bull Mar Sci 45: 531–538 Riutort M, Field KG, Raff RA, Baguna J (1993) 18S rRNA sequences and phylogeny of Plathyhelminthes. Biochem Syst Ecol 21: 71–77 Robertson AL (2000) Lotic meiofaunal community dynamics: colonisation, resilience and persistence in a spatially and temporally heterogeneous environment. Freshw Biol 44: 135–147 Robertson AL (2002) Changing times: the temporal dynamics of freshwater benthic microcrustacea. In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys, Leiden, The Netherlands, pp. 261–278 Robertson AL, Rundle SD, Schmid Araya JM (2000a) An introduction to a special issue on lotic meiofauna. Freshw Biol 44: 1–3 Robertson AL, Rundle SD, Schmid-Araya JM (2000b) Putting the meio into stream ecology: current findings and future directions for lotic meiofaunal research. Freshw Biol 44: 177–183 Robertson AL, Rundle SD, Schmid-Araya JM (eds) (2000c) The role of meiofauna in lotic ecosystems. Freshw Biol 44: 1–183
References
481
Robinson SMC, Chandler RA (1993) An effective and safe method for sorting small molluscs from sediment. Limnol Oceanogr 38: 1088–1091 Rocha-Olivares A, Fleeger JW, Foltz DW (2001) Decoupling of molecular and morphological evolution in deep lineages of a meiobenthic harpacticoid copepod. Mol Biol Evol 18: 1088–1102 Rodriguez JG, Incera M, de la Huz R, Lopez J, Lastra M (2007) Polycyclic aromatic hydrocarbons (PAHs), organic matter quality and meiofauna in Galician sandy beaches, 6 months after the Prestige oil-spill. Mar Pollut Bull 54: 1046–1052 Rohde K (1992) Latitudinal gradients in species diversity: the search for a primary cause. Oikos 65: 514–527 Rohde K, Hefford C, Ellis JT, Baverstock PR, Johnson AM, Watson NA, Dittmann S (1993) Contributions to the phylogeny of Platyhelminthes based on partial sequencing of 18S ribosomal DNA. Int J Parasitol 23: 705–724 Rokas A, Krüger D, Carroll SB (2005) Animal evolution and the molecular signature of radiations compressed in time. Science 310: 1933–1938 Romeyn K, Bouwman LA (1983) Food selection and consumption by estuarine nematodes. Hydrobiol Bull 17: 103–109 Rose A, Seifried S, Willen E, George KH, Veit-Kohler G, Brohldick K, Drewes J, Moura G, Arbizu PM, Schminke HK (2005) A method for comparing within-core alpha diversity values from repeated multicorer samplings, shown for abyssal Harpacticoida (Crustacea: Copepoda) from the Angola Basin. Organ Divers Evolut 5: 3–17 Rosenberg R, Davey E, Gunnarsson J, Norling K, Frank M (2007) Application of computer-aided tomography to visualize and quantify biogenic structures in marine sediments. Mar Ecol Prog Ser 331: 23–34 Rosenzweig ML (1995) Species diversity in space and time. Cambridge University Press, Cambridge Rossi L (1961) Morfologia e riproduzione vegetativa di un Madreporario nuovo per il Mediterraneo. Boll Zool 28: 261–272 Rota E, Erséus C (1996) Six new species of Grania (Oligochaeta, Enchytraeidae) from the Ross Sea, Antarctica. Antarct Sci 8: 169–183 Rota E, Martin P, Erséus C (2001) Soil-dwelling polychaetes: enigmatic as ever? Some hints on their phylogenetic relationships as suggested by a maximum parsimony analysis of 18S rRNA gene sequences. Contr Zool 70: 127–138 Rouch R (1968) Contribution à la connaissance des harpacticides hypogés (Crustacés-Copépodes). Ann Spéléol 23: 13–158 Rouch R, Danielopol DL (1997) Species richness of microcrustacea in subterranean freshwater habitats: comparative analysis and approximate evaluation. Int Rev Ges Hydrobiol 82: 121–145 Round FE (1971) Benthic marine diatoms. Oceanogr Mar Biol Annu Rev 9: 83–139 Rouse GW, Fauchald K (1997) Cladistics and polychaetes. Zool Scr 26: 139–204 Rouse GW, Pleijel F (2001) Polychaetes. Oxford University Press, Oxford, p. 354 Rouse G, Pleijel F (eds) (2006) Reproductive biology and phylogeny of Annelida. In: Jamieson BGM (ed) Reproductive biology and phylogeny. Vol. 4. Science Publishers, Enfield, UK, p. 688 Rousset V, Pleijel F, Rouse GW, Erséus C, Sidall ME (2007) A molecular phylogeny of annelids. Cladistics 23: 41–63 Rudnick DT (1989) Time lags between the deposition and meiobenthic assimilation of phytodetritus. Mar Ecol Prog Ser 50: 231–240 Rundle SD, Bilton DT, Shiozawa DK (2000) Global and regional patterns in lotic meiofauna. Freshwat Biol 44: 123–134 Rundle SD, Ramsay PM (1997) Microcrustacean communities in streams from two physiographically contrasting regions of Britain. J Biogeogr 24: 101–111 Rundle SD, Robertson AL, Schmid-Araya JME (eds) (2002) Freshwater meiofauna: biology and ecology. Backhuys, Leiden, The Netherlands, p. 369 Runnegar B (1991) Oxygen and the early evolution of the metazoa. In: Bryant C (ed) Metazoan life without oxygen. Chapman & Hall, London, pp. 65–87
482
References
Ruppert EE (1972) An efficient, quantitative method for sampling the meiobenthos. Limnol Oceanogr 17: 629–631 Ruppert EE (1977) Zoogeography and speciation in marine Gastrotricha. Mikrofauna Meeresboden 61: 231–251 Ruppert EE (1982) Comparative ultrastructure of the gastrotrich pharynx and the evolution of myoepithelial foreguts in Aschelminthes. Zoomorphology 99: 181–220 Rusch A, Huettel M (2000) Advective particle transport into permeable sediments—Evidence from experiments in an intertidal sandflat. Limnol Oceanogr 45: 525–533 Rutledge PA, Fleeger JW (1988) Laboratory studies on core sampling with application to subtidal meiobenthos collection. Limnol Oceanogr 33: 274–280 Rutledge PA, Fleeger JW (1993) Abundance and seasonality of meiofauna, including harpacticoid copepod species, associated with stems of the salt-marsh cord grass, Spartina alterniflora. Estuaries 16: 760–768 Ruttner-Kolisko A (1961) Biotop und Biozönose des Sandufers einiger österreichischer Flüsse. Verh Internat Verein Limnol 14: 362–368 Ruttner-Kolisko A (1962) Porenraum und kapillare Wasserströmung im Limnopsammal, ein Beispiel für die Bedeutung verlangsamter Strömung. Schweiz Z Hydrol 24: 444–458 Rysgaard S, Christensen PB, Sørensen MV, Funch P, Berg P (2000) Marine meiofauna, carbon and nitrogen mineralization in sandy and soft sediments of Disko Bay, West Greenland. Aquat Microb Ecol 21: 59–71 Rzeznik-Orignac J, Fichet D, Boucher G (2004) Extracting massive numbers of nematodes from muddy marine deposits: efficiency and selectivity. Nematology 6: 605–616 Rzeznik-Orignac J, Boucher G, Fichet D, Richard P (2008) Stable isotope analysis of food source and trophic position of intertidal nematodes and copepods. Mar Ecol Prog Ser 359: 145–150 Saager PM, Sweerts JP, Ellermeijer HJ (1990) A simple pore water sampler for coarse, sandy sediments of low porosity. Limnol Oceanogr 35: 747–750 Saburova MA, Polikarpov IG, Burkovsky IV (1995) Spatial structure of an intertidal sandflat microphytobenthic community as related to different spatials scales. Mar Ecol Prog Ser 129: 229–239 Sach G, van Bernem KH (1996) Spatial patterns of harpacticoid copepods on tidal flats. Senckenb marit 26: 97–106 Salvat B (1964) Les conditions hydrodynamiques interstitielles des sediments meubles intertidaux et la répartiion verticale de la fauna endogée. C R Hebd Séanc Acad Sci Paris 259: 1576–1579 Salvini-Plawen L von (1966) Zur Kenntnis der Cnidaria des nordadriatischen Mesopsammon. Veröff Inst Meeresforsch Bremerhaven Sbd 2: 165–186 Salvini-Plawen L von (1968) Neue Formen im marinen Mesopsammon: Kamptozoa und Aculifera. Ann Naturhist Mus Wien 72: 231–272 Salvini-Plawen L von (1972) Zur Taxonomie und Ökologie mediterraner Holothuroidea-Apoda. Helgol wiss Meeresunters 23: 459–466 Salvini-Plawen L von (1985) New interstitial Solenogastres (Mollusca). Stygologia 1: 101–108 Salvini-Plawen L von (1987) Mesopsammic Cnidaria from Plymouth (with systematic notes). J Mar Biol Ass UK 67: 623–637 Sambugar B, Giani N, Martínez-Ansemil E (1999) Groundwater oligochaetes from SouthernEurope. Tubificidae with marine phyletic affinities: new data with description of a new species, review and consideration on their origin. Mém Biospéol 26: 107–116 Sanders HL (1955) The Cephalocarida, a new subclass of Crustacea from Long Island Sound. Proc Natn Acad Sci 41: 61–66 Sanders HL (1958) Benthic studies in Buzzards Bay. I. Animal–sediment relationships. Limnol Oceanogr 3: 245–258. Sanders HL (1959) The significance of the Cephalocarida in crustacean phylogeny. In: Hewer HR, Riley ND (eds) Proceedings of 15th International Congress of Zoology London 1958, pp. 337–340 Sanders HL, Hessler RR, Garner SP (1985) Hirsutia bathyalis, a new unusual deep-sea benthic peracaridan crustacean from the tropical Atlantic. J Crust Biol 5: 30–57 Sandulli R, Nicola MD (1991) Responses of meiobenthic communities along a gradient of sewage pollution. Mar Poll Bull 22: 463–467
References
483
Santangelo G, Lucchesi P (1995) Spatial distribution pattern of ciliated Protozoa in a Mediterranean interstitial environment. Aquat Microb Ecol 9: 47–54 Santos, GAP dos, Derycke S, Genevois VGF, Coelho LCBB, Correia MTS, Moens T (2008a) Differential effects of food availability on population development and fitness of three species of estuarine, bacterial-feeding nematodes. J Exp Mar Biol Ecol 355: 27–40 Santos, GAP dos, Derycke S, Genevois VGF, Coelho LCBB, Correia MTS, Moens T (2008b) Interactions among bacterial-feeding nematode species at different levels of food availability. Mar Biol (submitted) Santos PJP, J. C, Souza-Santos LP (1995) Microphytobenthic patches and their influence on meiofaunal distribution. Cah Biol Mar 36: 133–139 Särkkä J (1996a) Meiofauna ratios as environmental indicators in the profundal depths of large lakes. Environ Monit Assess 42: 229–240 Särkkä J (1996b) Meiofauna of the profundal zone of the northern part of Lake Ladoga as an indicator of pollution. Hydrobiologia 322: 29–38 Särkkä J, Levonen L, Mäkelä J (1997) Meiofauna of springs in Finland in relation to environmental factors. Hydrobiologia 347: 139–150 Sarvala J (1998) Ecology and role of benthic copepods in northern lakes. J Mar Systems 15: 75–86 Sasekumar A (1994) Meiofauna of a mangrove shore on the west coast of peninsular Malaysia. Raffl Bull Zool 42: 901–915 Sassuchin DN, Kabanov NM, Neizvestnova ES (1927) K izuceniju mikroskopiceskogo nasalenija nanosnych peskov v rusle reki Oki. Ueber die mikroskopische Pflanzen- und Tierwelt der Sandfläche des Okaufers bei Murom. Russ Gidrobiol Zh 6: 59–83 (in Russian and German) Saunders GR, Moore CG (2004) In situ approach to the examination of the impact of copper pollution on marine meiobenthic copepods. Zool Studies 43: 350–365 Sayles FL, Mangelsdorf PC, Wilson TRS, Hume DN (1976) A sampler for the in situ collection of marine sedimentary pore waters. Deep-Sea Res 23: 259–264 Schallenberg M, Kalff J (1993) The ecology of sediment bacteria in lakes and comparisons with other aquatic ecosystems. Ecology 74: 919–934 Scheltema AH (1985) The aplacophoran family Prochaetodermatidae in the North American Basin, including Chevroderma n.g. and Spathoderma n.g. (Mollusca; Chaetodermomorpha). Biol Bull 169: 484–529 Scheltema AH (1998) A new class of Mollusca discovered off Sweden. Abstr Proc 32nd Eur Mar Biol Symp, Lysekil (Sweden), 131 Scheltema AH (2000) Two new hydrothermal vent species, Helicoradomenia bisquama and Helicoradomenia acredema, from the Eastern Pacific ocean (Mollusca, Aplacophora). Argonauta 14: 15–25 Scheltema AH, Ivanov DL (2002) An aplacophoran postlarva with iterated dorsal groups of spicules and skeletal similarities to paleozoic fossils. Invert Biol 121: 1–10 Scherer B (1985) Annual dynamics of a meiofauna community from the sulfide layer of a North Sea sand flat. Microfauna Mar 2: 117–161 Schewe I, Soltwedel T (2000) Deep-sea meiobenthos of the central Arctic Ocean: distribution patterns and size-structure under extreme oligotrophic conditions. Vie Milieu 49: 79–92 Schiemer F (1982) Food dependence and energetics of free-living nematodes. I. Respiration, growth and reproduction of Caenorhabditis briggsae at different levels of food supply. Oecologia (Berl) 54: 108–121 Schiemer F, Duncan A (1974) The oxygen consumption of a freshwater benthic nematode, Tobrilus gracilis (Bastian). Oecologia (Berl) 15: 121–126 Schiemer F, Ott J (2001) Metabolic levels and microhabitat of an interstitial cephalocarid and micro-isopod. PSZN Mar Ecol 22: 13–22 Schiemer F, Novak R, Ott J (1990) Metabolic studies on thiobiotic free-living nematodes and their symbiotic microorganisms. Mar Biol 106: 129–137 Schizas NV, Shirley TC (1996) Seasonal changes in structure of an Alaskan intertidal meiofaunal assemblage. Mar Ecol Prog Ser 133: 115–124 Schizas NV, Street GT, Coull BC, Chandler GT, Quattro JM (1997) An efficient DNA extraction method for small metazoans. Mol Mar Biol Biotechnol 6: 381–383
484
References
Schizas NV, Street GT, Coull BC, Chandler GT, Quattro JM (1999) Molecular population structure of the marine benthic copepod Microarthridion littorale along the southeastern and Gulf coasts of the USA. Mar Biol 135: 399–405 Schizas NV, Chandler GT, Coull BC, Klosterhaus SL, Quattro JM (2001) Differential survival of three mitochondrial lineages in a marine copepod exposed to a mixture of pesticides. Environ Sci Technol 35: 535–538 Schizas NV, Coull BC, Chandler GT, Quattro JM (2002) Sympatry of distinct mitochondrial DNA lineages in a copepod inhabiting estuarine creeks in the southeastern USA. Mar Biol 14: 585–594 Schmid PE, Schmid-Arraya JM (1997) Predation on meiobentic assemblages: resource use of a tanypodid guild (Chironomidae, Diptera) in a gravel stream. Freshw Biol 38: 67–91 Schmid-Araya JM (1995) Disturbance and population dynamics of rotifers in bed sediments. Hydrobiologia 313/314: 279–290 Schmid-Araya JM (1998) Small-sized invertebrates in a gravel stream: community structure and variability of benthic rotifers. Freshw Biol 39: 25–39 Schmid-Araya JM, Schmid PE (2000) Trophic relationships: integrating meiofauna into a realistic benthic food web. Freshw Biol 44: 149–163 Schmidt C (1989) Zur Ökologie des sulfidtoleranten Nematoden-Genus Tobrilus Andrássy 1959 aus dem Elbwatt. Dipl Thesis, University of Hamburg, Hamburg, pp. 1–145 Schmidt H, Westheide W (1997/98) RAPD-PCR experiments confirm the distinction between three morphologically similar species of Nerilla (Polychaeta: Nerillidae). Zool Anz 236: 277–285 Schmidt JL, Deming JW, Jumars PA, Keil RG (1998) Constancy of bacterial abundance in surficial marine sediments. Limnol Oceanogr 43: 976–982 Schmidt P, Teuchert G (1969) Quantitative Untersuchungen zur Ökologie der Gastrotrichen im Gezeiten-Sandstrand der Insel Sylt. Mar Biol 4: 4–23 Schmidt-Rhaesa A (2002) Two dimensions of biodiversity research exemplified by Nematomorpha and Gastrotricha. Integr Comp Biol 42: 633–640 Schminke HK (1973) Evolution, System und Verbreitungsgeschichte der Familie Parabathynellidae (Bathynellacea, Malacostraca). Mikrofauna Meeresboden 24: 1–192 Schminke HK (1981) Adaptation of Bathynellacea (Crustacea, Syncarida) to life in interstitial (Zoea Theory). Int Rev Ges Hydrobiol 66 (4): 575–637 Schminke HK (1986) Syncarida. In: Botosaneanu L (ed) Stygofauna Mundi. Brill/ Backhuys, Leiden, pp. 389–404 Schnack-Schiel SB, Dieckmann GS, Gradinger R, Melnikov IA, Spindler M, Thomas DN (2001) Meiofauna in sea ice of the Wedell Sea (Antarctica). Polar Biol 24: 724–728 Schockaert ER, Hooge M, Sluys R, Schilling S, Tyler S, Artois T (2008) Global diversity of free living flatworms (Platyhelminthes, “Turbellaria”) in freshwater. Hydrobiologia 595: 41–48 Scholz DS, Matthews LL, Feller RJ (1991) Detecting selective digestion of meiobenthic prey by juvenile spot Leiostomus xanthurus (Pisces) using immunoassays. Mar Ecol Prog Ser 72: 59–67. Schratzberger M, Warwick RM (1998a) Effects of physical disturbance on nematode communities in sand and mud: a microcosm experiment. Mar Biol 130: 643–650 Schratzberger M, Warwick RM (1998b) Effects of the intensity and frequency of organic enrichment on two estuarine nematode communities. Mar Ecol Prog Ser 164: 83–94 Schratzberger M, Warwick RM (1999a) Differential effects of various types of disturbances on the structure of nematode assemblages: an experimental approach. Mar Ecol Prog Ser 181: 227–236 Schratzberger M, Warwick RM (1999b) Impact of predation and sediment disturbance by Carcinus maenas (L.) on free-living nematode community structure. J Exp Mar Biol Ecol 235: 255–271 Schratzberger M, Gee JM, Rees HL, Boyd SE, Wall CM (2000) The structure and taxonomic composition of sublittoral meiofauna assemblages as an indicator of the status of marine environments. J Mar Biol Ass UK 80: 969–980 Schratzberger M, Dinmore TA, Jennings S (2002a) Impacts of trawling on the diversity, biomass and structure of meiofauna assemblages. Mar Biol 140: 83–93
References
485
