No title

Advances in

MARINE BIOLOGY VOLUME 19

This Page Intentionally Left Blank

Advances in

MARINE BIOLOGY VOLUME 19 Edited by

J. H. S. BLAXTER Dunstafnage Marine Research Laboratory, Oban, Scotland

SIR FREDERICK S. RUSSELL Plymouth, England

and

SIR MAURICE YONGE Edinburgh, Scotland

Academic Press

1982

A Subsidiary of Harcourt Brace Jovanovich, Publishers

London

New York

Toronto

Sydney

San Francisco

ACADEMIC PRESS INC. (LONDON) LTD 24-28 OVAL ROAD LONDON N W l 7 D X

U.S. Edition published by ACADEMIC PRESS INC. 111 FIFTH AVENUE NEW YORK, NEW YORK 10003

Copyright

0 1982 by Academic Press Inc. (London) Ltd.

All rights reserved

NO PART O F THIS BOOK MAY BE REPRODUCED IN ANY FORM BY PHOTOSTAT, MICROFILM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS

British Library Cataloguing in Publication Data Advances in marine biology. V O l . 19 1. Marine biology 574.92’05 QH91.Al TSSN 0065-2881 ISBN 0-12-026119-7 TJCCK 63- 14040 Typeset and printed in Great Britain by John Wright & Sons (Printing) Ltd. a t the Stonebridge Press, Bristol

CONTRIBUTORS TO VOLUME 19 R. S. BAILEY, Department of Agriculture and Fisheries for Scotland, Marine Laboratory, Aberdeen, Scotland.

J. DAVENPORT, N.E.R.C. Unit of Marine Invertebrate Biology, Marine Science Laboratories (University College of North Wales),Menai Bridge, Gwynedd, United Kingdom. C. C. EMIG,Station Marine d’Endoume (Laboratoire associd a u C.N.R.S.), Rue de la Batterie-des-Lions, 13007 Marseille, France. P. W. GLYNN,Smithsonian Tropical Research Institute, APO Miami 34002, U . S . A .

This Page Intentionally Left Blank

CONTENTS . .

. .

. . . . . . I. Introduction . . * . . . . . . . 11. Systematics . . . . 111. Reproduction and Embryonic Development. A. Sexual patterns and gonad morphology . . . . B. Oogenesis . . . . . . . . C. Spermiogenesis . . . . . . D. Release of spermatozoa . . . . . . E. Fertilization . . . . . . F. Spawning . . . . . . G. Embryonic development . . . . H. Embryonic nutrition . . . . . .

. . . .

. . . . IV. Actinotroch Larvae . . . . . . A. General account . . . . B. Development of the actinotroch species C. Larval settlement and metamorphosis . . D. Metamorphosis. . . . . .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. . . .

. . . _

CONTRIBUTORS TO VOLUME 19 . .

. .

. .

V

The Biology of Phoronida

c. c. E M I G

. . . . V. Ecology . . A. Tube . . . . . . B. Biotopes. . . . . . C. Ecological effects . . D. Predators of Phoronida E . Geographical distribution VI. Fossil Phoronida . . . .

. .

. . . .

. . . .

. . . .

. . . .

. .

. . . .

. . . . . . . .

. . . .

. .

. .

2 2 5 5 8 8 9 13 14 14 17 17 17 21 31 33

38 38 43 47 49 50 50

. . . . . . . VTI. Feeding . . A. Lophophore and epistome .. . B. Mechanisms of feeding. . . . . C. The alimentary canal . . . . . D. Food particles ingested by Phoronida E. Uptake of dissolved organic matter .

. . .

. . . .

.

. . . . . .

53 53 56 57 61 62

. . . . . . VIII. Circulatory System . . . . . . . . . . A. General structure . . B. Circulation and function . , . . , . C. Wall structure of the circulatory apparatus. . D. Blood corpuscles . . . . . . . .

63 63 64 66 69

. . IX. Phylogenetic Relationships of Phoronida . . A. Archimeric subdivisions, morphological adaptations and phylogenetic relationships . . B. Other phylogenetic expression . . . . C. Relation of the Phoronida to the other Lophophorata . . . . . . . . D. Relation of the Lophophorata to the other related phyla. . . . . . . . . .

71

X. References

.

.

. .

. .

. .

.

. .

. .

71 75

76 80

. .

81

Coral Communities and Their Modifications Relative to Past and Prospective Central American Seaways

P. W. GLYNN I . Introduction

. .

. .

. .

. .

. .

11. Panamic Isthmian Setting . . . . . . A. Paleoecological background . . . . B. Character of extant reefs .. . . C. Availability of colonists . . . . D. Access through the Panama Canal and proposed inter-ocean seaway . .

. .

92

. .

93 93 94 100

. . . .

. . the

. .

102

111. Theoretical Considerations . . . . . . . . . . . . A. Attributes of good colonists . . B. Establishment in relation to the biotic community . . . . . . . . . .

103 103 105

ix

IV. Speculations on some . . Interactions. . A. Feeding relations B. Competition . . C. Symbiosis . . D. Diseasedorganisms E. Biotic disturbance V. Conclusions

. .

VI. Acknowledgements VTI.

References

. .

Potential

. . . . . .

. . . .

Ecological

. .

. .

. .

. .

. .

. .

. . . .

. .

.

. . . . . .

. .

. . . .

.

. .

..

. .

106 107 113 118 119 120

. .

121

. .

. .

. .

. .

. .

. .

. .

. .

122 . .

. .

122

Environmental Simulation Experiments upon Marine and Estuarine Animals

J . DAVENPORT I. Introduction . .

..

. .

. .

. .

133

11. Variability of the Inshore Environment . . . . A. Temperature and salinity fluctuations at an intertidal estuarine site . . . . . . B. Rock pool physico-chemical conditions . . 111. Development of Simulation Equipment IV. Regimes

. .

. .

. .

V. Temperature Experiments . . . A. Survival. . . . . . . B. Development . . . . . C. Reproduction . . . . . D. Adaptation . . . . . E. Interaction with other factors

. .

. . .

137 139 143

. .

. .

145

. .

. .

160

. . . .

. .

166 166 175 177 I79 182

. .

. . . . . .

. . . . .

. . . . .

X

('ONTEN'I'S

. VI. Salinity Studies . . . A . Survival. . . . B. Behavioural responses. C. Reproduction . . . . D. Growth . . .. E. Feeding . . . . . F. Osmotic/ionic responses G. Oxygen consumption .

. . . .

. .

. .

. .

. .

. . . .

. .

. .

. .

. .

. .

..

.

.. ..

..

.. . .

. .

. . . .

.

. .

. .

VII. Oxygen Tension Studies

. .

* .

. .

. .

. .

. .

. .

. .

236

. .

. .

. .

. .

. .

240

. .

. .

. .

. .

. .

. .

VIII. Pollutant Studies IX. Conclusions

.

.

X . Acknowledgements XI. References

. .

.

. . . .

184 185 190 207 208 209 209 230 233

242

. .

242

The Population Biology of Blue Whiting in the N o r t h Atlantic

R. S. BAILEY

. .

. .

. .

. .

258

. . . .

. .

. .

. .

. . . . . .

. .

. .

. .

. .

. .

. .

259 259 266 266

111. The Ecological Role of Blue Whiting. A. Food and feeding . . . . . . . . B. Predators . . C. Parasites and diseases . . . . . D. Competition . . . . . .

. . . . . .

. .

. .

. .

..

. .

I . Introduction . .

. .

11. The Life History . . A. The planktonic stages B. The immature phase C. The adult phase

. . . .

. . . .

276 276 280 283 284

(‘ONTENTS

IV

Population Dynamics. . A, Introduction . . B. Age determination C. Growth . . D. Mortality . . E . Fecundity . . F. Condition . . G. Stock discrimination I

v

,

XI

.. . . . .

.. . . . .

..

. .

.. . .

. .

1

.

. .

. .

. .

. .

. . . . .

. . . . .

. . . .

. . . . .

. . . . Distribution . . . . A. Eggs and larvae . . . . B. lmmatures . . . . . . C. Adult distribution and migrations D. Ecological correlates . . . . E . Depth distribution . . . .

. . VI. Abundance and Stock Size . . A. Trends . . . . . . . . B. Absolute estimates of stock size

. . . . .

. . . .

.. . .

. .

. .

. . . .

. . . .

.

. .

.

. .

. .

. . . .

286 286 286 290 297 300 302 302 306 306 309 314 322 323

. . . . ..

. . . .

. .

326 327 329

..

. .

. .

. .

334

VIII. The Southern Blue Whiting . .

. .

. .

. .

337

. .

. .

. .

. .

. .

340

.

.

. .

. .

. .

. .

. .

. .

. .

VII. Exploitation

IX. Summary

. .

..

X. Acknowledgements

XI. References

..

..

342 . .

342

This Page Intentionally Left Blank

Adz). Ma? . Bid .. V0l . 19. 1982. pp . 1-89 .

THE BIOLOGY OF PHORONIDA C . C . EMIG Station Marine d'Endoume (Laboratoire associe' au C . N . R . S . 4 1 ) , 13007 Marseille. France

I . Introduction . . . . . . . . . . .. I1. Systematics . . . . . . . . . . .. I11. Reproduction and Embryonic Development .. A . Sexual patterns and gonad morphology .. B. Oogenesis . . . . . . . . . . .. C. 'Spermiogenesis . . . . . . . . .. D . Release of spermatozoa . . . . . . .. E . Fertilization . . . . . . . . .. F. Spawning . . . . . . . . . . .. G . Embryonic development . . . . . .. H . Embryonic nutrition . . . . . . .. IV . Actinotroch Larvae . . . . . . . . .. A . General account . . . . . . . . .. B . Development of the actinotroch species .. C . Larval settlement and metamorphosis . .. D . Metamorphosis . . . . . . . . .. V . Ecology . . . . . . . . . . . . .. A . Tube . . . . . . . . . . . .. B . Biotopes . . . . . . . . . . .. C. Ecological effects. . . . . . . . .. D . Predators of Phoronida . . . . . . .. E . Geographical distribution . . . . .. VI . Fossil Phoronida . . . . . . . . . .. VII. Feeding . . . . . . . . . . . . .. . . . . A . Lophophore and epistome .. B . Mechanisms of feeding . . . . . . .. C. The alimentary canal . . . . . . .. D . Food particles ingested by Phoronida . .. E . Uptake of dissolved organic matter . . .. VIII . Circulatory System . . . . . . . . .. A . General Structure . . . . . . . .. B . Circulation and function . . . . . C. Wall structure of the circulatory apparatus. . D . Blood corpuscles . . . . . . . . . .

.. ..

. .

..

..

..

..

. .

..

..

..

. .

..

..

.. ..

..

.. ..

..

..

.. .. ..

..

..

. . ..

..

.. ..

.. ..

..

.. .. ..

..

.. .. .. ..

..

.. .. ..

..

..

..

. . ..

..

.. .. .. .. ..

.. ..

..

.. ..

.. .. . .

.. .. ..

..

..

..

.. ..

..

..

.. . . . . . .

. . . . . .

.. . . . .

. . . .

..

. . .. . .

. .

..

..

..

.. ..

.. ..

.. ..

..

.. ..

.. ..

..

..

..

..

..

..

. .

. .

..

..

..

. . . . .

.. ..

.. ..

..

. . . . .

..

.. . . .. . .

.. .. ..

2 2 5 5 8 8 9 13 14 14 17 17 17 21 31 33 38 38 43 47 49 50 50 53 53 56 57 61 62

63 63 64 66 69

2

C. C. EMIG

IX. Phylogenetic relationships of Phoronida .

. . . . . . . . . . . A. Archimeric subdivisions, morphological adaptations and phylogenetic relationships . . . . . . . . . . . . . . . . . . . . B. Other phylogenetic expression. . . . . . . . . . . . . . C. Relation of the Phoronida to the other Lophophorata . . . . . . D. Relation of the Lophophorata to the other related phyla . . . . . X . References. . . . . . . . . . . . . . . . . . . . .