Schratzberger M, Wall CM, Reynolds WJ, Reed J, Waldock MJ (2002b) Effects of paint-derived tributyltin on structure on estuarine nematode assemblages in experimental microcosms. J Exp Mar Biol Ecol 272: 217–235 Schratzberger M, Daniel F, Wall CM, Kilbride R, Macnaughton SJ, Boyd SE, Rees HL, Lee K, Swanell RPJ (2003) Response of estuarine meio- and macrofauna to in situ bioremediation of oil-contaminated sediment. Mar Pollut Bull 46: 430–443 Schratzberger M, Bolam SG, Whomersley P, Warr K, Rees HL (2004) Development of a meiobenthic nematode community following the intertidal placement of various types of sediment. J Exp Mar Biol Ecol 303: 79–96 Schratzberger M, Bolam S, Whomersley P, Warr K (2006) Differential response of nematode colonist communities to the intertidal placements of dredged material. J Exp Mar Biol Ecol 334: 244–255 Schratzberger M, Warr K, Rogers SI (2007a) Functional diversity of nematode community in the southwestern North Sea. Mar Environ Res 63: 368–389 Schratzberger M, Lampadariou N, Somerfield PJ, Vandepitte L, Vanden Berghe E (2007b) The impact of seabed disturbance on nematode communities—linking field and laboratory observations. Mar Ecol Prog Ser (accepted) Schrijvers J, Van Gansbeke D, Vincx M (1995) Macrobenthic infauna of mangroves and surrounding beaches at Gazi Bay, Kenya. Hydrobiologia 306: 53–66 Schrijvers J, Schallier R, Silence J, Okondo JP, Vincx M (1997) Interactions between epibenthos and meiobenthos in a high intertidal Avicennia marina mangrove forest. Mangrov Salt Marsh 1: 137–154 Schrijvers J, Vincx M (1997) Cage experiments in an East African mangrove forest: a synthesis. J Sea Res 38: 123–133 Schünemann H, Werner I (2005) Seasonal variations in distribution patterns of sympagic meiofauna in Arctic pack ice. Mar Biol 146: 1091–1102 Schulz E (1950) Psammohydra nanna, ein neues solitäres Hydrozoon in der westlichen Beltsee. Kieler Meeresforsch 7: 122–137 Schumann R, Rentsch D (1998) Staining particulate organic matter with DTAF—a fluorescence dye for carbohydrates and protein: a new approach and application of a 2D image analysis system. Mar Ecol Prog Ser 163: 77–88 Schuster R (1965) Die Ökologie der terrestrischen Kleinfauna des Meeresstrandes. Verh Dtsch Zool Ges Kiel 1964: 492–521 Schuster R (1979) Soil mites in the marine environment. Rec Adv Acarology 1: 593–602 Schuster RO, Nelson D, Grigarick A, Christenberry D (1980) Systematic criteria of the Eutardigrada. Trans Am Microsc Soc 99: 284–303 Schwinghamer P (1981a) Characteristic size distributions of integral benthic communities. Can J Fish Aquat Sci 38: 1255–1263 Schwinghamer P (1981b) Extraction of living meiofauna from marine sediments by centrifugation in a silica sol–Sorbitol mixture. Can J Fish Aquat Sci 38: 476–478 Schwinghamer P (1983) Generating ecological hypotheses from biomass spectra using causal analysis: a benthic example. Mar Ecol Prog Ser 13: 151–166 Schwinghamer P, Hargrave B, Peer D, Hawkins CM (1986) Partitioning of production and respiration among size groups of organisms in an intertidal benthic community. Mar Ecol Prog Ser 31: 131–142 Schwoerbel J (1961a) Über die Lebensbedingungen und die Besiedlung des hyporheischen Lebensraumes. Arch Hydrobiol Suppl 25: 182–214 Schwoerbel J (1961b) Subterrane Wassermilben (Acari: Hydrachnellae, Porohalacaridae und Stygotrombiidae), ihre Ökologie und Bedeutung für die Abgrenzung eines aquatischen Lebensraumes zwischen Oberfläche und Grundwasser. Arch Hydrobiol Suppl 25: 242–306 Schwoerbel J (1967) Das hyporheische Interstitial als Grenzbiotop zwischen oberirdischem und subterranem Ökosystem und seine Bedeutung für die Primär-Evolution von Kleinsthöhlenbewohnern. Arch Hydrobiol Suppl 33: 1–62
486
References
Scott FJ, Marchant HJ (2005) Antarctic marine protists. CSIRO Publishing, Collingwood, VICT, Australia, p. 572 Sebastian S, Raes M, De Mesel I, Vanreusel A (2007) Comparison of the nematode fauna from the Weddell Sea Abyssal Plain with two North Atlantic abyssal sites. Deep-Sea Res II 54: 1727–1736 Sedlacek L, Thistle D (2006) Emergence on the continental shelf: differences among species and between microhabitats. Mar Ecol Prog Ser 311: 29–36 Segers H (2008) Global diversity of rotifers (Rotifera) in freshwater. Hydrobiologia 595: 49–59 Segers H, Martens K (2005) Aquatic biodiversity II. Springer, Berlin, p. 390 Sen Gupta BK (ed) (2002) Modern Foraminifera. Kluwer, Dordrecht, p. 384 Sepers ABJ (1977) The utilisation of dissolved organic compounds in aquatic environments. Hydrobiologia 52: 39–54 Serban M (1960) La néotenie, et le problème de la taille chez les Copépodes. Crustaceana 1: 77–83 Sergeeva NG (1999) Morphological abnomalies of free-living Nematoda in extremal conditions (The Black Sea). International Conference on Oceanography of the Eastern Mediterranean Black Sea, Athens, pp. 105–106 Sergeeva NG (2003) Meiobenthos of deep-water anoxic hydrogen sulphide zone of the Black Sea. In: Yilmaz A (ed) Oceanography of the Eastern Mediterranean and Black Sea. Similarities and differences of two interconnected basins. Tübitak Publ., Ankara, pp. 880–887 Sergeeva NG, Burkatsky ON (2002) Meiobenthos of the eastern part of the Azov Sea in spring 2001. Ecol Morja 59: 37–41 Sergeeva NG, Gulin MB (2007) Meiobenthos from an active methane seepage area in the NW Black Sea. Mar Ecol PSZN 28: 152–159 Service SK, Bell SS (1987) Density-influenced active dispersal of harpacticoid copepods. J Exp Mar Biol Ecol 114: 49–62 Service SK, Feller RJ, Coull BC, Woods R (1992) Predation effect of three fish species and a shrimp on macrobenthos and meiobenthos in microcosms. Estuar Coast Shelf Sci 34: 277–293 Shain DH, Mason TA, Farrell AH, Michalewicz LA (2001) Distribution and behavior of ice worms (Mesenchytraeus solifugus) in south-central Alaska. Can J Zool 79: 1813–1821 Shanks AL, Edmonson EW (1990) The vertical flux of metazoans (holoplankton, meiofauna, and larval invertebrates) due to their association with marine snow. Limnol Oceanogr 35: 455–463 Shanks AL, Walters K (1997) Holoplankton, meroplankton, and meiofauna associated with marine snow. Mar Ecol Prog Ser 156: 75–86 Shannon CE (1949) The mathematical theory of communication. In: Shannon CE, Weaver W (eds) The mathematical theory of communication. University of Illinois Press, Urbana, pp. 3–91 Sheldon RW, Prakash A, Sutcliffe WHJ (1972) The size distribution of particles in the ocean. Limnol Oceanogr 17: 327–340 Sheppard CRC (1999) How large should my sample be? Some quick guides to sample size and the power of tests. Mar Pollut Bull 38: 439–447 Sherman KM, Coull BC (1980) The response of meiofauna to sediment disturbance. J Exp Mar Biol Ecol 46: 59–71 Shiells GM, Anderson KJ (1985) Pollution monitoring using nematode/copepod ratio. A practical application. Mar Pollut Bull 16: 62–68 Shimanaga M, Kitazato H, Shirayama Y (2000) Seasonal patterns of vertical distribution between meiofaunal groups in relation to phytodetritus deposition in the bathyal Sagami Bay, Central Japan. J Oceanogr 56: 379–387 Shimanaga M, Nomaki H, Suetsugu K, Murayama M, Kitazato H (2007) Standing stock of deepsea metazoan meiofauna in the Sulu Sea and adjacent areas. Deep-Sea Res II 54: 131–144 Shirayama Y (1984) The abundance of deep-sea meiobenthos in the western Pacific in relation to environmental factors. Oceanol Acta 7: 113–121 Shirayama Y (1992) Respiration rates of bathyal meiobenthos collected using a deep-sea submersible SHINKAI 2000. Deep-Sea Res 39: 781–788 Shirayama Y (1995) Ingestion rates of bathyal deep-sea meiobenthos collected from Suruga Bay, Central Japan. Vie Milieu 45: 11–15
References
487
Shirayama Y, Fukushi T (1995) Comparisons of deep-sea sediments and overlying water collected using multicorer and box corer. J Oceanogr 51: 75–82 Shirayama Y, Horikoshi M (1989) Comparison of the benthic size structure between sublittoral, upper-slope and deep-sea areas of the western Pacific. Int Rev Ges Hydrobiol 74: 1–13 Shirayama Y, Ohta S (1990) Meiofauna in a cold-seep community off Hatsushima, Central Japan. J Oceanogr Soc Japan 46: 118–124 Shirayama Y, Swinbanks DD (1986) Oxygen profile in deep-sea calcareous sediment calculated on the basis of measured respiration rates of deep-sea meiobenthos and its relevance to manganese diagenesis. La Mer 24: 75–80 Shirayama Y, Kaku T, Higgins RP (1993) Double-sided microscopic observation of meiofauna using Hs-slides. Benthos Res 44: 41–44 Sich H (1990) Die benthische Ciliatenfauna bei Gabelsflach (Kieler Bucht) und deren Beeinflussung durch Bakterien. Eine Studie über Menge, Biomasse, Produktion, Bakterieningestion und Ultrastruktur von Mikroorganismen. Ber Inst Meeresk 191: 1–215 Sikora JP, Sikora WB, Erkenbrecher CW, Coull BC (1977) Significance of ATP, carbon, and caloric content of meiobenthic nematodes in partitioning benthic biomass. Mar Biol 44: 7–14 Sikora WB, Sikora JP (1982) Ecological implication of the vertical distribution of meiofauna in salt marsh sediments. In: Kennedy VS (ed) Estuarine comparisons. Academic, New York, pp. 269–282 Simon M, Grossart H-P, Schweitzer B, Ploug H (2002) Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28: 175–211 Simonini R, Molinari F, Pagliai AM, Ansaloni I, Prevedelli D (2003) Karyotype and sex determination in Dinophilus gyrociliatus (Polychaeta: Dinophilidae). Mar Biol 142: 441–445 Sket B (1996) The ecology of anchihaline caves. Trends Ecol Evolut 11: 221–225 Skowronski RS, Corbisier TN (2002) Meiofauna distribution in Martel Inlet, King George Island (Antarctica): sediment features versus food availability. Polar Biol 25: 126–134 Smith DJ, Underwood GJC (1998) Exoploymer production by intertidal epipelic diatoms. Limnol Oceanogr 43: 1578–1591 Smith EP, Stewart PM, Cairns Jr J (1985) Similarities between rarefaction methods. Hydrobiologia 120: 167–170 Smith LD, Coull BC (1987) Juvenile spot (Pisces) and grass shrimp predation on meiobenthos in muddy and sandy substrata. J Exp Mar Biol Ecol 105: 123–136 Smith T, Wharton DA, Marshall CJ (2008) Cold tolerance of an Antarctic nematode that survives intracellular freezing: comparisons with other nematode species. J Comp Physiol B 178: 93–100 Smol N, Willems KA, Govaere JCR, Sandée AJJ (1994) Composition, distribution and biomass of meiobenthos in the Oosterschelde Estuary (SW Netherlands). Hydrobiologia 282/283: 197–217 Snelgrove PVR, Butman CA (1994) Animal–sediment relationships revisited: Cause versus effect. Oceanogr Mar Biol Ann Rev 32: 111–177 Snelgrove PVR, Smith CR (2002) A riot of species in an environmental calm: the paradox of the species-rich deep-sea. Oceanogr Mar Biol Ann Rev 40: 311–342 Sørensen MV, Pardos F (2008) Kinorhynch systematics and biology—an introduction to the study of kinorhynchs, inclusive identification keys to the genera. Meiofauna Mar 16: 21–73 Sørensen MV, Hebsgaard MB, Heiner I, Glenner H, Willerslev E, Kristensen RM (2008) New data from an enigmatic phylum: evidence from molecular sequence data supports a sistergroup relationship between Loricifera and Nematomorpha. J Zool Syst Evol Res 46: 231–239 Sørensen MV, Sterrer W, Giribet G (2006) Gnathostomulid phylogeny inferred from a combined approach of four molecular loci and morphology. Cladistics 22: 32–58 Sørensen MV, Tyler S, Hooge M, Funch P (2003) Organization of the pharyngeal hard parts and musculature in Gnathostomula armata (Gnathostomulida, Gnathostomulidae). Can J Zool 81: 1463–1470 Soetaert K, Heip C (1989) The size structure of nematode assemblages along a Mediterranean deep-sea transect. Deep-Sea Res I 36: 93–102
488
References
Soetaert K, Muthumbi A, Heip C (2002) Size and shape of ocean margin nematodes: morphological diversity and depth-related patterns. Mar Ecol Prog Ser 242: 179–193 Soetaert K, Vincx M, Wittoeck J, Tulkens M (1995) Meiobenthic distribution and nematode community structure in five European estuaries. Hydrobiologia 311: 185–206 Soltwedel T (1997) Meiobenthos distribution pattern in the tropical East Atlantic: indication for fractionated sedimentation of organic matter to the sea floor?. Mar Biol 129: 747–756 Soltwedel T, Miljutina M, Mokievsky V, Thistle D, Vopel K (2003) The meiobenthos of the Molloy Deep (5600 m), Fram Strait, Arctic Ocean. Vie Milieu 53: 1–13 Soltwedel T, Portnova D, Kolar I, Mokievsky V, Schewe I (2005) The small-sized benthic biota of the Håkon Mosby Mud Volcano (SW Barents Sea slope). J Mar Syst 55: 271–290 Somerfield PJ, Clarke KR (1995) Taxonomic levels, in marine community studies, revisited. Mar Ecol Prog Ser 127: 113–119 Somerfield PJ, Clarke KR (1997) A comparison of some methods commonly used for the collection of sublittoral sediments and their associated fauna. Mar Environ Res 43: 145–156 Somerfield PJ, Jeal F (1995) Vertical distribution and substratum association of Halacaridae (Acari. Prostigmata) on sheltered and exposed Irish shores. J Nat Hist 29: 909–917 Somerfield PJ, Warwick RM (1996) Meiofauna in marine pollution monitoring programmes. A laboratory manual. Ministry of agriculture, fisheries and food. Directorate of Fisheries Research, Lowestoft, UK, p. 71 Somerfield PJ, Gee JM, Warwick RM (1994) Soft sediment meiofauna community structure in relation to a long-term heavy metal gradient in the Fal estuary system. Mar Ecol Prog Ser 105: 79–88 Somerfield PJ, Warwick RM, Moens T (2005) Meiofauna techniques. In: Eleftheriou A, McIntyre A (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 229–272 Somerfield PJ, Dashfield SL, Warwick RM (2007) Three-dimensional spatial structure: nematodes in a sandy tidal flat. Mar Ecol Prog Ser 336: 177–186 Somero GN, Childress JJ, Anderson AE (1989) Transport, metabolism, and detoxification of hydrogen sulfide in animals from sulfide-rich marine environments. Rev Aquat Sci 1: 591–614 Sommer S, Gutzmann E, Ahlrichs W, Pfannkuche O (2003) Rotifers colonising sediments with shallow gas hydrates. Naturwissensch 90: 273–276 Sommer S, Gutzmann E, Pfannkuche O (2007) Sediments hosting gas hydrates: oases for metazoan meiofauna. Mar Ecol Prog Ser 337: 27–37 Soosten C, Schmidt H, Westheide W (1998) Genetic variability and relationships among geographically widely separated populations of Petitia amphophthalma (Polychaeta: Syllidae). Results from RAPD-PCR investigations. Mar Biol 131: 659–669 Sopott B (1973) Jahreszeitliche Verteilung und Lebenszyklen der Proseriata (Turbellaria) eines Sandstrandes der Nordseeinsel Sylt. Mikrofauna Meeresboden 15: 253–358 Sopott-Ehlers B (1989) Coelogynopora visurgis nov. spec. (Proseriata) und andere freilebende Plathelminthes mariner Herkunft aus Ufersanden der Weser. Microfauna Mar 5: 87–93 Sørensen MV, Hebsgaard MB, Heiner I, Glenner H, Willerslev E, Kristensen RM (2008) New data from an enigmatic phylum: evidence from molecular sequence data supports a sistergroup relationship between Loricifera and Nematomorpha. J Zool Syst Evol Res (accepted) Souza-Santos LP, Castel J, Santos PJP (1996) The role of phototrophic sulfur bacteria as food for meiobenthic harpacticoid copepods inhabiting eutrophic coastal lagoons. Hydrobiologia 329: 79–89 Spindler M (1994) Notes on the biologiy of sea ice in the Arctic and Antarctic. Polar Biol 14: 319–324 Spindler M, Dieckmann GS (1994) Ecological significance of the sea ice biota. In: Hempel G (ed) Antarctic science. Springer, Berlin, pp. 60–68 Stal LJ (1991) The sulfur metabolism of mat-building cyanobacteria in anoxic marine sediments. Kiel Meeresforsch Sdh. 8: 152–157 Starink M, Krylova IN, Bär-Gilissen M-J, Bak RPM, Cappenberg TE (1994a) Rates of benthic protozoan grazing on free and attached sediment bacteria measured with fluorescently stained sediment. Appl Environ Microbiol 60: 2259–2264 Starink M, Bär-Gilissen MJ, Bak RPM, Cappenberg TE (1994b) Quantitative centrifugation to extract benthic Protozoa from freshwater sediments. Appl Environ Microbiol 60: 167–173
References
489
Starink M, Bär-Gilissen MJ, Bak RPM, Cappenberg TE (1996) Bacterivory by heterotrophic nanoflagellates and bacterial production in sediments of a freshwater littoral system. Limnol Oceanogr 41: 62–69 Staton JL, Schizas NV, Chandler GT, Coull BC, Quattro JM (2001) Ecotoxicology and population genetics: the emergence of “phylogeographic and evolutionary ecotoxicology”. Ecotoxicology 10: 217–222 Staton JL, Wickliffe LC, Garlitska L, Villanueva SM, BC C (2005) Genetic isolation discovered among previously described sympatric morphs of a meiobenthic copepod. J Crust Biol 25: 551–557 Stead TK, Schmid-Araya JM, Hildrew AG (2003) All creatures great and small: patterns in the stream benthos across a wide range of metazoan body size. Freshw Biol 48: 532–547 Stead TK, Schmid-Araya JM, Hildrew AG (2005) Secondary production of a stream metazoan community: does the meiofauna make a difference? Limnol Oceanogr 50: 398–403 Steele JH, Baird IE (1968) Production ecology of a sandy beach. Limnol Oceanogr 13: 14–25 Stephens GC (1982) Recent progress in the study of “Die Ernährung der Wassertiere und der Stoffhaushalt der Gewässer”. Am Zool 22: 611–619 Stephens GC, Volk MJ, Wright SH, Backlund PS (1978) Transepidermal transport of naturally occurring amino acids in the sand dollar, Dendraster excentricus. Biol Bull 154: 335–347 Sterrer W (1971) Gnathostomulida: Problems and procedures. Smiths Contr Zool 76: 9–15 Sterrer W (1972) Systematics and evolution within the Gnathostomulida. Syst Zool 21: 151–173 Sterrer W (1973) Plate tectonics as a mechanism for dispersal and speciation in interstitial sand fauna. Neth J Sea Res 7: 200–222 Sterrer W, Ax P (eds) (1977) The meiofauna species in time and space. Proceedings of a workshop symposium, Bermuda Biological Station, 1975. Microfauna Meeresboden 61:1–316 Sterrer W, Rieger R (1974) Retronectidae—a new cosmopolitan marine family of Catenulida (Turbellaria). In: Riser NW, Morse MP (eds) Biology of the Turbellaria. McGraw-Hill, New York, pp. 63–92 Steward CC, Pinckney J, Piceno Y, Lovell CR (1992) Bacterial numbers and activity, microalgal biomass and productivity, and meiofaunal distribution in sediments naturally contaminated with biogenic bromophenols. Mar Ecol Prog Ser 90: 61–72 Stewart MG (1979) Absorption of dissolved organic nutrients by marine invertebrates. Oceanogr Mar Biol Ann Rev 17: 163–192 Steyaert M, Herman PMJ, Moens T, Widdows J, Vincx M (2001) Tidal migration of nematodes on an estuarine tidal flat (the Molenplaat, Schelde Estuary, SW Netherlands). Mar Ecol Prog Ser 224: 299–304 Steyaert M, Moodley L, Nadong T, Moens T, Soetaert K, Vincx M (2007) Responses of intertidal nematodes to short-term anoxic events. J Exp Mar Biol Ecol 345: 175–184 Stock JH (1976) A new genus and two new species of the crustacean order Thermosbaenacea from the West Indies (with full bibliography and biogeographic notes). Bijdr Dierk 46: 47–70 Stock JH (1986) Deep sea origin of cave faunas, an unlikely supposition. Stygologia 2: 105–111 Stock JH (1994) Biogeographic synthesis of the insular groundwater faunas of the (sub)tropical Atlantic. Hydrobiologia 287: 105–117 Strawbridge S, Coull BC, Chandler GT (1992) Reproductive output of a meiobenthic copepod exposed to sediment-associated Fenvalerate. Arch Environ Contam Toxicol 23: 295–300 Strayer D (1985) The benthic micrometazoans of Mirror Lake, New Hampshire. Arch Hydrobiol Suppl 72: 287–426 Strayer D (1986) The size structure of a lacustrine zoobenthic community. Oecologia (Berl) 69: 513–516 Strayer D, Likens GE (1986) An energy budget for the zoobenthos of Mirror Lake, New Hampshire. Ecology 67: 303–313 Strayer DL (1991) Perspectives on the size structure of lacustrine zoobenthos, its causes, and its consequences. J N Am Benthol Soc 10: 210–221 Strayer DL (1994) Limits to biological distributions in groundwater. In: Gibert J, Danielopol DL, Stanford DL (eds) Groundwater ecology. Academic, San Diego, pp. 287–310
490
References
Strayer DL (2006) A beginner’s key to freshwater meiofauna. http://www.ecostudies.org/meiofauna_key.html Strayer DL, May SE, Nielsen P, Wollheim W, Hausam S (1995) An endemic groundwater fauna in unglaciated eastern North America. Can J Zool 73: 502–508 Strayer DL, May SE, Nielsen P, Wollheim W, Hausam S (1997) Oxygen, organic matter, and sediment granulometry as controls on hyporheic animal communities. Arch Hydrobiol 140: 131–144 Struck T, Hessling R, Purschke G (2002) The phylogenetic position of the Aelosomatidae and Parergordrilidae, two enigmatic oligochaete-like taxa of the ‘Polychaeta’, based on molecular data from 18S rDNA sequences. J Zool Syst Evol Res 40: 155–163 Struck T, Purschke G (2005) The sister group relationship of Aeolosomatidae and Potamodrilidae (Annelida). Zool Anz 243: 281–293 Struck TH (2006) Progenetic species in polychaetes (Annelida) and problems assessing their phylogenetic affiliation. Integr Comp Biol 46: 558–568 Suderman K, Thistle D (2004) The relative impacts of spills of two alternative fuels on the microalgae of a sandy site: a microcosm study. Mar Pollut Bull 49: 473–478 Suess E (1969) Interaction of organic compounds with calcium carbonate—I. Association phenomena and geochemical implications. Geochim Cosmochim Acta 34: 157–168 Suess E (1973) Interaction of organic compounds with calcium carbonate—II. Organo-carbonate association in recent sediments. Geochim Cosmochim Acta 37: 2435–2447 Sun B, Fleeger JW (1991) Spatial and temporal patterns of dispersion in meiobenthic copepods. Mar Ecol Prog Ser 71: 1–11 Sun B, Fleeger JW (1995) Sustained mass culture of Amphiascoides atopus, a marine harpacticoid copepod, in a recirculated system. Aquaculture 136: 313–321 Suresh K, Ahamed MS, Durairaj G, Nair KVK (1992) Ecology of interstitial meiofauna at Kalpakkam coast, east coast of India. Ind J Mar Sci 21: 217–219 Sutherland TF, Shepherd PCF, Elner RW (2000) Predation on meiofaunal and macrofaunal invertebrates by western sandpipers (Calidris mauri): evidence for dual foraging modes. Mar Biol 137: 938–993 Sutherland T, Levings C, Petersen S, Poon P, Piercey B (2007) The use of meiofauna as an indicator of benthic organic enrichment associated with salmonid aquaculture. Mar Pollut Bull 54: 1249–1261 Suzuki MT, Sherr EB, Sherr BF (1993) DAPI direct counting underestimates bacterial abundances and average cell size compared to AO direct counting. Limnol Oceanogr 38: 1566–1570 Swedmark B (1964) The interstitial fauna of marine sand. Biol Rev 39: 1–42 Swedmark B (1967) Gwynia capsula (Jeffreys), an articulate brachiopod with brood protection. Nature 213: 1151–1152 Swedmark B (1968) The biology of interstial Mollusca. Symp Zool Soc London 22: 135–149 Swedmark B (1971) A review of Gastropoda, Brachiopoda, and Echinodermata in marine meiobenthos. In: Hulings NC (ed) Proceedings Of the 1st International Conference On Meiofauna, Tunis, 1969. Smiths Contr Zool 76: 41–45 Swedmark B, Teissier G (1958) Armorhydra janowiczi, n.g. n.sp. Hydroméduse benthique. C R Acad Sci Paris 247: 133–135 Swinbanks DD, Shirayama Y (1986) High levels of natural radionuclides in a deep-sea infaunal xenophyophore. Nature 320: 354–358 Szymelfenig M, Kwasniewski S, Weslawski JM (1995) Intertidal zone of Svalbard. 2. Meiobenthos density and occurrence. Polar Biol 15: 137–141 Tabacchi E (1990) A sampler for interstitial fauna in alluvial rivers. Regul Rivers Res Manag 5: 177–182 Takashima Y, Mawatari SF (1998) Mitinokuidrilus excavatus n.g., n.sp., a marine tubificid (Oligochaeta) with a unique mode of reproduction. Zool Sci 15: 593–597 Tarjan AC, Keppner EJ (1999) Illustrated key to the genera of free-living marine nematodes in the superfamily Chromadoroidea exclusive of the Chromadoridae. UF/IFAS Extension Publ # EENY82, Gainsville, FL, p. 39 (see http://creatures.ifas.ufl.edu/nematode/marine_nematodes.htm)
References
491
Taylor RB, Blackburn RI, Evans JH (1995) A portable battery-powered suction device for the quantitative sampling of small benthic invertebrates. J Exp Mar Biol Ecol 194: 1–7 Taylor WR, Gebelein CD (1966) Plant pigments and light penetration in intertidal sediments. Helgol wiss Meeresunters 13: 229–237 Tchesunov AV, Portnova DA (2005) Free-living nematodes in seasonal coastal ice of the White Sea. Description of Hieminema obliquorum gen. et sp n. (Nematoda, Monhysteroidea). Zool Zh 84: 899–914 Teal JM, Kanwisher J (1961) Gas exchange in a Georgia salt marsh. Limnol Oceanogr 6: 388–399 Teal JM, Wieser W (1966) The distribution and ecology of nematodes in a Georgia salt marsh. Limnol Oceanogr 11: 217–222 Tempel D, Westheide W (1980) Uptake and incorporation of dissolved amino acids by the interstitial Turbellaria and Polychaeta and their dependence on temperature and salinity. Mar Ecol Prog Ser 3: 41–50 Tenore KR (1977a) Food chain pathways in detrital feeding benthic communities: a review,with new observations on sediment resuspension and detrital recycling. In: Coull BC (ed) Ecology of marine benthos. University of South Carolina Press, Columbia, SC, pp. 37–54 Tenore KR (1977b) Utilization of aged detritus derived from different sources by the polychaete Capitella capitata. Mar Biol 44: 51–55 Tenore KR, Rice DL (1980) A review of trophic factors affecting secondary production of deposit-feeders.. In: Tenore KR, Coull BC (eds) Marine benthic dynamics. University of South Carolina Press, Columbia, SC, pp. 325–340 Tenore KR, Tietjen JH, Lee JJ (1977) Effect of meiofauna on incorporation of aged eelgrass, Zostera marina, detritus by the polychaete Nephthys incisa. J Fish Res Board Can 34: 563–567 Tenore KR, Cammen L, Findlay SEG, Phillips N (1982) Perspectives of research on detritus: do factors controlling the availability of detritus to macroconsumers depend on its source?. J Mar Res 40: 473–490 Teuchert G (1977) The ultrastructure of the marine gastrotrich Turbanella cornuta Remane (Macrodasyoidea) and its functional and phylogenetical importance. Zoomorph 88: 189–246 Teuchert G (1978) Strukturanalyse von Bewegungsformen bei Gastrotrichen. Zool Jb Anat 99: 12–22 Teuchert G, Lappe A (1980) Zum sogenannten “Pseudocoel” der Nemathelminthes.—Ein Vergleich der Leibeshöhlen von mehreren Gastrotrichen. Zool Jb Anat 103: 424–438 Thiel H (1972) Meiofauna und Struktur der benthischen Lebensgemeinschaft des Iberischen Tiefseebeckens. Meteor Forsch-Ergebn D 12: 36–51 Thiel H (1975) The size structure of the deep-sea benthos. Int Rev Ges Hydrobiol 60: 575–606 Thiel H, Pfannkuche O, Schriever G, Lochte K, Gooday AJ, Hemleben C, Mantoura RFG, Turley CM, Patching JW, Riemann F (1988) Phytodetritus on the deep.sea floor in a central oceanic region of the Northeast Atlantic. Biol Oceanogr 6: 203–239 Thiermann R, Windoffer R, Giere O (1994) Selected meiofauna around shallow water hydrothermal vents off Milos (Greece). Ecological and ultrastructural aspects. Vie Milieu 44: 215–226 Thiermann F, Akoumianaki I, Hughes JA, Giere O (1997) Benthic fauna of a shallow-water gaseohydrothermal vent area in the Aegean Sea (Milos, Greece). Mar Biol 128: 149–159 Thiermann F, Vismann B, Giere O (2000) Sulphide tolerance of the marine nematode Oncholaimus campylocercoides—a result of internal sulphur formation? Mar Ecol Prog Ser 193: 251–259 Thistle D (1982) Aspects of the natural history of the harpacticoid copepods of San Diego Trough. Biol Oceanogr 1: 225–238 Thistle D (1988) Temporal difference in harpacticoid-copepod abundance at a deep-sea site: caused by benthic storms? Deep-See Res 35: 1015–1020 Thistle D, Eckman JE (1990) The effect of a biologically produced structure on the benthic copepods of a deep-sea site. Deep-Sea Res 37: 541–554 Thistle D, Sedlacek L (2004) Emergent and non-emergent species of harpacticoid copepods can be recognized morphologically. Mar Ecol Prog Ser 266: 195–200
492
References
Thistle D, Hilbig B, Eckman JE (1993) Are polychaetes sources of habitat heterogeneity for harpacticoid copepods in the deep sea? Deep-Sea Res 40: 151–157 Thistle D, Lambshead PJD, Sherman KM (1995a) Nematode tail-shape groups respond to environmental differences in the deep-sea. Vie Milieu 45: 107–115 Thistle D, Weatherly GL, Ertman SC (1995b) Shelf harpacticoid copepods do not escape into the seabed during winter storms. J Mar Res 53: 847–863 Thistle D, Levin LA, Gooday AJ, Pfannkuche O, Lambshead PJD (1999) Physical reworking by near-bottom flow alters the metazoan meiofauna of Fieberling Guyot (northeast Pacific). Deep-Sea Res I 46: 2041–2052 Thistle D, Carman KR, Sedlacek L, Brewer PG, Fleeger JW, Barry JP (2005) Deep-ocean experimental tests of the sensitivity of sediment-dwelling animals to imposed CO2 gradients. Mar Ecol Prog Ser 289: 1–4 Thistle D, Sedlacek L, Carman KR, Fleeger JW, Brewer PG, Barry JP (2006) Simulated sequestration of industrial carbon dioxide at a deep-sea site: effects on species of harpacticoid copepods. J Exp Mar Biol Ecol 330: 151–158 Thistle D, Sedlacek L, Carman KR, Fleeger JW, Brewer PG, Barry JP (2007a) Exposure to carbon dioxide-rich seawater is stressful for some deep-sea species: an in situ, behavioural study. Mar Ecol Prog Ser 340: 9–16 Thistle D, Sedlacek L, Carman KR, Fleeger JW, Barry JP (2007b) Emergence in the deep-sea: evidence from harpacticoid copepods. Deep-Sea Res I 54: 1008–1014 Thistle D, Eckman JE, Paterson GLJ (2008). Large, motile epifauna interact strongly with harpacticoid copepods and polychaetes at a bathyal site. Deep-Sea Res I 55: 324–331 Thomas ALR (1997) The breath of life—did increased oxygen levels trigger the Cambrian explosion? Trends Ecol Evolut 12: 44–45 Thomas DN, Dieckmann GS (2003) Sea ice. An introduction to its physics, chemistry, biology and geology. Blackwell, Oxford Thomas JD (1997) The role of dissolved organic matter, particularly free amino acids and humic substances, in freshwater ecosystems. Freshw Biol 38: 1–36 Thomas MB, Edwards NC, Higgins RP (1995) Cryptohydra thieli n.gen. n.sp. a meiofaunal marine hydroid (Hydroida, Athecata, Capitata). Invertebrate Biology 114: 107–118 Thomas MLH (1986) A physically derived exposure index for marine shorelines. Ophelia 25: 1–13 Thomsen L (1991) Treatment and splitting of samples for bacteria and meiofauna biomass determinations by means of a semi-automatic image analysis system. Mar Ecol Prog Ser 71: 301–306 Thomson L, Altenbach AV (1993) Vertical and areal distribution of foraminiferal abundance and biomass in microhabitats around inhabited tubes of marine echiurids. Mar Micropaleontol 20: 303–309 Tietjen JH (1980a) Microbial–meiofaunal interrelationships: a review. Proceedings of 8th Congress American Society for Microbiology, Washington, DC, pp. 335–338 Tietjen JH (1980b) Population structure and species distribution of the free-living nematodes inhabiting sands of the New York Bight apex. Estuar Coast Mar Sci 10: 61–73 Tietjen JH (1984) The use of free-living nematodes as a bioassay for estuarine sediments. Mar Environ Res 11: 233–251 Tietjen JH (1989) Ecology of deep-sea nematodes from the Puerto Rico Trench area and Hatteras abyssal plain. Deep-Sea Res 36: 1579–1594 Tietjen JH (1992) Abundance and biomass of metazoan meiobenthos in the deep sea. In: Rowe GT, Pariente V (eds) Deep-sea food chains and the global carbon cycle. Kluwer, Leiden, pp. 45–62 Tietjen JH, Lee JJ (1975) Axenic cultures and uptake of disolved organic substances by the marine nematode Rhabditis marina Bastian. Cah Biol Mar 16: 685–693 Tietjen JH, Lee JJ (1977) Feeding behaviour of marine nematodes. In: Coull BC (ed) Ecology of marine benthos. South Carolina Press, Columbia, SC, pp. 21–35 Timm T (1996) Oligochaeta of Lake Taimyr: a preliminary survey. Hydrobiologia 334: 89–95 Tipton K, Bell SS (1988) Foraging patterns of two syngnathid fishes: importance of harpacticoid copepods. Mar Ecol Prog Ser 47: 31–43
References
493
Tita G, Desrosiers G, Vincx M (2000a) New type of hand-held corer for meiofauna sampling and vertical profile investigation: a comparative study. J Mar Biol Ass UK 80: 171–172 Tita G, Desrosiers G, Vincx M, Nozais C (2000b) Predation and sediment disturbance effects of the intertidal polychaete Nereis virens (Sars) on associated meiofaunal assemblages. J Exp Mar Biol Ecol 243: 261–282 Tita G, Vincx M, Desrosiers G (1999) Size spectra, body width and morphotypes of intertidal nematodes: an ecological interpretation. J mar biol Ass UK 79: 1007–1015 Todaro MA (1998) Meiofauna from the Meloria Shoals: Gastrotricha, biodiversity and seasonal dynamics. Biol Mar Medit 5: 587–590 Todaro MA, Shirley TC (2003) A new meiobenthic priapulid (Priapulida, Tubiluchidae) from a Mediterranean submarine cave. Ital J Zool 70: 79–87 Todaro MA, Hummon WD (2008) An overview and a dichotomous key to genera of the phylum Gastrotricha. Meiofauna Mar 16: 3–20 Todaro MA, Fleeger JW, Hu YP, Hrincevich AW, Foltz DW (1996) Are meiofauna species cosmopolitan? Morphological and molecular analysis of Xenotrichula intermedia (Gastrotricha: Chaetonotida). Mar Biol 125: 735–742 Todaro MA, Bernard JM, Hummon WD (2000) A new species of Urodasys (Gastrotricha, Macrodasyida) from dysoxic sediments of the Santa Barbara Basin (California, U.S.A.). Bull Mar Sci 66: 467–476 Todaro MA, Leasi F, Bizzarri N, Tongiorgi P (2006) Meiofauna densities and gastrotrich community composition in a Mediterranean sea cave. Mar Biol 149: 1079–1091 Todo Y, Kitazato H (2005) Simple Foraminifera flourish at the ocean’s deepest point. Science 307: 4 Todorov M, Golemansky V (2007) Seasonal dynamics of the diversity and abundance of the marine interstitial testate Amoebae (Rhizopoda: Testacealobosia and Testaceafilosia) in the Black Sea supralittoral. Acta Protozool 46: 169–181 Tranvik LJ, Jørgensen NOG (1995) Colloidal and dissolved organic matter in lake water: carbohydrate and amino acid composition, and ability to support bacterial growth. Biogeochemistry 30: 77–97 Traunsburger W, Bergtold M, Goedkoop W (1997) The effects of nematodes on bacterial activity and abundance in a freshwater sediment. Oecologia (Berl) 112: 118–122 Traunspurger W (1997a) Distribution, seasonal occurence and vertical pattern of Tobrilus gracilis (Bastian, 1865) and T. medius (Schneider, 1916). Nematologica 43: 59–81 Traunspurger W (1997b) Bathymetric, seasonal and vertical distribution of feeding-types of nematodes in an oligotrophic lake. Vie Milieu 47: 1–7 Traunspurger W (2000) The biology and ecology of lotic nematodes. Freshw Biol 44: 29–45 Traunspurger W (2002) Nematoda. In: Rundle SD, Robertson AL, Schmid-Araya JM (eds) Freshwater meiofauna: Biology and ecology. Backhuys, Leiden, pp. 63–104 Traunspurger W, Drews C (1996) Toxicity analysis of freshwater and marine sediments with meio- and macrobenthic organisms: a review. Hydrobiologia 328: 215–261 Travizi A, Zavodnik N (2004) Phenology of Caulerpa taxifolia and temporal dynamics of its epibiontic meiofauna in the port of Malinska (Croatia, northern Adriatic Sea). Sci Mar 68, Suppl. 1: 145–154 Tseitlin VB, Mokievsky VO, Azovsky AI, Soltwedel T (2001) The study of meiobenthos size structure with the sieving method (case study of the Arctic Basin free-living nematodes). Oceanology 41: 712–717 Tsurumi M, Tunnicliffe V (2003) Tubeworm-associated communities at hydrothermal vents on the Juande Fuca Ridge, northeast Pacific. Deep-Sea Res I 50: 611–629 Tsurumi M, de Graaf RC, Tunnicliffe V (2003) Distributional and biological aspects of copepods at hydrothermal vents on the Juan de Fuca Ridge, north-east Pacific Ocean. J Mar Biol Ass UK 83: 469–477 Tudorancea C, Taylor WD (eds) (2002) Ethiopian rift valley lakes. Backhuys, Leiden, p. 284 Tudorancea C, Zullini A (1989) Associations and distribution of benthic nematodes in the Ethiopian Rift Valley lakes. Hydrobiologia 179: 81–96 Turbeville JM (2002) Progress in nemertean biology: Development and phylogeny. Integr Comp Biol 42: 692–703