71 71 75 76 80

81

I. INTRODUCTION Since the last decade, the view that the Phoronida form a “minor phylum” has changed on account of their world-wide distribution, their ecological interest and their phylogenetic relationships. Known since the Devonian, the Phoronida, an exclusively marine group, are regarded as a class of the phylum Lophophorata (Emig, 1977a).As a result of the development of ecological investigations, our knowledge of the biology of the Phoronida has advanced in different disciplines. It is only recently that the variability in the taxonomic characteristics has become sufficiently known to establish the systematics of those phoronid species which are currently recognized (Emig, 1971a, 1974a, 1979). Larval development and the systematics of the actinotroch larvae also need detailed study. The extensive controversies concerning the phylogenetic relationships of the Phoronida have been in general due to lack of knowledge of embryonic and larval morphology and development. I n addition, some basic aspects of the biology of the Phoronida still need to be studied in detail. Thus, the aim of the present review is to stimulate questions which have to be answered in future investigations, and have become necessary since the previous reviews by Cori (1939) and Hyman (1959).

11. SYSTEMATICS The possession of common characters, especially that of the lophophore, proves an affinity between Brachiopoda, Bryozoa (Ectoprocta) and Phoronida, which is implied by several authors by referring them to Lophophorate phyla. Others, including myself (Emig,. 1977a), group them to form a phylum Lophophorata, of which each group then constitutes a class. As suggested by Hyman (1959), the name Tentaculata, proposed by Hatschek (1888), “is unfortunate, for tentacles occur in many unrelated animal groups”,

THE BIOLOGY OF PHOKONIDA

3

and has to be rejected; only the name Lophophorata should now be used. The diagnosis of the class Phoronida is as follows (Emig, 1977a): free-living, solitary, in a cylindricaJ tube of their own secretion; three body parts in larval and adult forms (archimeric regionalization); presence of a lophophore; trunk slender and cylindrical with an endbulb, the ampulla; U-shaped digestive tract; nervous centre between mouth and anus, a ring nerve at the basis of the lophophore, one or two giant nerve fibres; metanephridia; closed-type circulatory system with red blood corpuscles. I n the Phoronida only two genera-Phoronis Wright 1856 and Phoronopsis Gilchrist 1907-and some ten species are currently recognized. The former genus is identified by the absence of the epidermal collar-fold below the lophophore, while the genus Phoronopsis has such a collar-fold (Fig. 1). The following characteristics are used to distinguish the species: habitat, lophophore shape, nephridial morphology, number of giant nerve fibres, longitudinal muscle formulae, gonads and accessory sex glands, when available. Some other additional features are sometimes used: absence of one or two lateral mesenteries, unusual trunk muscle disposition and differences in the circulatory system (Emig, 1974a). On the bases of all those taxonomic characteristics the systematics of the adult species have been established and several previously described species may therefore be considered as synonyms (Table I). For accurate identification adult phoronids need histological sections at different levels of the animal, usually the whole of the anterior region and posterior third of the trunk, both of which contain the main taxonomic features. Phoronids must be fixed quickly to prevent lophophore autotomy. Good results are obtained with Bouin’s fixative, paraffin wax embedding, sectioning at 7 pm and Azan staining after Heidenhain’s method (Emig, 1971a, 1979). In several recent papers on Phoronida, particularly of American investigators, some synonyms (Phoronis architecta, P . vancouverensis, Phoronopsis viridis) are still cited as species: such usage should cease so as t o avoid confusion and misinterpretation, or the species status must be established by a new description on the basis of the cited taxonomic features. The larva of Phoronida, named Actinotrocha by Miiller (18461, was described before the discovery of the adult form. But the International Commission of Zoological Nomenclature accepted as valid the name Phoronis; thus the actinotroch keeps a separate name considered as a technical one, which is sometimes still different from

4

&err;

ganglion

LNephridiurn 'Diaphragm

Oesophagus

----Intestine

-Prestomach

Median vessel

Lateral vessel with caeca

r

\--Ovary

FIG.1. Diagram of

B

Testis

phoronid, showing the main anatomical features.

the adult species name (SilBn, 1952). A first review of the Actinotrocha, related to the adult form, with taxonomic characteristics is proposed and discussed in Section IV, B.

Genus Phoronis Wright 1856

Species

Synonyms

Not a n actinotroch: SilBn, 1954a

ovalis Wright, 1856 hippocrepia Wright, 1856 ijimai Oka, 1897

Actinotrocha*

I

gracilis kowalewskii caespitosa capensis vancouverensis

A . hippocrepia S i l h , 1954a

A . vancouverensis Zimmer, 1964

australis buskii Haswell, 1883 ( 2 bhadurii Ganguly and Majumdar, 1967) muelleri Selys-Longchamps, 1903 sabatieri psammoyhila architecta Cori, 1889 pattida SilBn, 1952

I

Phoronopsis albomacutata Gilchrist, 1907 Gilchrist, 1907 harmeri Pixell, 1912

pacijca viridis striata

,4.branchiata Muller, 1846

A , sabatieri Roule, 1896 A . paltida SilBn, 1952

A . harmeri Zirnmer, 1964

californica Hilton, 1930 *The adult form of Actinotrmha wilsoni has not yet been established while some larval forms remain unknown.

111. REPRODUCTION AND EMBRYONIC DEVELOPMENT A . Sexual patterns and gonad morphology Phoronid species are obviously either hermaphrodite or dioecious (Table 11, Fig. 3) though several previous authors, such as Roule (1900),Torrey (1901),Brooks and Cowles (1905),Selys-Longchamps (1907), Pixell (1912) and Cori (1939),suggested a possible protandric condition owing to the presence of spermatozoa in the metacoelom and around the ovary of females, or to the apparent succession of male-female over the reproductive period. Such a possibility can be ruled out; the presence of spermatozoa in females results from

internal fertilization which occurs in all phoronid species. A considerable range of gonad maturation occurs among the individuals of a population over the whole reproductive period (Fig. 2); evidence for protandry has never been found.

1 80 -

0

60-

I

Y)

3

.

8’

P

->

V

40-

e f

0

z

20 -

Ok

. .-,

RP

,

,

,

,

. .

,

-

.

1970

RP ,

,

.; , ,

,

,

1971

PI(:.2. Distribution (in yo)of mature individuals in a population of Phorrmisp~ammophilaovpr one half year (Marseilles, Prado Reach a t 4 m deep). The present data (unpublished) were obtained during the study of Emig and Emig (1975).