494
References
Turner PN (1993) Distribution of rotifers in a Floridian saltwater beach, with a note of rotifer dispersal. Hydrobiologia 255: 435–439 Tyler S (1976) Comparative ultrastructure of adhesive systems in the Turbellaria. Zoomorphology 84: 1–76 Tyler S (1977) Ultrastructure and systematics: an example from turbellarian adhesive organs. Mikrofauna Meeresboden 61: 271–286 Tyler S, Hooge MD (2001) Musculature of Gnathostomula armata Riedl 1971 and its ecological significance. PSZN Mar Ecol 22: 71–83 Tzetlin AB, Saphonov MV (1995) A new finding of intracellular bacterial symbionts in the Nerillidae (Annelida: Polychaeta). Russ J Aquat Ecol 4: 55–60 Tzschaschel G (1979) Marine Rotatoria aus dem Interstitial der Nordseeinsel Sylt. Mikrofauna Meeresboden 71: 1–62 Tzschaschel G (1980) Verteilung, Abundanzdynamik und Biologie mariner interstitieller Rotatoria. Mikrofauna Meeresboden 81: 1–56 Udalov AA, Azovsky AI, Mokievsky VO (2005) Depth-related pattern in nematode size: What does depth itself really mean? Prog Oceanogr 67: 1–23 Uhlig G (1964) Eine einfache Methode zur Extraktion der vagilen mesopsammalen Mikrofauna. Helgol wiss Meeresunters 11: 178–185 Uhlig G, Heimberg SHH (1981) A new versatile compression chamber for examination of living microorganisms. Helgol Meeresunters 34: 251–256 Uhlig G, Thiel H, Gray JS (1973) The quantitative separation of meiofauna. A comparison of methods. Helgol wiss Meeresunters 25: 173–195 Ullberg J, Ólafsson E (2003) Free-living marine nematodes actively choose habitat when descending from the water column. Mar Ecol Prog Ser 260: 141–149 Underwood AJ (1989) The analysis of stress in natural populations. Biol J Linn Soc 37: 51–78 Underwood AJ, Chapman MG (2005) Design and analysis in benthic surveys. In: Eleftheriou A, McIntyre A (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 1–42 Underwood GJC, Paterson DM, Parks RJ (1995) The measurement of microbial carbohydrate exopolymers from intertidal sediments. Limnol Oceanogr 40: 1243–1253 Urban-Malinga B, Moens T (2006) Fate of organic matter in Arctic intertidal sediments: is utilisation by meiofauna important? J Sea Res 56: 239–248 Urban-Malinga B, Kotwicki L, Gheskiere TLA, Jankowska K, Opalinski K, Malinga M (2004) Composition and distribution of meiofauna, including nematode genera, in two contrasting. Arctic beaches Polar Biol 27: 447–457 Urban-Malinga B, Wiktor J, Jablonska A, Moens T (2005) Intertidal meiofauna of a high-latitude glacial Arctic fjord (Kongsfjorden, Svalbard) with emphasis on the structure of free-living nematode communities. Polar Biol 28: 940–950 Urban-Malinga B, Gheskiere T, Degraer S, Derycke S, Opalinski KW, Moens T (2008) Gradients in biodiversity and macroalgal wrack decomposition rate across a macrotidal, ultradissipative sandy beach. Mar Biol 155: 79–90 Uthicke S (1998) Population structure of Holothuria (Holodeima) atra (Jäger, 1833) and Stichopus chloronotus (Brandt, 1835) (Holothuroidea: Aspidochirotida) and their role in nutrient recycling in coral reef ecosystems. Ph.D. thesis, Univ Hamburg, Hamburg (unpubl) Uthicke S, Klumpp DW (1998) Microphytobenthos community production at near-shore coral reef: seasonal variation and response to ammonium recycled by holothurians. Mar Ecol Prog Ser 169: 1–11 Valesini FJ, Clarke KR, Eliot I, Potter IC (2003) A user-friendly quantitative approach to classifying nearshore marine habitats along a heterogeneous coast. Estuar Coast Shelf Sci 57: 163–177 Van Damme D, Heip C, Willems KA (1984) Influence of pollution on the harpacticoid copepods of two North Sea estuaries. Hydrobiologia 112: 143–160 Van de Velde MC, Coomans A (1989) A putative new hydrostatic skeletal function for the epidermis in monhysterids (Nematoda). Tissue and Cell 21: 525–534 Van der Land J (1970) Systematics, zoogeography and ecology of the Priapulida. Zool Verhandel 112: 1–118+115 pl
References
495
Van der Meer J, Heip CH, Herman PJM, Moens T, Van Oevelen D (2005) Measuring the flow of energy and matter in marine benthic animal populations. In: Eleftheriou A, McIntyre A (eds) Methods for the study of marine benthos. Blackwell, Oxford, pp. 326–407 Van Doninck K, Schön I, De Bruyn L, Martens K (2004) A general purpose genotype in an ancient asexual. Oecologia (Berl) 132: 205–212 Van Gaever S, Vanreusel A, Hughes JA, Bett BJ, Kiriakoulakis K (2004) The macro- and microscale patchiness of meiobenthos associated with the Darwin Mounds (north-east Atlantic). J Mar Biol Ass UK 84: 547–556 Van Gaever S, Moodley L, De Beer D, Vanreusel A (2006) Meiobenthos at the Arctic Håkon Mosby Mud Volcano, with a parental-caring nematode thriving in sulphide-rich sediments. Mar Ecol Prog Ser 321: 143–155 Van Gaever S, Galéron J, Sibuet M, Vanreusel A (2008) Deep-sea habitat heterogeneity influence on meiofaunal communities in the Gulf of Guinea Deep-Sea Res II (accepted) Van Gemerden HV, Tughan CS, Wit RD, Herbert RA (1989) Laminated microbial ecosystems on sheltered beaches in Scapa Flow, Orkney Islands. FEMS Microbiol Ecol 62: 87–102 Van Harten D (1992) Hydrothermal vent Ostracoda and faunal association in the deep-sea. DeepSea Res 39: 1067–1070 Van Hoek AHAM, van Alen TA, Sprakel VSI, Leunissen JAM, Brigge T, Vogels GD, Hackstein JHP (2000) Multiple acquisition of methanogenic archaeal symbionts by anaerobic ciliates. Mol Biol Evol 17: 251–258 Van Oevelen D, Middelburg JJ, Soetaert K, Moodley L (2006a) The fate of bacterial carbon in an intertidal sediment: modeling an in situ isotope tracer experiment. Limnol Oceanogr 51: 1302–1314 Van Oevelen D, Moodley L, Soetaert K, Middelburg JJ (2006b) The trophic significance of bacterial carbon in a marine intertidal sediment: results of an in situ stable isotope labeling study. Limnol Oceanogr 51: 2349–2359 Vanaverbeke J, Martinez Arbizu P, Dahms H-U, Schminke HK (1997) The metazoan meiobenthos along a depth gradient in the Arctic Laptev Sea with special attention to nematode communities. Polar Biol 18: 391–401 Vanaverbeke J, Steyaert M, Soetaert K, Rousseau V, Van Gansbeke D, Parent J-Y, Vincx M (2004) Changes in structural and functional diversity of nematode communities during a spring phytoplankton bloom in the southern North Sea. J Sea Res 52: 281–292 Vanaverbeke J, Franco MA, Lampadariou N, Muthumbi A, Steyaert M, Vandepitte L, Vanden Berghe E, Soetaert K (2007) Nematode morphometry from the shelf to the deep-sea in European marine water. In: Santos PJP dos (ed) Abstr 13th Int Meiofauna Conf, Recife, Brazil, O 21 Vandepitte L, Vanaverbeke J, Vanhoorne B, Hernandez F, Bezerra TN, Mees J, Vanden Berghe E (2008) The Manuela database: an integrated database on meiobenthos from European marine waters. Meiofauna Mar (submitted) Vanhove S (1993) Size spectra of nematode assemblages in an east African Mangrove. Acad Analecta 55: 129–142 Vanhove S, Vincx M, Van Gansbeke D, Gijselinck W, Schram D (1992) The meiobenthos of five mangrove vegetation types in Gazi Bay, Kenya. Hydrobiologia 247: 99–108 Vanhove S et al. (1995) Deep-sea meiofauna communities in Antarctica: structural analysis and relation with the environment. Mar Ecol Prog Ser 127: 65–76 Vanhove S, Wittoeck J, Beghyn M, Van Gansbeke D, Van Kenhove A, Coomans A, Vincx M (1997) Role of the meiobenthos in Antarctic ecosystems. In: Caschetto S (ed) Marine biogeochemistry and ecodynamics. Belgian Research Programme on the Antarctic. Scientific results of phase III (1992–1996), Brussels, pp. 1–59 Vanhove S, Lee HJ, Beghyn M, Van Gansbeke D, Brockington S, Vincx M (1998) The metazoan meiofauna in its biogeochemical environment: the case of an Antarctic coastal sediment. J Mar Biol Ass UK 78: 387–409 Vanhove S, Beghyn M, Van Gansbeke D, Bullough LW, Vincx M (2000) A seasonally varying biotope at Signy Island, Antarctic: implications for meiofaunal structure. Mar Ecol Prog Ser 202: 13–25
496
References
Vanhove S, Lee HJ, Van de Velde J, Dewicke A, Timmermann B, Janssens T, Vincx M (2003) Meiobenthic biodiversity and fluxes within the Antarctic biogeochemical environment. Instituut voor Dierkunde, Mariene Biologie, Gent, Belgium, Research Contract A4/DD/B01, 8–69 Vanreusel A (1990) Ecology of the free-living marine nematodes from the Voordelta (Southern Bight of the North Sea). 1. Species composition and structure of the nematode communities. Cah Biol Mar 31 (4): 439–462 Vanreusel A (1991) Ecology of the free-living marine nematodes from the Voordelta (Southern Bight of the North Sea). 2. Habitat preferences of the dominant species. Nematologica 37: 343–359 Vanreusel A, Vincx M, Bett BJ, Rice AL (1995a) Nematode biomass spectra at two abyssal sites in the NE Atlantic with a contrasting food supply. Int Rev Ges Hydrobiol 80: 287–296 Vanreusel A, Vincx M, Schram D, van Gansbeke D (1995b) On the vertical distribution of the metazoan meiofauna in shelf break and upper slope habitats of the NE Atlantic. Int Rev Ges Hydrobiol 80: 313–326 Vanreusel A, van den Bossche I, Thiermann F (1997) Free-living marine nematodes from hydrothermal sediments: similarities with communities from diverse reduced habitats. Mar Ecol Prog Ser 157: 207–219 Varó I, Taylor AC, Amat F (1993) Comparison of two methods for measuring the rates of oxygen consumption of small aquatic animals (Artemia). Comp Biochem Physiol A 106: 551–555 Vassalo P, Fabiano M, Vezzulli L, Sandulli R, Marques JC, Jørgensen SE (2006) Assessing the health of coastal ecosystems: a holistic approach based on sediment micro- and meiobenthic measures. Ecol Indicat 6: 525–542 Veit-Köhler G (2004) Kliopsyllus andeep sp. n. (Copepoda: Harpacticoida) from the Antarctic deep sea—a copepod closely related to certain shallow-water species. Deep-Sea Res II 51: 1629–1641 Veit-Köhler G, Gerdes D, Quiroga E, Hebbeln D, Sellanes J (2009) Metazoan meiofauna within the oxygen-minimum zone off Chile: results of the 2001-PUCK expedition. Deep-Sea Research II (in press) Vernberg WB and Coull BC (1974) Respiration of an interstitial ciliate and benthic energy relationships. Oecologia (Berl) 16: 259–264 Vernberg WB, Coull BC, Jorgensen DD (1977) Reliability of laboratory metabolic measurements of meiofauna. J Fish Res Board Can 34: 164–167 Viets K (1927) Die Halacaridae der Nordsee. Z wiss Zool 130: 83–173 Villiers L, Bodiou JY (1996) Community structure of harpacticoid copepods in a tropical reef lagoon (Fangataufa Atoll-French Polynesia). Oceanol Acta 19: 155–162 Villora-Moreno S (1996a) Ecology and distribution of the Diurodrilidae (Polychaeta), with redescription of Diurodrilus benazzii. Cah Biol Mar 37: 99–108 Villora-Moreno S (1996b) A new genus and species of the deep-sea family Coronarctidae (Tardigrada) from a submarine cave with a deep-sea like condition. Sarsia 81: 275–283 Villora-Moreno S (1997) Environmental heterogeneity and the biodiversity of interstitial Polychaeta. Bull Mar Sci 60: 494–501 Vincx M, Meire P, Heip C (1990) The distribution of nematode communities in the Southern Bight of the North Sea. Cah Biol Mar 31: 107–129 Vincx M, Bett BJ, Dinet A, Ferrero T, Gooday AJ, Lambshead PJD, Pfannkuche O, Soltwedel T, Vanreusel A (1994) Meiobenthos of the deep northeast Atlantic. Adv Mar Biol 30: 1–88 Visman B (1991) Sulfide tolerance: physiological mechanisms and ecological implications. Ophelia 34: 1–28 Visman B (1996) Sulfide species and total sulfide toxicity in the shrimp Crangon crangon. J Exp Mar Biol Ecol 204: 141–154 Visscher PT, Beukema J, van Gemerden H (1991) In situ characterization of sediments. Measurement of oxygen and sulfide profiles with a novel combined needle electrode. Limnol Oceanogr 36: 1476–1479 Voigt M, Koste W (1978) Rotatoria (Überordnung Monogononta). Die Rädertiere Mitteleuropas. Ein Bestimmungswerk. 2nd edn, vols 1+2. Borntraeger, Berlin, p. 1160
References
497
Volk CJ, Volk CB, Kaplan LA (1997) The chemical composition of biodegradable dissolved organic matter in streamwater. Limnol Oceanogr 42: 39–44 Völkel S, Grieshaber MK (1996) Mitochondrial sulfide oxidation in Arenicola marina. Evidence for alternative electron pathways. Eur J Biochem 235: 231–237 Volz DC, Chandler GT (2004) An enzyme-linked immunosorbent assay for lipovitellin quantification in copepods: a screening toll for endocrine toxicity. Environ Toxicol Chem 23: 298–305 Vopel K, Dehmlow J, Arlt G (1996) Vertical distribution of Cletocamptus confluens (Copepoda, Harpacticoida) in relation to oxygen and sulphide microprofiles of a brackish water sulphuretum. Mar Ecol Prog Ser 141: 129–137 Vorobyova LV (1999) Meiobenthos of the Ukrainian shelf of the Black Sea and the Sea of Azov. (in Russian) Naukova Dumka, Kiev, p. 300 Vranken G, Heip C (1986) The productivity of marine nematodes. Ophelia 26: 429–442 Vranken G, Tiré C, Heip C (1988a) The toxicity of paired metal mixtures to the nematode Monhystera disjuncta (Bastian, 186). Mar Environ Res 26: 161–179 Vranken G, Herman PMJ, Heip C (1988b) Studies of the life-history and energetics of marine and brackish-water nematodes. I. Demography of Monhystera disjuncta at different temperature and feeding conditions. Oecologia (Berl) 77: 296–301 Wägele J-W, Misof B (2001) On quality of evidence in phylogeny reconstruction: a reply to Zrzavy’s defence of the ‘Ecdysozoa’ hypothesis. J Zool Syst Evol Res 39: 165–176 Wagner HP (1994) A monographic review of the Thermosbaenacea (Crustacea: Peracarida). A study on their morphology, taxonomy, phylogeny and biogeography. Zool Verh Leiden 291: 1–338 Walossek D (1996) Rehbachiella, der bisher älteste Branchiopode. Stapfia 42: 21–28 Walossek D, Müller KL (1990) Upper Cambrian stem-lineage crustaceans and their bearing upon the monophyletic origin of Crustacea and the position of Agnostus. Lethaia 23: 409–427 Waloszek D, Chen J, Maas A, Wang X (2005) Early Cambrian arthropods—new insights into arthropod head and structural evolution. Arthrop Struct Dev 34: 189–205 Walters K (1991) Influences of abundance, behavior, species composition, and ontogenetic stage on active emergence of meiobenthic copepods in subtropical habitats. Mar Biol 108: 207–215 Walters K, Bell SS (1986) Diel pattern of active vertical migration in seagrass meiofauna. Mar Ecol Prog Ser 34: 95–103 Walters K, Bell SS (1994) Significance of copepod emergence to benthic, pelagic and phytal linkages in a subtidal seagrass bed. Mar Ecol Prog Ser 108: 237–249 Walters K, Moriarty DJW (1993) The effects of complex trophic interactions on a marine microbenthic assemblage. Ecology 74: 1475–1489 Walters K, Shanks AL (1996) Complex trophic and non-trophic interactions between meiobenthic copepods and marine snow. J Exp Mar Biol Ecol 198: 131–145 Walters K, Jones E, Etherington L (1996) Experimental studies of predation on metazoans inhabiting Spartina alterniflora stems. J Exp Mar Biol Ecol 198: 251–265 Ward JV, Palmer MA (1994) Distribution patterns of interstitial freshwater meiofauna over a range of spatial scales, with emphasis on alluvial river-aquifer systems. Hydrobiologia 287: 147–156 Ward JV, Voelz NJ (1990) Gradient analysis of interstitial meiofauna along a longitudinal stream profile. Stygologia 5: 93–99 Warén A, Gofas S (1996) A new species of Monoplacophora, redescription of the genera Veleropilina and Rokopiella, and new information on three species of the class. Zool Scr 25: 215–232 Warwick RM (1984) Species size distribution in marine benthic communities. Oecologia (Berl) 61: 32–41 Warwick RM (1986) A new method for detecting pollution effects on marine macrobenthic communities. Mar Biol 92: 557–562 Warwick RM (1987) Meiofauna—their role in marine detrital systems. In: Moriarty DJW, Pullin, RSV (eds) Detritus and microbial ecology in aquaculture. Bellagio, Como, Italy, pp. 282–295 Warwick RM (1988a) The level of taxonomic discrimination required to detect pollution effects on marine benthic communities. Mar Pollut Bull 19: 259–268
498
References
Warwick RM (1988b) Effects on community structure of a pollutant gradient—summary. Mar Ecol Prog Ser 46: 207–211. Warwick RM (1989) The role of meiofauna in the marine ecosystem: evolutionary considerations. Zool J Linn Soc 96: 229–241 Warwick RM (1993) Environmental impact studies on marine communities: pragmatical considerations. Aust J Ecol 18: 63–80 Warwick RM (2000) Are Loriciferans paedomorphic (progenetic) priapulids? Vie Milieu 50: 191–193 Warwick RM, Clarke KR (1993a) Comparing the severity of disturbance: a meta-analysis of marine macrobenthic community data. Mar Ecol Prog Ser 92: 221–231 Warwick RM, Clarke KR (1993b) Increased variability as a symptom of stress in marine communities. J exp mar Biol Ecol 172: 215–226. Warwick RM, Clarke KR (1995) New ‘biodiversity’ measures reveal a decrease in taxonomic distinctness with increasing stress. Mar Ecol Prog Ser 129: 301–305 Warwick RM, Clarke KR (2001) Practical measures of marine biodiversity based on relatedness of species. Oceanogr Mar Biol Ann Rev 39: 207–231 Warwick RM, Gee JM (1984) Community structure of estuarine meiobenthos. Mar Ecol Prog Ser 18: 97–111 Warwick RM, Price R (1979) Ecological and metabolic studies on free-living nematodes from an estuarine mud-flat. Estuar Coast Mar Sci 9: 257–271 Warwick RM, Joint IR, Radford PJ (1979) Secondary production of the benthos in an estuarine environment. In: Jeffries RLD, Davy AJ (eds) Ecological processes in coastal environments. Blackwell, Oxford, pp. 429–450 Warwick RM, Collins NR, Gee JM, George CL (1986a) Species size distributions of benthic and pelagic Metazoa: evidence for interaction? Mar Ecol Prog Ser 34: 63–68 Warwick RM, Gee JM, Berge JA, Ambrose jr W (1986b) Effects of the feeding activity of the polychaete Streblosoma bairdi (Malmgren) on meiofaunal abundance and community structure. Sarsia 71: 11–16 Warwick RM, Carr MR, Clarke KR, Gee JM, Green RH (1988) A mesocosm experiment on the effects of hydrocarbon and copper pollution on a sublittoral soft-sediment meiobenthic community. Mar Ecol Prog Ser 46: 181–191 Warwick RM, Platt HM, Clarke KR, Agard J, Gobin J (1990a) Analysis of macrobenthic and meiobenthic community structure in relation to pollution and disturbance in Hamilton Harbour, Bermuda. J Exp Mar Biol Ecol 138: 119–142 Warwick RM, Clarke KR, Gee JM (1990b) The effect of disturbance by soldier crabs Mictyris platycheles H. Milne Edwards on meiobenthic community structure. J Exp Mar Biol Ecol 135: 19–33 Warwick RM, Platt HM, Somerfield PJ (1998) Free-living marine nematodes. Pt. 3. Monhysterids. Linnean Society, London, p. 296 (Synopses of the British Fauna, vol 53) Warwick RM, Dashfield SL, Somerfield PJ (2006) The integral structure of a benthic infaunal assemblage. J Exp Mar Biol Ecol 330: 12–18 Wasmund N (1989) Live algae in deep sediment layers. Int Rev Ges Hydrobiol 74: 589–597 Watanabe H, Iguchi T (2006) Using ecotoxicogenomics to evaluate the impact of chemicals on aquatic organisms. Mar Biol 149: 107–115 Watanabe T, Kitajima C, Fujita S (1983) Nutritional values of live organisms used in Japan for mass propagation of fish: a review. Aquaculture 34: 115–143 Watling L (1988) Small-scale features of marine sediments and their importance to the study of deposit-feeding. Mar Ecol Prog Ser 47: 135–144 Watling L (1991) The sedimentary milieu and its consequences for resident organisms. Am Zool 31: 789–796 Watzin MC (1985) Interactions among temporary and permanent meiofauna: observations on the feeding and behavior of selected taxa. Biol Bull 169: 397–416 Watzin MC (1986) Larval settlement into marine soft-sediment systems: interactions with the meiofauna. J Exp Mar Biol Ecol 98: 65–113 Webb DG, Parsons TR (1991) Impact of predation-disturbance by large epifauna on sediment-dwelling harpacticoid copepods: field experiments in a subtidal seagrass bed. Mar Biol 109: 485–491
References
499
Weber M, Faerber P, Meyer V, Lott C, Eickert G, Fabricius KE, De Beer D (2007) In situ applications of a new diver-operated motorized microsensor profiler. Environ Sci Technol 41: 6210–6215 Webster IT (1992) Wave enhancement of solute exchange within empty burrows. Limnol Oceanogr 37: 630–643 Weinstein F (1961) Psammostyela delamarei n.g. n.sp.: Ascidie interstitielle des sables à Amphioxus. C R Acad Sci Paris, Sér D 252: 1843–1844 Weise W, Rheinheimer G (1978) Scanning electron microscopy and epifluorescense investigations of bacterial colonization of marine sand sediments. Microb Ecol 4: 175–188 Weiss MJ (2001) Widespread hermaphroditism in freshwater gastrotrichs. Invert Biol 120: 308–341 Weissenberger J, Dieckmann G, Gradinger R, Spindler M (1992) Sea ice—a cast technique to examine and analyze brine pockets and channel structure. Limnol Oceanogr 37: 179–183 Wells JBJ (1971) A brief review of methods of sampling the meiobenthos. In: Hulings NC (ed) Proceedings of First International Conference on Meiofauna. Smiths Contr Zool 76:183–186 Wells JBJ (2007) An annotated checklist and keys to Copepoda Harpacticoida (Crustacea). Zootaxa 1568: 1–872 Wentworth CK (1922) A scale of grade and class terms for clastic sediments. J Geol 30: 377–392 Wenzhöfer F, Glud RN (2004) Small-scale spatial and temporal variability in coastal benthic oxygen dynamics: effects of fauna activity. Limnol Oceanogr 49: 1471–1481 Werner I (2005) Seasonal dynamics, cryo-pelagic interactions and metabolic rates of Arctic packice and under-ice fauna. A review. Polarforsch/Polar Res 75: 1–19 Westheide W (1968) Zur quantitativen Verteilung von Bakterien und Hefen in einem Gezeitenstrand der Nordseeküste. Mar Biol 1: 336–347 Westheide W (1972) Räumliche und zeitliche Differenzierungen im Verteilungsmuster der marinen Interstitialfauna. Verh Dt Zool Ges 65: 23–32 Westheide W (1979) Hesionides riegerorum n.sp. a new interstitial freshwater polychaete from the United States. Int Rev Ges Hydrobiol 64: 273–280 Westheide W (1981) Interstitielle Fauna von Galapagos. XXVI. Questidae, Cirratulidae, Acrocirridae, Ctenodrilidae (Polychaeta). Mikrofauna Meeresboden 82: 59–79 Westheide W (1984) The concept of reproduction in polychaetes with small body size: adaptations in interstitial species. Fortschr Zool 29: 265–287 Westheide W (1985) The systematic position of the Dinophilidae and the archiannelid problem. In: Conway MS, George JD, Gibson R, Platt HM (eds) The origins and relationships of lower invertebrates. Clarendon, Oxford., pp. 310–326 Westheide W (1987a) Progenesis as a principle in meiofauna evolution. J Nat Hist 21: 843–854 Westheide W (1987b) The interstitial polychaete Hesionides pettiboneae n.sp. (Hesionidae) from the U.S. East coast and its transatlantic relationships. Biol Soc Wash Bull 7: 131–139 Westheide W (1990) Polychaetes: interstitial families. Key and notes for the identification of the species. Universal Book Services/Dr. W. Backhuys, Qegstgeest. The Netherlands, p. 152 (Synopses of the British Fauna, vol 44) Westheide W (1991) The meiofauna of the Galápagos. A review. In: James MJ (ed) Galápagos marine invertebrates. Taxonomy, biogeography, and evolution in Darwin’s islands. Plenum, New York, pp. 37–73 Westheide W, Haß-Cordes E, Krabusch M, Müller MCM (2003) Ctenodrilus serratus (Polychaeta: Ctenodrilidae) is a truly amphi-Atlantic meiofauna species — evidence from molecular data. Mar Biol 142: 637–642 Westheide W, Purschke G, Mangerich W (1994) Sinohesione genitaliphora gen. et sp.n. (Polychaeta, Hesionidae), an interstitial annelid with unique dimorphous external genital organs. Zool Scr 23: 95–105 Westphalen D (1993) Stromatolithoid microbial nodules from Bermuda — a special micro habitat for meiofauna. Mar Biol 117: 145–157 Wetzel M, Jensen P, Giere O (1995) Oxygen/ sulfide regime and nematode fauna associated with Arenicola — burrows: new insights in the thiobios case. Mar Biol 124: 301–312 Wetzel MA, Fleeger JW, Powers SP (2001) Effects of hypoxia and anoxia on meiofauna: a review with new data from the Gulf of Mexico. In: Rabalais NN, Turner RE (eds) Coastal hypoxia:
500
References
consequences for living resources and ecosystems. American Geophysical Union, Coastal and Estuarine Studies, Washington, DC, pp. 165–184 Wetzel MA, Weber A, Giere O (2002) Re-colonization of anoxic/sulfidic sediments by marine nematodes after experimental removal of macroalgal cover. Mar Biol 141: 679–689 Wharton DA (1986) A functional biology of nematodes. Chapman & Hall, London, p. 224 Whiteside MK, Williams JB, White CP (1978) Seasonal abundance and pattern of chydorid Cladocera in mud and vegetative habitats. Ecology 59: 1177–1188 Whitlatch RB, Zajac RN (1985) Biotic interactions among estuarine infaunal opportunistic species. Mar Ecol Prog Ser 21: 299–311 Whitman RL, Clark WJ (1984) Ecological studies of the sand-dwelling community of an east Texas stream. Freshwater Invert Biol 3: 59–79 Wickham SA, Gieseke A, Berninger U-G (2000) Benthic ciliate identification and enumeration: an improved methodology and its application. Aquat Microb Ecol 22: 79–91 Widbom B (1984) Determination of average individual dry weights and ash-free dry weights in different sieve fractions of marine meiofauna. Mar Biol 84: 101–108 Widbom B, Frithsen JB (1995) Structuring factors in a marine soft bottom community during eutrophication—an experiment with radio-labelled phytodetritus. Oecologia (Berl) 101: 156–16 Widdows JA, Brinsley MD, Salkeld PN, Lucas CH (2000) Influence of biota on spatial and temporal variation in sediment erodability and material flux on a tidal flat (Westerschelde, The Netherlands). Mar Ecol Prog Ser 194: 23–37 Wiederholm T, Erikson L (1977) Effects of alcohol preservation on the weight of some benthic invertebrates. Zoon 5: 29–31 Wieser W (1953) Die Beziehung zwischen Mundhöhlengestalt, Ernährungsweise und Vorkommen bei freilebenden marinen Nematoden. Ark Zool 2: 439–484 Wieser W (1959a) The effect of grain size on the distribution of small invertebrates inhabiting the Beaches of Puget Sound. Limnol Oceanogr 4: 181–194 Wieser W (1959b) Zur Ökologie der Fauna mariner Algen mit besonderer Berücksichtigung des Mittelmeeres. Int Rev Ges Hydrobiol 44: 137–180 Wieser W (1960) Benthic studies in Buzzards Bay. II. The meiofauna. Limnol Oceanogr 5: 121–137 Wieser W, Kanwisher J (1961) Ecological and physiological studies on marine nematodes from a small salt marsh near Woods Hole, Massachusetts. Limnol Oceanogr 6: 262–270 Wieser W, Ott J, Schiemer F, Gnaiger E (1974) An ecophysiological study of some meiofauna species inhabiting a sandy beach at Bermuda. Mar Biol 26: 235–248 Wigley RL, McIntyre AD (1964) Some quantitative comparisons of offshore meiobenthos and macrobenthos south of Martha’s Vineyard. Limnol Oceanogr 9: 485–493 Wild C, Rasheed M, Jantzen C, Cook P, Stuck U, Huettel M, Boetius A (2005) Benthic metabolism and degradation of natural particulate organic matter in carbonate and silicate reef sands of the northern Red Sea. Mar Ecol Prog Ser 298: 69–78 Wilkens H, Culver DC, Humphreys WF (eds) (2000) Subterranean ecosystems. In: Goodall DW (ed) Ecosystems of the world 30. Elsevier, Amsterdam, p. 789 Wilkens H, Parzefall J, Iliffe TM (1986) Origin and age of the marine stygofauna of Lanzarote, Canary Islands. Mitt Hamb Zool Mus Inst 83: 223–230 Will KP, Mishler BD, Wheeler QD (2005) The perils of DNA barcoding and the need for integrative taxonomy. System Biol 54: 844–851 Williams DD (1984) The hyporheic zone as habitat for aquatic insects and associated arthropods. In: Resh VH, Rosenberg DM (eds) The ecology of aquatic insects. Praeger, New York, pp. 430–455 Williams DD (1989) Towards a biological and chemical definition of the hyporheic zone in two Canadian rivers. Freshwat Biol 22: 189–208 Williams DD (2003) The brackish water hyporheic zone: invertebrate community structure across a novel ecotone. Hydrobiologia 510: 153–173 Williams DD, Hynes HBN (1974) The occurrence of benthos deep in the substratum of a stream. Freshwat Biol 4: 233–256
References
501
Williams DD, Williams NE (1974) A counterstaining technique for use in sorting benthic samples. Limnol Oceanogr 19: 152–154 Williams PL, Dusenbery DB (1990) Aquatic toxicity testing using the nematode, Caenorhabditis elegans. Environ Tox Chem 9: 1285–1290 Williams R (1971) A technique for measuring the interstitial voids of a sediment based on epoxy resin impregnation. In: Hulings NC (ed) Proceedings of First International Conference on Meiofauna. Smiths Contr Zool 76:199–205 Williams R (1972) The abundance and biomass of the interstitial fauna of a graded series of shell gravels in relation to available space. J Anim Ecol 41: 623–646 Williams-Howze J (1997) Dormancy in the free-living copepod orders Cyclopoida, Calanoida, and Harpacticoida. Oceanogr Mar Biol Annu Rev 35: 257–321 Williams-Howze J, Coull BC (1992) Are temperature and photoperiod necessary cues for encystment in the marine benthic harpacticoid Heteropsyllus nunni Coull? Biol Bull 182: 109–116 Williams-Howze J, Fleeger JW (1987) Pore pattern: a possible indicator of tube-building in Stenhelia and Pseudostenhelia (Copepoda: Harpacticoida). J Crust Biol 7: 148–157 Williamson FA, Palframan KR (1989) An improved method for collecting and staining microorganisms for enumeration by fluorescence light microscopy. J Microscopy 154: 267–272 Wilson TRS (1978) Evidence for denitrification in aerobic pelagic sediments. Nature 274: 354–356 Wilson WH (1991) Competition and predation in marine soft-sediment communities. Annu Rev Ecol Syst 21: 221–241 Wiltshire KH (2000) Algae and associated pigments in intertidal sediments, new observations and methods. Limnologica 30: 205–214 Wiltshire KH, Blackburn J, Paterson DM (1997) The cryolander: a new method for fine-scale in situ sampling of intertidal surface sediments. J Sed Res 67: 977–981 Winberg GG (1971) Methods of estimation of production of marine animals. Academic, London Winner RW, Boesel MW, Farrell MP (1980) Insect community structure as an index of heavymetal pollution in lotic ecosystems. Can J Fish Aquat Sci 37: 647–655 Wirth EF, Chandler GT, DiPinto LM, Bidleman TF (1994) Assay of polychlorinated biphenyl bioaccumulation from sediments by marine benthic copepods using a novel microextraction technique. Environ Sci Technol 28: 1609–1614 Wirz A, Pucciarelli S, Micelli C, Tongiorni P, Balsamo M (1999) Novelty in phylogeny of Gastrotricha: evidence from 18SrRNA gene. Mol Phylogen Evol 13: 314–318 Wiszniewski J (1934) Recherches écologiques sur le psammon et spécialement sur les rotifères psammiques (in Polish and French). Archw Hydrobiol Rybact 8: 149–271 Witte U et al. (2003) In situ experimental evidence of the fate of a phytodetritus pulse at the abyssal sea floor. Nature 424: 763–766 Woodin SA, Jackson JBC (1979a) Facilitative and inhibitory interactions among estuarine meiobenthic harpacticoid copepods. Ecology 68: 1906–1919 Woodin SA, Jackson JBC (1979b) Interphyletic competition among marine benthos. J Mar Res 34: 25–41 Woodward G, Ebenman B, Emmerson M, Montoya JM, Olesen JM, Valido A, Warren PH (2005) Body size in ecological networks. Trends Ecol Evolut 20: 402–409 Wormald AP (1976) Effects of a spill of marine diesel oil on the meiofauna of a sandy beach at Picnic Bay, Hong Kong. Environ Pollut 11: 117–130. Worsaae K, Kristensen RM (2005) Evolution of interstitial Polychaeta (Annelida). Hydrobiologia 535536: 319–340 Worsaae K, Sterrer W (2006) Description of two new interstitial species of Psammodrilidae (Annelida) from Bermuda. Mar Biol Res 2: 431–445 Wotton RS (2005) The essential role of exopolymers (EPS) in aquatic systems. In: Gibson RN, Atkinson RJA, Gordon JDM (eds) Oceanogr Mar Biol Annu Rev. pp. 57–94 Woyke T et al. (2006) Symbiosis insights through metagenomic analysis of a microbial consortium. Nature 443: 950–955 Wray GA, Levinton JS, Shapiro LH (1996) Molecular evidence for deep Precambrian divergences among metazoan phyla. Science 274: 568–573
502
References