The gonads are applied to the lateral blood vessel and its large caeca in the left oral cavity of the metacoelom at stomach level and in the ampulla (Figs 1 and 3). They are intimately associated with the vasoperitoneal tissue. In hermaphrodite species, the testis lies on the oral side of the lateral vessel and the ovary on the anal side (Fig. 3b). In Phoronis pallida this disposition can be reversed ( S i l h , 1952; Emig, 1969). Ovary and testis are very close to each other, being only separated by a narrow distinct vasoperitoneal cell layer; both are clearly simultaneously active in one animal (Fig. 3b). I n dioecious species (Fig. 3c, d), the gonads can extend into the right oral cavity of the metacoelom, where a secondary lateral blood vessel generally occurs, and which is also filled by vasoperitoneal tissue, and sometimes extends into the anal cavities. The sexes cannot be distinguished externally, although the ampulla seems sometimes whitish in males.

FIG.3. Cross-sections at the gonad level. (a) Phoronis australis: development of vasoperitoneal tissue around the lateral vessel; (b) P. a.ustraZis: gonad maturation in a hermaphroditic species, showing the important development of the vasoperitoneal tissue in all coelomic compartments, presence of the secondary lateral vessel; (c)P. psammophila: mature female; (d)P. psammophila: mature male. bp: blood plexus; i: intestine; Iv: lateral blood vessel; mv: median. blood vessel; ov: ovary; slv: secondary lateral vessel; sp: spermatids: spz: spermatozoa; st: stomach; te: testis; vpt: vasoperitoneal tissue.

8

C C' EMIG

Gonads become mature at different seasons, often extending over 8-10 months. The peak of reproduction occurs in late spring and summer (Fig. 2), according to most investigators. It seems that individuals which metamorphose in spring show a reproductive period in autumn, in Phoronis psammophila (cf. Emig and Emig, 1975). B. Oogenesis The ovary differentiates from the peritoneum along the lateral blood vessel and its capillary caeca which seem to be of great importance in gonad development. The germ cells in different stages of development are arranged in groups around and along the blood caeca. They grow inside the vasoperitoneal tissue which then degenerates gradually. The oocytes become somewhat flattened, and the first meiotic division begins and proceeds to a metaphase arrangement; at this stage the division stops until the ova leave the ovary to enter into the trunk coelomic fluid. The vasoperitoneal tissue arises from the peritoneum. Its development starts just before that of the gonads. The tissue rapidly fills the oral cavities of the metacoelom and sometimes the anal ones through the numerous small holes distributed here and there in the mesenteries (Fig. 3). It extends over the posterior third of the trunk and reaches its greatest development at the breeding season. The vasoperitoneal tissue is considered as a nutrient layer owing to the richness of the yolk-like substance which nourishes the growing oocytes while at the same time the follicle widens. After the spawning of the oocytes, the vasoperitoneal tissue is said to be almost eliminated, and a new reproductive cycle can begin. According to Ohuye (1943), the vasoperitoneal tissue seems also to be a hematopoietic organ. Several authors considered the vasoperitoneal tissue to be unpaired (Cori, 1939; SilBn, 1952; Forneris, 1959), but, like SelysLongchamps (1907), I suggest that this tissue has a paired origin, coming from the peritoneal cells of the blood vessels in each oral cavity (along the lateral vessel in the left oral and the secondary lateral vessel in the right oral). This disposition occurs especially in dioecious species, but is less distinct in hermaphrodite ones where an unpaired origin cannot be excluded.

C. Spermiogenesis Spermiogenesis, like oogenesis, develops within the vasoperitoneal tissue. The male germ cells arise in the wall of the blood-vessels

THE BIOLOGY OF PHOROK1I)A

9

from the peritoneum; they meet first near the lateral vessel, anlage of the testis. At this stage, small spermatogonia and oogonia are almost identical in shape and aspect and cannot be distinguished. Then, the spermatogonia increase in number, around and between the large caeca; they aggregate more or less loosely to one another to form either radial strings or small masses containing cells at about the same stage (Fig. 3b, d). The development process of spermiogenesis has never become known owing to the great difficulty in following the germinal cell sequence. The formed spermatozoa appear usually on the periphery of the testis in cohesive clumps: heads are together and tails free, both being of about equal length (Ikeda, 1901; S i l h , 1952; F r a n z h , 1956; Zimmer, 1972; and my own unpublished observations). As those previous authors found, the V-shaped spermatozoa of Phoronida (Fig. 4)are of a highly “modified” type (in contrast to the primitive type: Franzbn, 1956, 1977). Such a sperm structure is connected with internal fertilization and spermatophore production.

FIG.4. Spermatogenesis of Phoronis pattida: (a)-(c) spermatids; (d) sperm (after FranzGn. 1956).

D. Release of spermatozoa Mature spermatozoa break away from the testis into the metacoelom and aggregate into a loose spherical mass near the nephridial funnels by currents created by their heavy ciliation. The sperm mass is compacted within the nephridial ducts where

10

C. C . EMIG

Ring ne

Nephridiopore

.

__--

._

Anus

- _ _ - .

(b)

( 0 )

FIG. 5. Lophophoral organs (accessory spermatophoral organs). (a) Looking into the lophophoral concavity of a mature Phoronis psan~mophilawith large and glandular lophophoral organs (after Emig, 1979); (h) lophophoral concavity of a mature Phoronis harmeri with large and membranous organs showing their innervation (left side) and their morphology with the three regions demarcated by dotted lines (right side) (after Zimmer, 1964). The small lophophoral organ type is represented in Fig. 7 .

(b)

( 0 )

FIG.6. Sperrnatophores: (a)of type A (Phoronis ijimai); (b) of type B (Phoronopsis harmeri) (after Zirnmer, 1964).