Wright JC (2001) Cryptobiosis 300 years on from van Leuwenhoek: what have we learned about tardigrades?. Zool Anz 240: 563–582 Wu J, Fu C, Liang Y, Chen J (2004) Distribution of the meiofaunal community in a eutrophic shallow lake of China. Arch Hydrobiol 159: 555–575 Wu RSS (2002) Hypoxia from molecular responses to ecosystem responses. Mar Pollut Bull 45: 35–45 Yallop ML, de Winder B, Paterson DM, Stal LJ (1994) Comparative structure, primary production and biogenic stabilization of cohesive and non-cohesive marine sediments inhabited by microphytobenthos. Estuar Coast Shelf Sci 39: 565–582 Yamaguchi S, Endo K (2003) Molecular phylogeny of Ostracoda (Crustacea) inferred from 18S ribosomal DNA sequences: implication for its origin and diversification. Mar Biol 143: 23–38 Yamamuro M (2000) Abundance and size distribution of sublittoral meiobenthos along estuarine salinity gradients. J Mar Syst 26: 135–143 Yingst JY (1978) Patterns of micro- and meiofaunal abundance in marine sediments, measured with adenosine triphosphate assay. Mar Biol 47: 41–54 Yingst JY, Rhoads DC (1980) The role of bioturbation in the enhancement of bacterial growth in marine sediments. In: Tenore KR, Coull BC (eds) Marine benthic dynamics. University of South Carolina Press, Columbia, SC, pp. 407–421 Yozzo DJ, Smith DE (1995) Seasonality, abundance, and microhabitat distribution of meiofauna from a Chickahominy River, Virginia tidal freshwater marsh. Hydrobiologia 310: 197–206 Zajcev JP, Ancupova LV, Vorobyoya LV, Garkayaya GP, Kulakova II, Rusnak EM (1987) Nematodes from the deep-water zone of the Black Sea (in Russian). Dokl Acad Sci Ukr SSR, Ser B, Geol Chim Biol Nauki 11: 77–79 Zander CD (1993) The distribution and feeding ecology of small-size epibenthic fish in the coastal Mediterranean Sea. In: Eleftheriou A, Ansell AD, Smith CJ (eds) Biology and ecology of shallow coastal waters. Olsen & Olsen, Fredensborg, Denmark, pp. 369–376 Zekely J, Van Dover CL, Nemeschkal HL, Bright M (2006a) Hydrothermal vent meiobenthos associated with mytilid mussel aggregations from the Mid-Atlantic Ridge and the East Pacific Rise. Deep-Sea Res I 53: 1363–1378 Zekely J, Gollner S, Van Dover CL, Govenar B, Le Bris N, Bright M (2006b) The nematode community and trophic structure of three macrofaunal aggregations at 9° & 11°N East Pacific Rise. Cah Biol Mar 47: 477–482 Zeppilli D, Danovaro R (2007) Meiobenthic diversity in a shallow hydrothermal vent of the Pacific Ocean (North Sulawesi — Indonesia). In: Santos PJP dos (ed) Abstr 13th Int Meiofauna Conf, Recife, Brazil, P 7 Ziebis W, Huettel M, Forster S (1996) Impact of biogenic sediment topography on oxygen fluxes in permeable seabeds. Mar Ecol Prog Ser 140: 227–237 Zimmermann R, Iturriaga R, Becker-Birck J (1978) Simultaneous determination of the total number of aquatic bacteria and the number thereof involved in respiration. Appl Environ Microbiol 36: 926–935 Zinn DS, Found BW, Kraus MG (1982) A bibliography of the Mystacocarida. Crustaceana 43: 270–274 Zobrist EC, Coull BC (1992) Meiobenthic interactions with macrobenthic larvae and juveniles: an experimental assessment of the meiofaunal bottleneck. Mar Ecol Prog Ser 88: 1–8 Zobrist EC, Coull BC (1994) Meiofaunal effects on growth and survivorship of the polychaete Streblospio benedicti Webster and the bivalve Mercenaria mercenaria (L). J Exp Mar Biol Ecol 175: 167–179 Zrzavy J (2001) Ecdysozoa versus Articulata: clades, artifacts, prejudices. J Zool Syst Evol Res 39: 159–163 Zrzavy J (2003) Gastrotricha and metazoan phylogeny. Zool Scr 32: 61–82 Zühlke R, Blome D, Bernem KH, Dittmann S (1998) Effects of the tube-building polychaete Lanice conchilega (Pallas) on benthic macrofauna and nematodes in an intertidal sandflat. Senckenberg Mar 29: 131–138 Zullini A (1976) Nematodes as indicators of river pollution. Nematol Mediterr: 13–22
Glossary
The following abbreviations are used in this glossary: adj.: adjective; contr.: contrary term; corr.: corresponding; subst.: substantive; syn.: synonym. Abyssal Allopatric Alveolar (adj.) Amensalism Amphiatlantic Anchi(h)aline Apex Apomorphic Aquifer ATP (adenosine triphosphate) Aufwuchs Bell jar apparatus Bioassay Biodiversity
Biofilm
Depth zone of the sea, mostly between 4,000 and 6,000 m Living in or originating from different geographical areas; contr. sympatric A structure filled with liquid, often having a stiffening effect Active inhibition of a competing species through biological interactions Occurring on both sides of the Atlantic Ocean A coastal saltwater habitat separated from the open sea, mostly used in context with marine caves The tip of a (mollusc) shell Morphologically derived, not directly related to the ancestral structure An underground region that transmits groundwater, e.g., the aquifer of the Danube The prevailing energy store in living organisms Organisms (mostly plants) growing as a surface layer on hard substrates; a German word introduced into general ecology, syn. epigrowth Experimental gadget with bell-shaped jars enclosing an area of the bottom (mostly used for measurements of community respiration) A standardized experimental procedure mostly used in pollution tests Taxonomic or biodiversity (clearer: “species richness”); indicates number of taxa in comparisons of samples within a habitat (alpha-diversity), of habitats (beta-diversity), or of geographic areas (gamma-diversity); biodiversity should not be confused with ecological diversity (see diversity) Thin organic layer of microorganisms and excreted mucus
503
504
Bioirrigation Bioturbation Calcareous Cambrian Carboniferous Capitate Cavernicolous Circum-mundane Chemoautotrophy
Chemocline Cnidome Cretaceous Cryptic species Cryptobiosis Cuticularized Cystid
Desmoneme Destruents Devonian Diapause Diaphragm (of electrodes) Disjunct
Glossary
Water flux caused by the activity of organisms; adj. bioirrigative Mixing of sediment by biogenic activities, e.g., burrowing (Sediment) consisting of carbonates, mostly of biogenic origin Geological time period with intense radiation of animal phyla, about 540 million to 495 million years BP Geological time period, about 355 to 290 million years BP Ending with a small swelling, mostly used for tentacles Living in caves; syn. troglobitic Has a worldwide distribution A physiological pathway comparable to photoautotrophy, but uses the energy of a chemical compound (e.g., methane, hydrogen sulfide) instead of light for the assimilatory production of food substances Borderline between different chemical parameters, e.g., oxygen and hydrogen sulfide The complete set of differently shaped cnidocysts (nematocysts) Geological time period (140–65 million years BP) in the Mesozoic with a warm climate and, at around 93 million years BP, a long-lasting “oceanic anoxic event” A group of morphologically identical species whose differences are genetical or physiological and thus morphologically hidden, syn. sibling species A mode of passive endurance without signs of active life; organisms in this stage are resistant even to high radiation and a complete vacuum; adj. cryptobiotic Hardened by incorporation of hard, often brownish and resistant substances The body part of a bryozan individual (zooid) which is more or less solid, often also encrusted (syn. zooecium), and into which the soft tentacular apparatus, the lophophor or polypid, can be withdrawn Type of cnidocyst (nematocyst) with a long and thin thread coiled up in the cell The community of organisms that degrade organic matter, e.g., bacteria, fungi Geological time period, 415–355 million years BP A stage in the ontogeny of some animals in which the development halts, often occurs as an adaptation to unfavorable environmental conditions Connection of internal reference electrode with ambient medium, usually a porous ceramic disc With a geographically discontinuous distribution that precludes gene flow
Glossary
Dispersant Displacement Diversity
Dormancy Edaphic Ediacaran Endemic Ephemeral Epigean Epilithic Epimers Epiphytic
Estuary Euhaline Eulittoral zone Euryoecious Eurytopic Eutelic Eutrophication Exopolymer substances (EPS) or excretions Filiform
505
Chemical compound applied in oil spills to disperse and partly dissolve oil, comparable to detergents in private use In zoogeography and ecology, the removal of a taxon by another more successful one A. Taxonomic or biodiversity (clearer: “species richness”), indicates number of taxa in comparisons of samples within a habitat (alpha-diversity), of habitats (beta-diversity), or of geographic areas (gamma-diversity). B. Ecological diversity (various indices), a measure connecting the number of species with their abundance and distribution pattern (e.g., Shannon–Wiener index, Simpson index) A quiescence stage in some animals, often evolved as an adaptation to unfavorable environmental conditions; adj. dormant Living/occurring in soil Geological period in the late Precambrian, 600–540 million years BP, with mostly flat, two-layered metazoan fossils Occurring exclusively in the region investigated, not present in other regions; subst. endemism, endemist Has a very short, temporary existence Living on the surface of the Earth; contr. hypogean, subterranean Living attached to rock surfaces Lateral plate-like extensions of the body cuticle, mostly occuring in peracarid crustaceans Living as epigrowth (mostly filamentous algae) on other plants, often the blades of seagrasses or thalli of macroalgae; note that, in this sense, an “epiphytic meiofauna” does not exist, the correct term is “phytal meiofauna” River mouth opening into a tidal sea, comprises a wide range of brackish water conditions; adj. estuarine According to the Venice System, the brackish water classification system, an ecologically marine salinity range of between about 30 and 35 PSU Tidal zone Of a wide ecological range, ecologically fairly hard, contr. stenoecious Of a wide biotopical range; contr. stenotopic Having a fixed cell number, a cellular constancy reached at a juvenile stage Rich supply of a water body with nutrients, often resulting in an unwanted excess of organisms Complex mucoid substances released by organisms, consisting mainly of polysaccharides Thread-like
506
Fission Flotation Generalist
Gondwana Gonochoristic Haptic Hapto-sessile Hemisessile Heterogony Homonomous Hydrogenosome Hydrothermal vents Hyperbenthic Hypogean Idioadaptation Interstitial fauna Irrigation Iteroparous K-strategist
Karstic Lacustrine
Glossary
Mode of asexual reproduction in animals achieved by division of the body into two or several parts that become independent individuals Passive drifting of particles Ecological type of organism adapted to a wide array of habitat conditions; often also termed a “colonizer” due to its ability to rapidly populate disturbed habitats after organismic depletion. contr. specialist, persister Archaic southern supercontinent that existed before it was broken up by the Atlantic Ocean An unisexual animal; individuals with separate sexes; contr. bisexual, hermaphroditic Structures/animals that temporarily attach to a substrate Animals that attach to a substrate, but are capable of detaching their hold and slowly changing their position Slow-moving animals that temporarily fix their positions by anchoring to structures Regular alternation of generative modes, changing from a normal, heterosexually reproducing generation to one (or several) unisexual, parthenogenetic generations With a repeated sequence of similar structures, e.g., similarly sized segments in arthropods; subst. homonomy Cellular organelles (of some protozoans) that are metabolically involved in the production of hydrogen, probably derived from mitochondria Underwater, often deep-sea sites where hot water, usually rich in hydrogen sulfide, escapes under high pressure Living directly above the bottom Living under the Earth’s surface, often also “subterranean;” contr. epigean Adaptations of a high specificity for a particular organism Microscopically small fauna of aquatic sediments living in the interstices of sand The pumping of water via waving movements (mostly used by animals living in tubes or burrows) Reproducing several times during a lifetime, while semelparous animals have just one reproductive period Ecological type of organism highly adapted to specific biotopical conditions, fully utilizing the carrying capacity of a given ecological niche; often also coined a “specialist” or a “persister.” contr. r-strategist Pertaining to limestone strata rich in subterranean waters, caves and streams Pertaining to lakes
Glossary
Lagerstätten Laurasia Lebensformtype Limnogenous/ limnogenic Littoral Marsupium Meroplanktonic Mesobenthic Mesocosm Mesohaline Messinian Crisis Metameric Metasome Microtidal sea Mixocoelom Monotypic Necrophagous Neotenic Neozoa Nodulus Oligohaline
507
Paleontological site with numerous undisturbed layers of sedimented fossils; a German word introduced into general paleontology Archaic northern supercontinent that existed before being broken up by the Atlantic Ocean A specific combination of structural and ecological adaptations to a particular habitat, e.g., to life in narrow crevices; a German word introduced into general ecology Of freshwater origin In the marginal zone of the sea over the shelf; “eulittoral” refers to the tidal zone (here) The brood pouch of peracarid crustaceans made of characteristic cuticular plates, the oostegites Living temporarily, for a certain ontogenetic phase (mostly as larvae), as plankton; subst. meroplankton Benthos living between in the voids, interstitium of the sediment Experimental simulation of a more or less large and complex natural habitat and its organisms; on a smaller scale: microcosm According to the Venice System, the brackish water classification system, a salinity range of between 5 and 18 PSU A period in the late Tertiary, about 6 million years ago, with a hot, dry climate, when the Mediterranean Sea evaporated (possibly completely) Of segmental structure; subst. A “metamer” is a segment Posterior part of the body in chelicerates Sea basin with small tides, e.g., the western Baltic and Mediterranean The body cavity of an arthropod developed through the fusion of the primary blastocoel and the secondary coelom A taxon which alone represents the next highest taxonomic category, e.g., a genus with a monotypic species comprises just this single species Feeding on dying or dead animals Attaining sexual maturation at a stage that structurally corresponds to a larva, somatic development is retarded in comparison to sexual development; subst. neoteny; syn. progenetic Animal species that have only recently invaded a regional community, often spread by anthropogenic transport, invasive species; corr. neophyta A swelling in the setae of annelids According to the Venice System, the brackish water classification system, a salinity range of between 5 and 1 PSU
508
Oostegite Orthogenetic
Oxybiont Oxygen crisis
P/B value Paratomy Parthenogenesis Pelagos Pelos Peristaltic Permo/Triassic transition period Phenotypic sex determination Phoretic Physical lung
Phytal Phyton Pinger Plastron respiration
Glossary
Cuticular plates originating at the basis of the thoracic appendages in peracarid crustaceans and forming the brood pouch A “straight,” rectilinear evolution in one direction over a considerable period of time, commonly used with the implication that the direction is determined by an internal factor, not by natural selection; subst. orthogenesis Organism adapted to live under oxic conditions; adj. oxybiotic Periods of complete anoxia in the deep sea that lasted millions of years (e.g., the Permo/Triassic crisis; the Cretaceous crisis). The existence of these oceanic periods means that the benthos did not survive in the deep-sea since archaic times. Relative index of productivity relating Production (over a given time period) to the average standing stock, the Biomass Asexual division of body with complete subsequent regeneration of parts, a mode of asexual multiplication; see fission Development from unfertilized female gametes; adj. parthenogenetic Animal community living in the open ocean Aquatic animal community living in mud or silt A mode of motion with regular alternation of longitudinal and circular body muscles, producing progressive contractile waves Geological time period about 250 million years BP during which there was complete anoxia in wide parts of the deeper oceans Sex determination by external environmental factors and not by the chromosomal combination alone Being regularly transported by an animal of a different species; subst. phoresy Structure present in aquatic air-breathing animals used for the storage of an air bubble from which the oxygen is respired, and into which dissolved oxygen regularly diffuses from the ambient water; this type of respiration is also termed “plastron respiration” An aquatic habitat that is characterized and structured by plants (also used as an adjective, e.g., the phytal meiofauna community) Aquatic animal community living in plants, in the algal belt Device releasing an electronic signal upon contact, e.g., attached to grabs or nets signaling contact with the bottom See “Physical Lung”
Glossary
Plesiomorphic Plesiotope Pneumatophors Polyhaline Polypid Polyvoltine Precambrian Preadaptation Progenetic Protandrous Protogynous Proterozoic Psammon Pseudofertilization r-Strategist
Refractometer Regressive evolution Rhabdites
509
An archaic condition similar to the (assumed) ancestral one; contr. apomorphic The original, ancient biotope of an organism’s ancestors Aerial roots of mangroves specialized for gaseous exchange; they grow from the mud into the water/air According to the Venice System, the brackish water classification system, a salinity range of between about 18 and 30 PSU The soft part of a bryozoan zooecium, the lophophor (see cystid) Reproductive cycle with several reproductive periods per year; similarly univoltine (one reproductive period), bivoltine (two) Geological time period previous to the Cambrian, from about 4.5 billion years to 540 million years; including the Proterozoic as the first period with metazoan fossils Possession by an organism of traits that would favor its survival in a changed (new) environment; adj. preadaptive Attaining sexual maturity at a (morphologically) larval stage, somatic development retarded in comparison to sexual development; subst. progenesis; syn. neotenic, neoteny Condition in hermaphrodites where a functional male phase is antecedent to the subsequent female phase Condition in hermaphrodites where a functional female phase is antecedent to the subsequent male phase (rare) Geological time period in the Precambrian, about 2.5 billion years to 540 million years BP; the first animal fossils occurred in the Proterozoic, dating from about 1 billion years BP Aquatic animal community living in sand A stimulus elicited from sperm (or sexual products) that does not result in karyogamy but is required for the onset of parthenogenetic development (in some oligochaetes) Ecological type of organism adapted to a wide range of biotopical conditions; with its high reproductive potential it quickly colonizes impoverished habitats after destructive events; often also termed a “generalist” or “colonizer;” contr. K-strategist Instrument measuring the light-refractive effects of compounds dissolved in water Evolution leading to increasing simplification of structures or reduction of organs, notably frequent in caves or subterranean habitats Rod-shaped secretions in the turbellarian epidermis which dissolve rapidly when extruded
510
Rhinophore Semelparous Sessile Siliceous Skewness Specialist Stenoecious Stenothermal Stenotopic Stygobios Subterranean Sublittoral zone Sympagic Sympatric (Sym)plesiomorphy (Syn)apomorphy Syncytium Syntopic Terrigenous Tethys Sea
Thalassobionts Thalassogenous (thallasogenic) Thiobiont Triradiate Troglobitic Troglobiont
Glossary
Posterior pair of tentacles in opisthobranch molluscs with numerous chemoreceptors Reproducing only once during a lifetime; contr. iteroparous: with several reproductive periods Permanently attached/fixed to a substrate Sediment of silicate minerals, e.g., quartz The degree of asymmetry in a statistical distribution, e.g., in the sediment grain size distribution Species with a small adaptive range that fits well into one particular narrow ecological niche; often a K-selected species; contr. generalist Ecologically of a narrow range; contr. euryeocious With a narrow temperature range; contr. eurythermal With a narrow biotopical range; contr. eurytopic Community of animals in subterranean habitats, also “stygofauna;” adj. stygobiotic Living underground Subtidal zone Living in the crevices of ice, mostly sea water ice Co-occuring in the same area; subst. sympatry The common possession of an ancestral, homologous character; contr. A derived character specific to the taxon and not shared by the ancestral taxon; adj. plesiomorph, apomorph Cell mass or layer with numerous nuclei, but without separating cell walls; secondarily derived in eukaryonts Occurring in the same biotope Consisting of material originating from the land (also terrigenic); contr. thalassogenous (thallasogenic): originating from the sea The mesozoic circumglobal ocean that separated Pangaea into the northern Laurasia and the southern Gondwanda supercontinents; the Tethys Sea was largest during the Jurassic and Cretaceous Organisms living in the sea, from (Greek) thalassa: sea Originating from the sea, of marine origin Organism adapted to live under sulfidic conditions; adj. thiobiotic Pertaining to an organ with a regular tripartite and trisymmetrical structure, i.e., the pharynx musculature in Nemathelminthes Living in caves and corresponding subterranean spaces, also “troglobiotic” Animal living in caves; as a community: “troglofauna”
Glossary
Turgescence Ubiquitous Undulatory Urosome Vascularization Verrucose Vicariant
Zooid
511
Stiffening by internal pressure (turgor), present in cells, organs, bodies; adj. turgescent Of worldwide occurrence, subst. ubiquity Wave-like (movements) The posterior part of the abdomen in crustaceans Filling a structure (cell, organ) with liquid; adj. vascular Warty, with small tubercles Geographical distribution of a taxon in separate areas without bridging connections; for instance, a meiofauna species that has amphi-Atlantic littoral populations; subst. vicariance An individual within a colonial aggregate, usually of minute size, e.g., in bryozoans, colonial ascidians
Index
2D-planar optodes, 19
A ABC method, 355 abiotic factors of hyporheic habitats, 332 abiotic structuring factors, 375 abrasion, 278 abundance, 388 Acari, 201 acidity, 25 acidosis, 26 Acochlidioidea, 226 Acoela, 120 acoelomates, 236 Actinopodia, 107 Adenophorea, 140 adhesion, 92 adhesive fibres, 177 glands, 130, 236 mucus, 92 adsorption to biofilms, 370 advective transport, 17 Aeolidiacea, 226 Aeolosomatida, 209 aggregations, 259 agitation, 14, 15 Alaskan ice worm, 23 algal belts, 317 fronds, 320 holdfasts, 320 structural complexity, 319 Allogromiida, allogromiids, 104, 106 allopatric speciation, 244, 343 alpha diversity, 267 Ameiridae, 184 amensalism, 56, 374 American meiobenthology, 4
ammonia, 106 Amoco Cadiz oil spill, 363 Amoebozoa, 107 amphiatlantic, 248 amphiatlantic distribution, 256 amphids, 141 amphi-oceanic distribution, 247 Amphioxus sand, 232 Amphipoda, 197 amphipods and isopods, 341 amphi-tethyan, 342 anaerobic life, 105 anaerobic metabolism, 31, 304 anaesthetization, 78 Anaspidacea, 191 anchihaline, 244 anchihaline caves, 164, 234, 343 anchoring devices, 92 Ancorabolidae, 184 anesthetization, 74 angularity, 7 anhydrobiosis, 168 Annelida, 208, 237 Annelida incertae sedis, 209 annelid-arthropod line, 170 annelids, 277 Anomopoda, 173 Anopla, 136 anorganic nutrients, 36 anoxia, 30 anoxia indicator, 302 anoxic, and sulfidic environments, 295 Anthozoa, 118 anthropogenic transport, 258 Anthuridea, 197 Anurida, 207 apex, 229 Aphanoneura, 220 Aplacophora, 225 apomorphic, 236 513
514 aquifers, 244 archaic thiobios, 312 Archiannelids, 237 Arctic meiobenthos, 268 Arenadiplosoma, 232 Arthrotardigrada, 166 Ascidiacea, 231 ash-free dry weight, 39, 391 Aspidosiphon, 223 Astomonema, 145, 300 ATP content, 40, 294 attachment capability, 318 available\detritus, 39 axopodia, 107
B Bacteria, 43 bacteria production, 48 symbiosis, 152, 217, 302 bacterial abundance and biomass, 43 consumption, 403 epigrowth, 9 number or biomass, 45 symbionts, 226 symbiosis, 144, 307 bacterivory, 44 ballast sand, 255 water, 255 Baltic Sea, 324 bathymetric gradient of biodiversity, 289 Bathynellacea, 191 Batillipes, 93 Batillipes mirus, 166 Batillipes pennaki, 168 Bdelloidea, 130 beach enhancement, 58 stratifiction pattern, 20 bell jar, 294 apparatus, 395 benthic flagellates, 52 benthos in permanent anoxia, 303 beveled perspex tube, 67 bioassays on reproduction, 366 bioavailability, 353 biocoenosis, 3 biodiversity, 267, 417 biofilm(s), 7, 9, 38, 42 biogenic mobilization, 19
Index sand, 277 structures, 37 bioirrigators, 16 biological interactions between size groups, 379 oxygen demand, 291 reworking, 18 biomass, 384 biomass and fixation, 390 bioremediation, 363 biostabilization, 43 biostabilizers, 16 biotic factors, 37, 62 habitat factors, 37 structuring factors, 374 biotopical differentiation of phytal, 317 bioturbation, 17, 18, 30, 52, 57, 240 birds preying on meiofauna, 409 biramous mandibles, 181 bivalved carapace, 173, 175 bivoltine, 168 Black Sea, 303, 324 body elongation, 89 size spectra, 378 spines, 162 Bogidiella, 341 Bogidiellidae, 198 Bogorov sorting tray, 80 bottleneck effect, 409 bottom landers, 284 water straining, 67 Bou–Rouch pump, 337 Brachiopoda, 227 brackish coastal groundwater, 244 water, 125 water sites, 324 water species minimum, 325 water zone, 23 branchial filter system, 231 Bray-Curtis dissimilarity, 359 bridging taxa, 103 brine basins, 303 seeps, 129 brominated compounds, 56 brood protection, 100 Bryozoa, 227 bubble-and-blot method, 158 buccal cavity, 143 buffered formalin, 78
Index Burgess shale, 239 burrowers, 8 burrowing, 95 burrows, 57
C Caenorhabditis, 141 Caenorhabditis elegans, 87 caging effect, 408 calcareous sands, 13, 233 calculation of diversity, 63 of meiofaunal biomass, 82 of production, 384 of respiration, 397 caloric values, 391 Cambrian explosion, 239 canonical correspondence analyses, 361 Canthocamptidae, 184 Canuella perplexa, 387 carbon content, 391 Cartesian diver, 395 cast techniques, 73 cavernicolous meiofauna, 344 centers of bacterial growth, 44 Centropyxis, 108 Cephalaspidea, 226 Cephalocarida, 172 Cephalothrix, 136 cerata, 226 Chaetodermomorpha, 226 Chaetognatha, 231 Chaetonotida, 162 Chaetonotus, 162 chemoautotrophic ecosystems, 313 chemocline, 261 Chiridotidae, 230 chironomid larvae, 335 chloroplastic pigments, 285, 300 Chordata, 231 chordoid tissues, 136 Ciliata, 108, 299 ciliates, bacterivorous, 113 ciliation, 95 Ciliophora, 108 Cirolanidae, 197 cladistic analysis, 236 Cladocera cladocerans, 173, 335 Cladocopa, 177 claws, 168 cleptocnides, 226 clepto-cnidocysts, 122 Cletodidae, 184
515 climatic change, 421 climbing, 96 Cnidaria, 114 cnidocyst(s), 226 coastal brackish groundwater, 191 groundwater, 324 lagoons, 324 Coelogynoporidae, 123 coelom, 136 coelomates, 236 cohesiveness, 52 Collembola, 207 colloids, 38, 43 colonization, 253 colonization in lakes, 342 colonizers, 357 community changes by metal pollution, 368 respiration, 294, 395 stress, 362 structure, 374 competitive displacement, 375 situation, 382 compilation of benthic life history data, 384 compound chaetae, 216 computer-based methods, 420 computerization, 417 conductivity electrodes, 25 measurements, 25 continental distribution, 341 groundwater system, 244, 338 convergence, 89, 101, 235 conversion factors, 390 formulae, 390 Convoluta, 122 copepod identification, 182 Copepoda, 181 Copidognathus, 203 copper compounds, 367 copulation, 100 counterstaining, 73 Craib corer, 70 Crangonyctidae, 199 crawling, 95 critical grain size, 8, 15 crude oil, 363 Crustacea, 171 cryolander, 50
516 cryptic species, 246 cryptobiosis, 101, 168 ctenidia, 225 Ctenodrilus, 214 culms, thalli, 54 culture animals, 214 current velocity, 14 currents, 15 cyanobacteria, 47 cyclomorphosis, 132 cyclomorphotic, 168 Cyclopina, 190 Cyclopina gracilis, 273 Cyclopininae, 189 cyclopoid copepods, 274 Cyclopoida, 188, 189, 190, 272 Cyclorhagida, 156 Cylindropsyllidae, 184 Cypridacea, 176 Cyprideis, 176 Cyprideis torosa, 302, 178 cysts, 168 cystid, 228 cytherissa, 176, 178, 345 cytochrome C oxidase, 33
D Dalyelloida, 122 DAPI staining, 45 Darwinuloidea, 176 dead/live differentiation, 73 decantation, 74 decaying macrofauna, 259 decompositional biotope, 282 decrease of oxygen, 332 of pH, 332 decreasing size, 287 deep-sea, 233, 234 brine seeps, 298 hydrothermal vents, 297 meiobenthos, 284, 287 meiofauna, biomass, production, 293 meiofauna, density, 291 meiofauna, distribution pattern, 291 meiofauna, diversity, 289 oxygen crisis, 245 denaturing gradient gel electrophoresis, 402 Derocheilocaris, 180 detergents, 77 detrital trophic complex, 402 detritus, 37, 38, 402 detritus/bacteria complex, 44
Index detritus/bacteria-based food chain, 357 diapause, 101 diatom feeders, 218 diatoms, 49, 271 diffusive transport, 17 dimethyl sulfoxide (DMSO), 79 dinoflagellates, 49 Dinophilidae, 212 Dinophilus, 92 Diosaccidae, 184 direct fertilization, 199 Dirivultidae, 190, 302, 315 dispersants, 365 dissipative beaches, 21 shore, 412 dissolved free amino acids, 40 organic matter, 40, 307, 353 distribution patterns, 243 distributional pathways, 243 disturbance, 55, 376 Diurodrilidae, 212 Diurodrilus, 92, 212 diversity, 58 diversity indices, 376 DNA barcoding, 154 extraction, 238, 402 DOM, 40 DOM in the sediment, 40 dominance indices, 354 shifts, 364 dormancy, 101 Dorvilleidae, 212 downward migration, 56 Draconematidae, 93 dredging, 67 Drescheriella glacialis, 187, 273, 387 drift, 334 drinking water, 330, 345 dual gland system, 92 duo-gland pattern, 133 Duoshantuo, 239 dwarf forms, 197 dwarfism, 87 dysoxic, 297
E East Pacific Rise, 314, 317 Ecdysozoa, 137, 164, 170, 172 Echiniscoidea, 166
Index Echiniscus, 166 Echinodermata, 230 ecological disequilibrium, 374 stability, 344 ecosystem models, 416 ecotoxicogenomics, 354, 417 Ectoprocta, 227 Ectinosomatidae, 184 Ediacaran, 240 Eh measurements, 27 Ekman grab, 66 electronic species registers, 6 elongation, 87 elutr(i)ation, 75 Embletoniidae, 226 emergence, 124, 250, 251, 252, 126 diurnal, 251 Enchytraeidae, 217 Enchytraeus, 217 encystment, 101 endemism, 270 endopsammon, 4 endosymbionts, 122 energetic budget, 399 divergence between meiofauna and macrofauna, 398 energy budget, 384 fluxes, 14, 16–18 webs, 401 energy-flux diagrams, 388 Enopla, 136 enrichment of nutrients, 36 Entoprocta, 229 ephippium, 174 epibenthic, 7 epigean fauna, 333 epigrowth, 321 epigrowth feeders, 144 epimeres, 196 epipsammon, 4 Epsilonematidae, 93, 300 erodibility, 18 erosion, 14, 15, 250, 334 erosive shore, 411 estuaries, 277, 326 ETS, 391 euryecious nature, 205 euryhalinity, 197 Eutardigrada, 167 eutelic, 168 eutelic development, 137
517 eutrophic lakes, 346 evolution, 235 exopolymer secretions (EPS), 42 exopolymers, 42, 43 exposure, 14 extraction, 73 extreme biotopes, 270 pH, 26
F facilitative effects, 57 factorial correspondence analyses, 360 fatty acids, 40 signature, 401 faunistic composition of the freshwater meiofauna, 330 fecal pellets, 15 Fenvalerate, 370 fetch-energy index, 14 field-generated life tables, 387 filamentous algae, 322 filiform, 123 body shape, 89 first discoveries, 2 fission, 122 fixation, 77 fixation with fluorescent dyes, 72 fixatives for molecular analyses, 79 Flabellifera, 197 flattening, 92 flatworms, 120 flexibility, 87, 89 floating natural detritus, 254 flocculent layer, 66, 70 matter, 13 flood, 334 flotation methods, 75 flowmeters, 17 flushing effects, 57 food spectrum, 307 supply, 260 Foraminifera, foraminiferans, 103, 271, 288, 299 Foraminifera/nematode ratio, 302 formalin, 78 fossil record, 239 fossils, 175 fractal scaling, 320 fractionated sieving procedure, 10 fractioning device, 69
518 free-living platyhelminths, 237 freshwater, 173 biotopes, 329 environmental science, 359 meiobenthology, 4 meiofauna, 329 shock, 174 functional diversity, 361 future toes, 162 techniques, 419 topics, 418
G Gamasida, 207 gardening, 404 gardening hypothesis, 144 Gastropoda, 162 Gastrotricha, 162, 299 general foraging theory, 407 gene sequencing, 235 generation time, 392 genome, 141 geochemical fluxes, 61 geothermal reduction, 298 germarium, 122 glacier worm, 273 glandular adhesion, 93 gliding motion, 164 global biodiversity assessment, 267 climate change, 5 glutaraldehyde, 78 Glyceel embedding, 80 Gnathifera, 126 gnathosoma, 203 Gnathostomula paradoxa, 127 Gnathostomulida, 126, 299 Gondwana, 199 grabs, 66 gradients between oxic (oxidized) and sulfidic layers, 307 grain shape, 7 size, 7 size analysis, 9 size composition, 11 structure, 8 surfaces, 47 Grania, 217 granulometric parameters, 11 granulometry, 10 Granuloreticulosa, 103
Index Graphic Mean, 12 graphical/distributional methods, 362 grasping, 118 grazing rates, 404 Gromiida, 104 gromiids, 106 groundwater, 181, 191, 197, 214 Amphipoda, 244 aquifer, 330 Isopoda, 244 gutless nematodes, 145, 300 oligochaetes, 220 Gwynia capsula, 227 Gymnamoebea, 107 Gymnolaemata, 228
H habitat enhancement, 377 heterogeneity, 286 Halacaridae, 201 Halacaroidea, 201 Halammohydra, 115 Halicyclophinae, 189 Halicyclops, 190 haptic, 318 haptic structures, 92 haptic tubes, 162 haptosessile, 1 Harpacticoida, 171, 181 harpacticoid families, 182 heterogonous, 130 heterogonous, generation cycle, 174 harpacticoids, 278 as prey, 188 sensitive to metal pollution, 369 headwater habitats, 330 heavy metals, 18 Heliozoa, 107 heme proteins, 305 hermaphrodites, 212, 213 hermaphroditic, 233 hermaphroditism, 100, 195 Hesionidae, 213 heterogonous, 126 heterogonous, generation cycle, 163 heterokaryotic, 104, 108 Heterotardigrada, 166 Higgins larva, 161 high diversity, 280 endemism, 316
Index silt flotation, 77 structural complexity, 279 history of meiobenthology, 2 Holothuroidea, 230 Homalorhagida, 157 homology, 235, 236 homonomous, 236 Hoplonemertinea, 135, 136 horizontal distributional patterns, 258 hot deep-sea vents, 313 spot of biodiversity, 289 spots of meiofauna diversity, 290 Huston model, 40 Hutchinsoniella, 172, 173 Hydracarida, 207 Hydrachnellae, 205 Hydrachnidia, 205 hydrodynamic pattern, 21 hydrodynamics in running waters, 333 hydrogen sulfide, 27, 33, 295 sulfide - development, 297 sulphide and anoxia - synergism, 305 Hydroida, 116 hydrogenosomes, 299, 113 hydromedusa, 116 hydropsammal, 345 hydrothermal vents, 190, 302 hygropsammal, 346 hyperbenthic, 125 hyperbenthic, life style, 187 Hydroida, 114 hyporheic, 4, 205, 330 refuge hypothesis, 334 research - methodological problems, 337 hyporheic zone as an ecotone, 332 Hypsibius, 167
I ice age, 344 channels, 273 floes, 254, 270 margin, 268 worm, 217 idiosoma, 186 identification guides, 6 idioadaptations, 225 immunological methods, 247 impact of (heavy) metals, 366 Inanidrilus, 302 indicator species, 353
519 infraciliation, 109 Ingolfiellidea, 199 initial losses, 366 insect larvae, 330 interactions between bacteria and meiofauna, 404 interference contrast microscopy, 139 internal crystals, 301 International Association of Meiobenthologists, 5 interpretation of diversity indices, 375 interrelations troglobitic/abyssal, 245 interstitial coenoses, 264 fauna, 3 intestinal granules, 301 introvert, 159, 223 Introverta, 137 invasive capacity, 258 inverse relation weight and production, 399 iron sulfides, 33 irrigational fluxes, 18 Irwin loops, 81 island theory of colonization, 377 isocommunities, 322 isotopic markers, 419 signature, 401 iterative contraction–elongation, 96