THE BIOLOGY OF PHOHONIUA

11

orientation of the spermatozoa occurs, and is then extruded through the nephridiopore along the spermatic groove to the lophophoral organs where the spermatophore gradually takes shape. Nephridia serve also as gonoducts, as Dyster (1859) first observed. Crossfertilization seems to be the rule; according to Zimmer (1964), the maturation of the spermatozoon is probably dependent on secretion from either the nephridia or the lophophoral organs, which could provide a mechanism for the avoidance of self-fertilization. The term “lophophoral organs” has previously been used to describe all glands which occur in the lophophoral concavity. Many hypotheses have been put forward as to their possible functions (sensory: Caldwell, 1882; McIntosh, 1888; Selys-Longchamps, 1907; Gilchrist, 1907; secretory: Benham, 1889; Masterman, 1900; sensory and secretory: Forneris, 1959; S i l h , 1954b; selection of sand grains for tube formation: Andrews, 1890); also correlations with gonad development have been suggested by Brooks and Cowles (1905), Selys-Longchamps (1907),Gilchrist (1907), Silhn (1952) and Hyman (1959). The true function of the “lophophoral organs” has only Lophophone /

:I

v0

t

\

and basal nidamenlal glands

0.5rnm

PIC.7. Nidamental glands: looking into the lophophoral concavity of mature phoronids with brooding patterns, viewed from the distal end. (a) Nidamental glands of type 2a (I’hormis hippocrepia, P. ijimai),developed on the floor of the concavity and on the inner tentacle row at the inner side of the horseshoeshaped end (respectively basal and tentacular nidamental glands); (b) of type 2c (Phormispsammophila), formed along the inner tentacle row; (b’) anal view of the anterior body part showing the position of the brood mass in the lophophoral concavity; (c) of type 2b (Phorais australis), extended from the floor of the concavity into the several coils of the lophophore at the inner surface of the inner tentacles (after Emig, 1977b).

Species

Sexes

Phoronis ovalis

8 ?

Phoronis hippocrepia Phoronis ijimai Phoronis australis

G

Egg Diameter in pm types

Number per individual u p to

Release

I n one time No spermatophore( 1 )

125

40

85-100

100

Continuous

90-110 100-130

400 300

Continuous Continuous

%

8(rl20

400

Periodic

Phoronopsis albomaculata

69

100

Phoronis muelleri Phoronis pallida Phoronopsis harmeri

J?

5&65

500

50-70

500

6@65

1000

Phoronis

$

1

2

Q

Spermato phore typm

Type A

- _psammophila - -- _ - - - _ - -- _- - - - - - - - - - - - - - - - - - - - - - - - - - - -

Q

3

S?

?

(Type R O

1

Continuous

Type B ~

Phoronop9is Californica

39

?

?

?

?

(Type B ? )

recently been discovered by Zimmer (1967): the general term lophophoral organs overlaps the male and female accessory sex glands, respectively lophophoral organs (seasu stricto) and nidamental glands (in brooding species). The development of both sex glands is correlated with gonad maturation. The lophophoral organs (s.s.)may be small (Fig. 7a, c) or large (glandular or membranous; Fig. 5 a , b ) and occur in males and hermaphrodite species, but are usually lacking in Phoronis ovalis (Table 11).They secrete the spermatophoral membrane and assist in spermatophore formation which is of general occurrence in Phoronida. However, several authors (Ikeda, 1903; Rattenbury, 1953; SilBn, 1954a) have observed direct release of spermatozoa into the sea water, but that seems t o be exceptional. In Phoronida, two types of spermatophores can be distinguished (Zimmer, 1967; Emig, 1980).The A type is an ovoid mass of spermatozoa produced by small

13

THE HIOLOGY OF P H 0 R O S I I ) A

Types of developmental patterns

Oviposition and embryonic development

Actinotrocha species

Pelagic l i f p

1

Brooding in Not a true parental tube actinotroch during 4-5 days

2

Brooding on A . hippocrepia 9-14 days nidamental glands A . vancouverensis (after during about brooding 7-8 days ? period)

Settlement on

Short stage 4 days Creeping stage 3 days Hard substrate (burrowing or encrusting

A . sabatieri

.------- - - - - - - - - - - - - - - - - _ (2)

3

?

Direct release into the

1

?

A . branchiata 18-22 days

ambient A . pallida sea water (no brooding) entirely A . harmeri pelagic existence ?

?

1

Soft substrate (embedded vertically)

?

lophophoral organs (Table 11; Fig. 6a) which is produced by burrowing or encrusting hermaphrodite species which are all living in intimate dense populations. The B type is a large spermatophore in two parts, a spherical mass of spermatozoa to which is attached a wide spiral float (Table 11; Fig. 6b). This type seems to be formed by species with large lophophoral organs, living embedded vertically in soft bottoms, often in sparse populations. The spermatophores are greatly assisted in their escape by water and lophophoral ciliary currents: those of A type are probably rapidly collected by one of the nearest individuals and those of B type can float away to other, sometimes far distant specimens.

E . Fertilization The transport of the sperm t o female or hermaphrodite species is effected by means of the spermatophore. The main mechanism of

14

C. V. EMIQ

insemination seems to be the penetration of the sperm mass into the metacoelom through the nephridial duct: this is the natural access to the ovary. It is corroborated by many observations of previous investigators, such as Brooks and Cowles ( 1905),Selys-Longchamps (1907), Kume (1953), Rattenbury (1953). Forneris (1959) and Zimmer (1967). Nevertheless, Zimmer (1972) observed the drawing into the lumen of a tentacle downwards to the ovary after perforation of the diaphragm. Fertilization in Phoronida appears to be internal. The presence of spermatozoa in the metacoelom and around the ovary of females (in dioecious species) has suggested protandry to several authors (see Section I, A). As indicated above, cross-fertilization seems to be a rule in hermaphroditic species. Fertilization occurs in the trunk coelom usually just after the egg escapes from the ovary. F . Spawning The ova rise into the nephridial funnels and are discharged into the lophophoral concavity through the nephridia: spawning usually takes place at all hours of the day and night. I n the majority of the phoronid species i t is more or less continuous over a number of days; however, spawning may be periodic in Phoronis psammophila (cf. Emig, 1974b, 1977b) and only once in Phoronis ovalis (cf. S i l h , 1954a). The ova are directly released into the ambient sea water, or brooded in nidamental glands or in the distal end of the tube in Phoronis ovalis (Table 11). Species with brooding patterns produce and release less eggs and the egg number decreases while the egg size increases; however, S i l h ( 1954a) suggested that the estimated number also increases with the body volume. The function of the nidamental glands which occur only in brooding species is the attachment of the ova (by means of mucous secretion) to the embryonic masses and the maintenance of the integrity of these brood masses. According to Zimmer (1964) and Emig (1977b) the nidamental glands are of three types (Table 11),which are illustrated in Fig. 7.