J jaw apparatus, 127, 133 jaws, 128 Jennaria pulchra, 222
K K-strategists, 338, 355 Kajak or Haps corer, 70 Kalyptorhynchia, 118 Kamptozoa, 229 kaolin powder, 77 Karaman–Chappuis method, 67 karstic, 197 fauna, 330 regions, 342 Karyorelictida, 110 k-dominance method, 355 kenozoids, 229 Kentrophorus, 113, 299 keystone taxa, 421 kinorhynch extraction, 158 kinorhynch movement, 157
520 Kinorhyncha, 156 Komokiacea, 106 K-selection, 355 K strategists, 244
L lab-cultured meiobenthic species, 354 Labidoplax, 230, 236 lacustrine meiobenthos, 5, 345 lagoons, 328 Lake Brunnsee, 346 Lake Mirror (USA), 346 lakes, 345 Laophontidae, 184 large-scale field experiments, 353 zonation, 259 larval structures, 237 lateral slits, 135 latitudinal diversity gradient, 291 Laurasian continent, 199 lebensformtype, 3 length/width ratios, 308 length-to-width ratio, 89 life history, 386 studies, 387 life-table parameters, 386 limnetic benthic size spectra, 348 Limnognathia maerski, 133 Limnohalacaridae, 201 littoral mesopsammal, 244 living fossils, 173 Labotocerebrida, 220, 221 Lobatocerebrum, 209, 220, 222 locomotive organs, 93 long distance transport, 258 long-term parameters, 353 looping, 96, 142 Lophotrochozoa, 208, 223 lorica, 159, 160 Loricifera, 160 loss of weight, 10 low permeability, 278 Loxodes, 305 Ludox, 76 Lumbricillus, 217 Lumbricillus lineatus, 246 Lumbriculidae, 218 Lunulitiform bryozoans, 229
M macroalgae, 54, 319
Index macroalgal grazers, 146 Macrobiotus, 167 Macrodasyida, 162 Macrostomida, 122 mainstream research, 421 maintenance dredging, 58 Malacostraca, 190 manganese nodules, 286, 288 mangroves, 280, 321 manmade erosive forces, 377 many brackish water species, 325 Marenda, 104 marine snow, 42, 255 Marionina, 92 marsupium, 193–195 mass culture, 187 extraction methods, 74 massive anthropogenic impact, 327 mastax, 130 maturity index, 153 maturity indices, 357 measurement of total organic matter, 39 mechanical adhesion, 93 disturbance, 55 mechanisms of dispersal, 249 median, 11 medusa, 115 meiobenthic communities, 264 meiobenthos, in the benthic energy flow, 384 definition, 1 derivation, 1 meiofauna, 1 biomass boundaries, 1 and pollution, 350 as an integrative benthic complex, 410 as prey, 323 dead-end hypothesis, 406 in hard substrate communities, 323 in polluted areas, 350 integration hypothesis, 406 interstitial, 1 paradox, 247 videofilms, 82 fossils, 239 grazing, 404 size boundaries, 1 meiophyton, 319 Meiopriapulus, 159 Meiopriapulus fijiensis, 158 meroplanktonic larvae, 380 Mesenchytraeus solifugus, 217, 273
Index mesobenthic, 7 mesodermal coelothel, 137 mesohaline region, 326 mesopsammic fauna, 87 mesopsammon, 4 mesothelium, 137 metabolic rates, 384 metabolic relevance of meiobenthos, 398 metal(s), 350 compounds, 366 pollution and insect larvae, 369 toxicity scale, 367 metapopulation, 246 methane, 298 methanogenic bacteria, 299 methanogens, 299 methodological problems, 344 methodology, 63 methods assessing population parameters, 384 microalgae, 49 microalgae-based food chain, 357 harpacticoids, 50 Microarthridion littorale, 387 microbial sulfate reduction, 298 micro-bivalves, 223 Microcerberoidea, 197 microchambers, 298 microcrustaceans, 335 microelectrodes, 30 microelectrometric measurement, 307 microgastropods, 223 Micrognathozoa, 126, 133 microgradients of oxygen, 30 Microhedyle, 236 microniches, 47 micro-oxic zones, 105 Microparasellidae, 196 microphytobenthos, 49 production, 405 microrespirometer, 395 microscopic observation chambers, 72 studies, 420 microtopography of the bottom, 39 microwave fixation, 79 Mictacea, 194 Mid-Atlantic Ridge, 314 migration around the chemocline, 308 migrations, vertical, 261 miniaturization, 87 minimal sample area, 64 Miraciidae, 184
521 mitochondria, 305 molecular biology, 237 clocks, 239 genetics, 417, 419 genetics in pollution studies, 371 methods, 5, 247 Mollusca, 223 Monhystera disjuncta, 392 Monobryozoon, 228 monociliated, 236 Monoplacophora, 225 monsoonal floods, 277 monsoons, 280 morpho groups, 105 morphological adaptations, 87 mosaic patch theory, 374 mountain streams, 336 mouth cone, 160 pipette 81 tentacles, 231 mouthless, 308 Movile Cave, 315 mucus, 15, 37, 42 muddy bathyal and abyssal plains, 284 mud-volcano, 314 multidimensional scaling, 359 multigeneration tests, 353 multiple Aberdeen corer, 284 corers, 66 SMBA corer, 70 multivariate methods, 362 statistics, 153, 359 mutual exclusion, 56 Myodocopa, 177 myoepithelial pharynx, 236 Myriotrochidae, 230 Mystacocarida, 180
N N/C index, 357 N/C ratio, 365 Naididae, 101 Nanaloricus mysticus, 160 natural oil seeps, 363 nauplius, 182 Nematalycidae, 89 Nematalycus nematoides, 201 Nemathelminthes, 136, 137 Nematoda, 299
522 nematode buccal cavity, 139 feeding groups, 348 groups, Traunspurger, 146 groups, Wieser, 144 identification Keys, 188 nematode/copepod ratio, 357 nematode-copepod index, 153 nematodes, biodiversity, 149 disturbance indicators, 153 dominate, 288 dominating, 268 diversity, 149 functional groups, 148 genetic diversity, 154 life history, 142 motion, 141 physiology, 154 tail shape, 148 width/length ratio, 147 Nematoplanidae, 123 nemerteans, 136 Nemertinea, 134 Neorhabdocoela, 122 neotenic, 237 neoteny, 101, 232 neozoa, 254 nephridia, 237 Nerillidae, 212 Nerillidium, 209 Niphargidae, 199 Nitokra, 187 non-selective deposit feeders, 143 novel microscopical methods, 249 nuclear apparatus, 109 Nudibranchia, 226 nutrient input by aquaculture, 37 nutritional selectivity, 62
O oil pollution and meiofauna, 362 seeps, 315 spills, 363 oil-degrading bacteria, 366 Olavius, 302 planus, 92 Oligochaeta, oligochaetes, 302, 328 oligotrophic lakes, 346 Oniscoidea, 197 Ophiuroida, juvenile, 230 Ophryotrocha, 214
Index Opisthobranchia, 223, 226 optical oxygen electrodes, 31 sensor, 17 optode(s), 17, 31, 395 optometric image analysis, 82 organic content, 259 film, 322 matter, 38, 39, 333 organismic space and time, 381 Orsten fauna, 239 orthogenetic, 235 Ostracoda, 175, 239 Otoplanidae, 95, 122, 123 Ototyphlonemertes, 136 oxic/anoxic chemocline, 308 microniches, 30 oxygen, 27, 28, 297 diffusion rate, 33 gradients, 297 microelectrodes, 31 minimum zone, 297, 310 minimum zones, 284
P P/B ratio, 384 packaging model, 407 paleontological findings, 235 Palpigradi, 206 Pancarida, 193 Parahaploposthia, 304 Paramesochridae, 184 Paramonhystera wieseri, 304 Paraonidae, 213 Parastenhelia megarostrum, 387 Parastenocaridae, 181 Parastenocaris, 346 Parergodrilidae, 213 parthenogenesis, 101, 137, 174, 177, 212 particle surface, 7 particulate organic matter, 38, 39 patchy distribution, 62, 259 pathways of refuge, 244 pelos, 3 Peltidiidae, 184 penetration of light, 49 Pentactula larva, 231 percolation, 17 Percoll, 77 Percoll flotation, 45, 110 periphyton, 164, 347
Index permanent anoxia, 303 meiofauna, 57 permeability, 7, 16 permeameter, 16 persisters, 357 pesticide(s), 350, 370 bioaccumulation, 371 toxicity, 370 petroleum hydrocarbons, 350, 362 pharyngeal bulb, 209, 237 pH electrode, 26 measurement, 26 value, 25 Phascolion, 223 phi intervals, 10 phi-conversion, 13 phi-notation, j-notation, 11 phreatic, 190, 330 Phreaticoidea, 197 Phyllognathopus, 346 phyllopodia, 173 phylogenetic aspects, 235 physical lungs, 208 physiographic regime, 259 physiological approaches, 420 phytal, 317 phytal habitats, 317 phytodetritus, 38, 285, 291, 293, 403 phyton, 3, 317 pictorial identification keys, 83 piercing plant cells, 322 Pilidium, 136 Pisionidae, 213 piston corer, 84 plankton sledge, 252 planktonic larva hypothesis, 381 phase, 237 plant roots, 54 plastron respiration, 208 plate tectonics, 256 Platycopida, 177 Platyhelminthes, 299 platyhelminths, free-living, 119 plesiomorphic, 235, 237 biotope, 312 plesiotope, 236, 343 Pliciloricus, 160 Ploimida, 130 plutonium pollution, 369 pneumatophores, 54, 280, 321 Podocopa, 176, 177
523 polar regions, 267 polar sea ice diatoms, 49 polarographic oxygen electrodes, 31 pole corer, 70 pollutants, 36 pollution and freshwater meiofauna, 351 pollution indicators, 188 studies with meiofauna vs. macrofauna, 351 Polychaeta, 208, 302 polychaetes, limnic, 212 troglobitic, 212 Polygordiidae, 211 Polygordius, 89 polyp, 114 polyphyletic, 237 polypid, 228 polyps, 116 polysulfur, 301, 306 polyunsaturated fatty acids, 41 fatty acids (PUFAs, 51 Pontonema vulgare, 300, 310 pore volume, 7 water, 83 water drainage, 16 water flow, 16, 17 water lance, 84 porosity, 16, 17 Posidonia, 321 Potamodrilida, 220 Practical Salinity Units (PSU), 24 predation, on meiofauna by small fish, 407 selective, 125 on meiofauna, 56 predators, 125 predators/ omnivores, 144 preservation, 77 preserving meiofauna, 72 Priapulida, 137, 158 priapulid larva, 160 PRIMER software, 360 primitive metazoan life, 311 Principal Response Calculation, 361 Proarthropoda, 170 proboscis, 135 proboscis, eversible, 122 processing of meiofaunal samples, 72
524 production calculated from respiration, 397 in lakes, 347 profundal, 347 progenesis, 101, 115, 161, 174, 178, 181, 191, 192, 195, 231, 246, 338 progenetic, 213, 237 prop roots, 282 Proseriata, 123 Prosobranchia, 226 protecting structures, 97 Proterozoic, 240, 311 Protista (Protoctista), 103 Protodrilidae, 211 Protodrilus, 92 protruding tubes and plant culms, 15 Psammodrilidae, 213 psammon, 4 Psammonalia, 5 Psammostyela, 96 pseudocoelom, 137 pseudofertilized, 246 pseudocoelomates, 246 pseudofertilized, 246 pseudosegmental protonephridia, 135 Pseudovermidae, 226 PSU, 24 Pycnogonida, 207
Q qualitative extraction, 73 sampling, 67 sampling problems, 67, 335 quartile, 11
R r- and K-strategists, 355,356 radioactive isotope, 402 radula, 225 rafting, 254 raphe, 225 reactions after oil spills, 366 recolonization, 58, 334, 365, 377 recovery period, 365 redox electrode, 27 potential, 27 potential and oxygen, 28 potential discontinuity layer, 28 values, 298
Index reduced competition, 344 egg number, 190 habitat, 297 vagility, 236 reducing layers, 281 reduction of eyes, 99 re-entry, active, 253 reflective beaches, 21 refuge area, 244 regressive evolution, 87, 235, 244 Rehbachiella, 239 reinforcing structures, 97 relict group, 192 relict groups, 190 relict refuge model, 256, 342 Remanella, 109, 111 remotely operated corer, 70 reproductive and larval stages, 353 reproductive dysfunctions, 370 resettlement, 252 resin embedding, 72 technique, 79 resource partitioning, 62 respiration, 388 equation, 394 rate, mass-specific, 394 rates, 396 resting eggs, 347 stages, 249 reticulopodia, 105 retrospect on meiofaunal research, 432 reworking of the sediment, 55 rhabdites, 125 Rhabditida, 141 Rhabditis marina, 141 Rhizaria: Granuloreticulosa, 103 Rhizopoda, 105, 288 rhizopodia, 105 Rhodopidae, 226 Rhombognathus, 203 rhynchocoel, 135 ribbon worms, 134 ripple marks, 17 river aquifers, 344 pollution, 332 Rose Bengal, 73 Rotatoria, 130 Rotifera, 126 rotifers, 330, 335, 346
Index RPD layer, 28 r-strategists, 355 S Sabatieria pulchra, 300 Saccocirridae, 212 salinity, 23 gradient, 24 refractometer, 25 tolerance, 326 sample size, 63 splitter, 80 staining, 73 sampling devices, 64 strategy, 63 sand grains, 47 saprobiotic meiofauna, 39 Scalidophora, 137, 156, 161 scarcity of food, 287 Scyphozoa, 118 sea grasses, 54 sea ice, 49, 111, 125 meiofauna, 270 mount fauna, 290 mounts, 234, 257 seagrass beds, 320, 323 seagrass meadows, 318 seasonal pattern, 277 variations, 263 seasonality, 284, 320 seawater ice method, 75 Secernentea, 140, 141 secretion of exopolymers, 58 sediment–water regime, 14 sediment cohesion, 16 pores, 7 sinks for pollutants, 36 sediment-water regime, 14 seeps, 313 selective deposit feeders, 143 grazing, 185 semiconductor probes, 23 separate spatial and temporal scales, 383 Seriata, 122, 123 serological antigen–antibody reaction, 402 sex determination, 195 Shannon diversity index, 354 shear strength, 15, 16 shelf connections, 257
525 shell hash, 227 shells, 173 shock freezing, 72 wave, 66 short-distance transport, 258 shortening of the core, 66 siliceous sands, 13 silt-clay fraction, 10 similarity indices, 354 Siphonolaimus, 300 Sipuncula, 223 size reduction, 325 spectra, 287 skeletal ossicles, 231 skewness, 12 sliders, 8 small-scale distribution, 259 small fishes and shrimps as meiofauna predators, 408 Smith–McIntyre grab, 66 small food web, 402 small-scale distribution, 66 Solenofilomorphidae, 124 Solenogastres, 225 sonication, 110 sonification, 45 Sorberacea, 233 Spadella, 231 Spartina, 321 salt marsh, 321 specialized K-strategists, 287 speciation in meiobenthos, 256 spectrophotometrical sulfide measurement, 34 spermatophores, 99 sperm injection, 132 sperm transfer, 99 spermatophore, 203 spicules, 225 spinneret glands, 178 square stainless steel tube, 69 squeeze sampling, 85 stable environments, 374 standardized bioassays, 353 standing stock, 384 standing waters, 345 standpipe corer, 337 static organs, 97 statocyst, 116, 122, 231 steep vertical gradient, 261 stepping stones, 257 sticky eggs, 100
526 sticky tunic, 232 Stilbonematinae, 144, 300 stimulated bacterial activity, 48 straining ground water, 67 stratified brackish water bodies, 326 stream beds, 330 strong variations, 270 structural complexity, 54, 317 stygobial P/B ratio, 393 stygobionts, 339 stygobios, 4, 217 stygobiotic, 136, 195, 199, 334 stygobitic, 338 crustacean meiofauna, 341 Stygocapitella, 213 stygofauna, 244 sublethal effects, 363 submarine canyons, 286 suboxic, 297 subterranean fauna, 329 subtropical regions, 276 sucker plates, 168 suction corer, 67 suction sampling of pore water, 83 sulfide biome, 122, 296 biome oldest biosystem, 311 layer, 152 microelectrodes, 35 nematodes, 281 toxicity, 307 sulfidic microniches, 34 sands, 129 sulfur bacteria, 44 sulphide fauna, 307 sulphur bacteria, 307 summer eggs, 174 suspended polymers, 43 suspension, 250 Syllidae, 213 symbiosis, zoochlorellae, 122 with bacteria, 306 symbiotic bacteria, 301 relationships, 99 sympagic, 123 sympagic meiofauna, 23, 50, 270 sympatric speciation, 246 symplesiomorphic, 238 synapomorphic, 236 Synaptidae, 230 Syncarida, 341
Index syncytial integument, 130 synecological, 37 perspectives, 374 synergistic effects, 353 syringe corer, 83
T tail formation, 89 Tanaidacea, 195 tannin content, 281 tanypodid midges, 412 Tardigrada, 166 taxonomic distinctness index, 359, 376 diversity, 278 minimalism, 356 Tegastidae, 185 temperature, 23 temporary meiobenthos, 252 meiofauna, 57, 209, 336 tentacular crown, 223 Tentaculata, 227 ternary diagrams, 11 terrestrial, 213 terrigenous Arthropoda, 207 Testacea, 107 Tethys Sea, 197 texture, 15 thalassobionts, 207 thalli/blades, 318 theristus anoxybioticus, 304 thermistor, 23 thermistors, 17 Thermosbaenacea, 193, 341 thiobios, 148, 220, 296 definition, 297 ecophysiology, 302 hypothesis, 240 thiobiota, 298 thiobiotic ciliates, 299 meiobenthos, 298 turbellarians, 299 Thioploca, 303 thiosulfate, 298 thiozoic, 240 thiozoon, 311 three quarters rule, 397 tidal changes in inundation, salinity and temperature, 280 tidal currents, 258, 261 tidal estuaries, 324
Index tidal pump, 16 time-stability hypothesis, 374 Tisbe, 246 Tisbidae, 185 Tobrilus, 300 gracilis, 304 toes, 130 tolerance capacity, 327 torpidity, 178 toxicants in sediments, 353 toxicity of hydrogen sulphide, 304 transepidermal uptake, 41, 92 triad in pollution research, 352 Tricladida, 123 triennial conferences, 5 tritiated thymidine method, 45 Trochozoa, 136 troglobiotic meiofauna, 344 troglobitic, 5, 190, 194 meiofauna, troglobionts, 338 Troglochaetus beranecki, 341 trophic niche-partitioning, 246 role of ciliates, 405 specialization, 186 tropical regions, 276 true anoxybionts, 299 Trump’s fixative, 78 tubes, 57 tubicolous, 195 tubificid and naidid oligochaetes, 281 Tubificidae, 217 Tubiluchus corallicola, 158 tun stage, 168 Tunicata, 231 tunicates, carnivorous, 233 Turbellaria, turbellarians, 120, 271, 327 turgoid cells, 97 turnover rate, 386 two phase model, 343 Typhloplanoida, 122
U ubiquitous, 247 Uhlig compression chamber, 82 Unela, 226 uniciliated, 127 unimodal size spectra in lakes, 381 univariate methods, 362
527 univoltine, 125 upwelling, 284 areas, 293 Urodasys, 162 Usnel corer, 70 utilization rate of microphytobenthos, 51 V vagile phytal meiofauna, 321 Van Veen grab, 66 vascularization, 97 ventral pharyngeal bulb, 220 vermiform, 225 crustaceans, 180 vertical distribution, 308 distribution pattern, 261 temperature gradient, 23 vibracularia, 229 vicariance, 243, 245, 256, 342 viviparity, 100 void system, 14 W wading birds, 408 Wandesiidae, 205 water bears, 166 column transportation, 249 content, 19 flow as an ecofactor, 332 saturation, 19 water-bound dispersal, 253 web of biotic ecofactors, 62 Wentworth scale, 10 wheel organ, 130 wrack zone, 217 wriggling, 95 writhing motion, 141 X Xenophyophorea, 106 Z zonites, 156 zoogeographical indicators, 244 zoogeography, 249 zooids, 229