G. Embryonic Development Only when the egg comes in contact with sea water does it start the expulsion of the polar bodies and the subsequent developmental stages. Phoronids show three different types of egg development (Table 11; Fig. 8 ) . The segmentation is similar in all species: total,

. 25prn

Protocoel

Ectoderrn

.arval tentacle n Gastral ;late

Mesoderm

'Y

Nephridial prirnordium

Anus

FIG.8. Egg cleavage and embryonic development in Phoronida. (a)Egg cleavage in species of type 2 (see Table 11);(b)blastula and gastrulation of developmental type 2. and (c) of type 3; (d) some stages of gastrula development (after Emig, 197410, 197713, 1979).

equal or subequal, and the cleavage is of typically radial type, though biradial in some stages. However, in egg developmental type 3, there occurs sometimes an apparent spiral arrangement which is induced by compression or variations in the orientation of the blastomeres (Zimmer, 1964; Emig, 197413, 1977b), and also egg cleavage within the metacoelom which must be considered as an abnormal pattern. The development reaches the blastula stage (Fig. 8b, c), a thickwalled ciliated coeloblastula in type 2 and a thin-walled one in type 3, but in both types the blastocoel has about the same diameter (3540pm). The gastrula arises by a typical invagination (Fig. 8). During this process the gastrula of type 2 virtually obliterates its blastocoel by wall compression, while this cavity remains extensive in type 3. With the elongation of the archenteron, the embryo acquires a new bilateral symmetry perpendicular to the polar axis of the egg. At the gastrula stage (Fig. 8d), the differentiation of the archenteron (endoderm) produces a stomach and an intestine, the exterior opening of which, the anus, arises by perforation of the ectoderm without the formation of a proctodaeum. The oesophagus is produced by an ectodermal penetration of the posterior part of the vestibule: this process pushes inside the blastopore which remains as the boundary between the ectodermal oesophagus and the endodermal stomach (Fig. 8d). The mouth marks later the entrance into the digestive tract. The anterior ectoderm differentiates (a characteristic feature of the phoronid larva) the preoral lobe, on which an epidermal thickening leads to the nervous ganglion. In brooding species, the embryos are attached to the mucous cord of the nidamental glands by the apical area of the preoral lobe. A t the postero-ventral region the tentacular ridge appears, and below in the midline the primordium of the protonephridia develops as an ectodermal invagination (Fig. 8d). According t o the recent interpretation of the mesoderm origin (cf. Emig, 1977b), the site and mode of mesoderm proliferation in Phoronida show marked similarities t o the enterocoelous mode: the mesoderm originates as isolated cells proliferated from the anterior and ventro-lateral areas of the archenteron in two phases. The pattern does not differ significantly from this latter mode and must be considered as a modified enterocoelous type. The differentiation of mesoderm begins in the gastrula, but only one coelomic cavity occurs, the protocoel. This arises from the anterior mesoderm cells either as a schizocoel (in Phoronopsis hurmeri: Zimmer, 1964) or by mesodermal wandering (in Phoronis ijimai and P . psammophila:

THE BIOLOGY O F PHOROKlD.4

17

Zimmer, 1964; Emig, 1974b; P. hippocrepia). The protocoel largely fills the preoral lobe (Fig. 8d). Several mesodermal cells budded off from the lateral archenteric areas proliferate to form in the posterior end of the gastrula a solid mass which later gives rise to the metacoel. With the development of the gastrula the blastocoelic cavity reappears rapidly in embryos of type 2. The embryos of brooding species escape from the brood masses with incipient tentacles, up to about six in number, according t o the species, usually at the beginning of the larval stage.

H. Embryonic nutrition The ova of types 1 and 2 are apparently supplied with sufficient yolk to last until the pelagic life without food; in non-brooding species (type 3), the amount of yolk is too small to allow a lecithotrophic mode of life during the same period of time: in all three types the larval size is about the same at the end of this period ( S i l h , 1954a). Thus, during pelagic existence embryo and larva ingest diverse organism's (as flagellates, diatoms, small larvae, etc). Digestion is always intracellular. The mode of embryonic nutrition has so far only been established by short and incomplete observations by several previous investigators, so that new careful studies are obviously needed on this topic.

IV. ACTIKOTROCH LARVAE

A. General Account The characteristic phoronid larva is termed Actinotrocha (or actinotroch) which must only be used as a technical name of the larval forms as stated by Sil6n (1952) in a footnote. The actinotroch has a pelagic existence: swimming near the sea surface for several days (Table 11). The larva is a familiar constituent of the plankton, with a world-wide distribution. Only Phoronis ovalis is a curious exception (Sil&n,1954a). The actinotroch seems to be photopositive, but its position at the sea surface depends upon the water movements, which if they are strong induce the larva to sink down (Hermann, 1976). The general form and the gross structure of the Actindrocha are familiar, established by several authors and also given in textbooks (e.g. Hyman, 1959; Emig, 1979, 1980). Thus, they are only briefly described here to facilitate the understanding of the different larval stages (Fig. 9) and the processes of metamorphosis.

Metarnorphosts

-

4 330

A. hippocrepio k

4

A. voncouverensis 4

4 220

A. sabotieri t i

6 220

I

,-.

6

1

.

6

I

350

8

IU

430

w M

10 470

-8

-

340

, _

.c

8

12 620

I

A

U

450

f

10 700

500

14

'H

9 0 0

10

12

550

800

i -18

Type3

4 200

A. po/lido I

A.bronchIafo

4

I

,80

A . harmer/ t

A. wi/soni t Appearance of

:

6 2oo ;

4

1

200

'

4

6 300

6

,

' ' 8

300

12

10

8

,

8

6 300

l i l

300

8 350 ,

'

400

, . 10

1 -

,

6-

,

12

'li

10 , 12 , 14 450 ' -475

, J6 W

14

,

660

'

, '

1-12

, 2 0 - , 22 , 24 ' IOOO-' 1200 "%406*

, 16 , 18 A500' 750 '

, 14

'

10 600

I

400

16 720

18 20 650 -760 I_

18 v'915

I-

L2 '0

. H '

26

42 2500

1500

,A

20

'v

1050

'

A

24 '

-9920

'

26 i - 3 8 ( ? ) 1500

*. Metosomol soc 0 Blood mass 0 Dorsal vessel

0 Adult tentacle 4

Piriform organ

H

Mesocoel

FIG.9. Developmental stages of the known actinotroch species, with indication of the stage number (hy number of larval tentacles: upper level, of the body length in pm: lower level) and of the appearance of the main features (by specific signs). See also Figs 12-19 where are represented the main larval stages in lateral view, without the ciliation of the perianal ring.

19

THE HIO1,OBY OF PHORONIDA

During development, the actinotroch elongates the larval trunk which ends in the perianal ciliated ring and increases regularly its number of tentacles. Both structures, tentacles and perianal ciliated ring, are the main locomotory organs of the larva, whilst the tentacles and the preoral lobe have feeding functions. The tentacles develop obliquely on each side of the midventral region, the longest being ventral; their maximum number varies with the species, but also somewhat within each species according to local ecological conditions, especially food availability. The definitive adult tentacles arise either as thickenings of the wall of the larval ones or as eversions under the bases of the larval tentacles (Figs 10, 19, 20). The preoral

Stomach diverticulum Larval tentacle Adult tentocle

entrol mesentery Pylorus h P e r i a n a l ciliated ring

-

130pm

F I G . 10. Diagram of the ventral view of a mature Actinotrocha snbntieri, showing the intrrnal anatomy.

lobe which overhangs the mouth like a hood is a characteristic feature of the actinotroch; it is entirely ciliated with a belt of cilia along the free margin and a strongly ciliated area (especially in larva of type 3) in the centre of the dorsal (anterior) surface oY the preoral lobe at the site of the apical plate, which is the larval nervous ganglion. The remaining epidermal body surface is also ciliated, especially the tentacles and the perianal ring. Just behind the mid-ventral tentacles there is an ectodermal invagination which gives rise to themetasomal

sac. This sac develops between the two leaves of the ventral mesentery and grows to occupy the largest space of the metacoel, sometimes virtually all the coelom (Fig. 10). The protonephridia originate by a single ectodermal invagination that bifurcates rapidly into two separate canals opening laterally on each side of t h e intestine by a tiny pore just behind the tentacles and the trunk septum (Fig. 8d, 10, 11). At the closed proximal end of each nephridial canal arise solenocytes arranged in one to three clusters and lying in the blastocoelic preseptal cavity. In Phoronida the body is divided into

. 10 Frn

- Solanocyte

. N e ~ h r i d i a l duct

,Epidermis

Protonephridium

Nephridiopore

F I ~11. : Protonephridium in a young actinotroch (Actinotrocha hippocrepia) disposition in the larva and cross-section (after Emig, 1980).

three major archimeric regions each with its own unpaired coelomic cavity. At first, the U-shaped protocoel occurs by schizocoely or cell wandering in the space of the preoral lobe (or protosome) and is separated from the blastocoelic collar space (or blastocoelic preseptal cavity) by a septum (preoral septum) just behind the apical plate (Fig. 10).According to Zimmer (1978) the extensive protocoel which occupies the cavity of the preoral lobe in the gastrula degenerates to persist only as a small vesicle situated between the apical plate and the oesophagus near the limit of the preoral lobe in the actinotroch stage. Then the metacoel (or trunk coelom) undergoes schizocoely; it establishes a ventral mesentery which unites the trunk wall to the digestive tract and anteriorly, at the level of the tentacles, a definitive septum (or trunk septum) which assumes the status of a mesentery with the development of the mesocoel (or collar coelom).

Between the preoral septum and the trunk a blastocoelic cavity persists until late in the actinotroch development (Fig. 10). The mesocoel develops into a horseshoe shape (probably by schizocoelic formation according to Ikeda, 1901; Goodrich, 1903) within the blastocoelic space over the trunk septum in well-developed larvae. The digestive tract has elongated with the trunk development and consists of three divisions: the oesophagus opening by the mouth; the stomach in three portions: anteriorly one or two vacuolated diverticula, then a large cavity and posteriorly a small funnel-shaped heavily ciliated cavity entering the intestine by a pylorus, and the intestine opening by the anus in the centre of a ciliated ring (Fig. 10). One to four solid masses of blood corpuscles appear in the blastocoelic collar space. Their number and disposition are used in the identification of actinotroch species. In the fully developed larva there is a dorsal blood vessel, the incipient median vessel, and a t the site of the pylorus a bunch of short blood caeca. The circulatory system is not functional in the actinotroch. The muscle arrangement and the nervous system are complex and vary in the different species. Several actinotroch species are provided with a piriform organ which appears shortly before metamorphosis (Fig. 20a) and is supplied by three long nerves from the ganglion. Several actinotroch species show from about the four tentacle stage a characteristic pigmentation of prime importance in identification (Figs 13,14).

B. Development of the actinotroch species The main developmental stages of the different known actinotroch species are shown in Figs 9 and 12-19, together with some characteristics helpful in their identification. The duration of the whole larval development a,verages probably 19-21 days in all species. Sil6n (1954a) suggested that the length of the actinotroch stage, elapsing between four-tentacle to metamorphosis, is about 12-14 days without difference between brooding and non-brooding phoronid species; only Phoronis ovalis is an exception (Table 11). A brief description of each actinotroch species follows and possible synonyms are proposed. It is suggested that the description of a number of actinotrochs is due t o different interpretations by previous investigators who have mostly studied A . branchiata, to an unintentional misunderstanding of statements by earlier workers, and to the fact that the early workers recognized about 20 phoronid species where there are presently about ten. There is no doubt t h a t

our knowledge of actinotroch species is far from satisfactory and the following list of actinotrochs needs particular attention in the future and probably some modifications, and also additions will have to be made. 1. Larva of Phoronis ovalis The sexual reproduction of Phoronis ovalis has only been described by Silkn (1954a). The embryo escapes from its tube in a gastrula stage about 4-5 days after the egg release. Its transformation and differentiation is so different from that of other phoronid larvae that this embryo cannot be called an actinotroch, as suggested by SilBn (1954a). The development is more direct: the planktonic existence is almost omitted and the short pelagic life of about 4 days serves for larval dispersal exclusively, and no true metamorphosis occurs after a creeping period of about 3 4 days. The whole development elapsing from the egg release to the transformation to the adult phoronid is of about 12-13 days. The external morphological characteristics of the larva of P. ovalis are shown in Fig. 12.

D ( C )

(d

1

FIG.12. Larval development of Phmonisozinlis (after Silen, 1954a).(a) Larva justescaped from

the parental tube, in ventral and lateral view; (b) 2 days after liberation; ( c ) 3 days after liberation; (d)after5 days, creeping stage; (e)just attached larva, 7 days efter liberation, in lateral view.

2. Larva of Phoronis hippocrepia: Actinotrocha hippocrepia

The larva of Phoronis hippocrepia Wright, 1856, was discovered by SilBn (1954a), and since found by Forneris (1959).

The body of Actinotroch,a hippocrepia is opaque; its pigmentation consists of very small pigment granules (dark brown in reflected light) probably contained in the epidermal cells. The granules are distributed in distinct patches at certain fixed points of the body which increase in number from the four-tentacle stage to the last actinotroch stage (Fig. 13).

Q 4-T

6-T

8-T

130pn 1

10-T

Frc. 13 Developmental stages of Artanotrorhn happocrepm, with its characteri
A . hippocrepia possesses two ventral blood masses which fuse in the oldest specimens at the level of the oesophagus, but blood globule clusters on each side of the trunk are situated near the insertion of the tentacles. The stomach diverticulum is unpaired. The tentacles are not more than ten in number; no adult tentacles occur (Fig. 9). According to Sikn (1954a) and Forneris (1959),A . hippocrepia is very similar to A . pallida in general appearance and behaviour and it is difficult to distinguish between the larvae unless they are placed side by side. However, the characteristic pigmentation of A . hippocrepia is the main feature for identification, as is the number of blood masses. 3. Larva of Phoronis ijimai: Actinotrocha vancouverensis Actinotrocha vancouverensis has been described by Zimmer ( 1964), especially its main developed stages. This larval form is the larva of Phoronis ijimai Oka, 1897 (synonym: Phoronis vancouverensis Pixell, 1912; established by Emig (1971b) and confirmed in Emig ( 1 9 7 7 ~ ) ) . A . vancouverensis has a opaque body which is heavily pigmented (two pairs of pigment patches on the preoral lobe, a rather uniform distribution on the collar, only interrupted at the tentacles, uniform

24

76 p n I

6-T

8-T

130gm

10-T

12-T

FK:.14. Main development stages of ActPnotrocha uaticouvere~i~is showing t h e chararteristic pigmentation (after Zimmer, 1964).

but variable in density on the trunk, see Fig. 14). There is a single blood mass on the anterior ventral surface of the stomach. The maximum number of tentacles is 14 in larvae ready to metamorphose, without indication of adult tentacles. The species A of the four actinotroch types described by Ikeda (1901) cannot be considered as the larval form of Phoronis ijimai especially in view of the presence of two masses of blood corpuscles. It seems also that any larva found by this author belongs to A . vancouverensis.

4. Larva of Phoronis psammophila: Actinotrocha sabatieri Actinotrocha sabatieri, discovered by Roule (1896) and described by this author in 1900 and by Selys-Longchamps (1907),is the larva of Phoronis psammophila. It has been recently studied by Veillet (1941) and Herrmann (1977). After the writing of the present paper, Herrmann (1979) published a note on the larval development and metamorphosis of Phoronis psarnmophila, results of which confirmed most of my own observations on A . sabatieri. A . sabatieri is large and transparent. Pigmentat,ion occurs until the six-tentacle stage; at first two pigment masses are located on both sides of the apical plate and later at the distal end of the tentacles. Herrmann (1979) considers that the pigment amoebocytes may represent a nutrient reserve used by the larva during a period of food shortage. The larva does not develop more than 12 larval tentacles. The adult tentacle are represented by a thickening of the wall of the larval tentacles at the end of the ten-tentacle stage. Three blood masses are distributed, two on each side of the stomach diverticulum

25

THE BIOLOGY OF PHORONI1)A

(which is unpaired), and one, unpaired, on the ventral midline j u s t above the insertion of the tentacles (Figs 10, 15). Herrmann (1979) shows the metasomal sac and the perianal ciliated ring during the eight-tentacle stage, the stomach diverticula and the blood masses at the ten-tentacle stage, and the adult tentacles and two longitudinal blood vessels along the stomach during the 12-tentacle stage.

mass

4-T 6 -T

FIG.16 Main

stages of Aclinotroeha ~ u b a f i ~ r i

According to various authors, several actinotroch species are to be considered as synonyms of A . sabatieri. However, the name sabatieri has been retained because this actinotroch has the best complete description and is without doubt the larval form of P. psammophila. The characteristics of Actinotrocha metschnikofJi discovered by Metschnikoff (1869, 1871) have been established by Selys-Longchamps (1907), all being similar t o those of A . sabatieri: 0.6 mm long, up to 16 larval tentacles with anlage of the adult ones as thickenings at the interior bases of the larval tentacles and three blood masses of characteristic disposition. On A . metschnikofji, the statement of Roule (1900)that probably only one actinotroch species occurs in the Mediterranean Sea must be refuted as suggested by Selys-Longchamps (1907), especially because several phoronid species live here and consequently several actinotroch species. Actinotrocha wikoni A , which was named by Selys-Langchamps (1907),is described by Wilson (1881),Cowles (1904a) and Brooks and Cowles (1905) and belongs probably to A . metschnikof’ (presently A . sabatieri); it is about 1 mm long; has up to 18 larval tentacles with definitive ones as thickenings; has no piriform organ; pigmentation is present especially as spots at the bases of the tentacles; has blood

masses until about the 12-tentacle stage, but there are only two of these masses, disposed ventro-laterally to the stomach. Actinotrocha hatscheki, figured by Hatschek (189l), has been briefly described by Selys-Longchamps (1907);all known features are similar to those of A . sabatieri, especially in the maximum number of tentacles (up to 16),no piriform organ and two stomach diverticula. Another species, Actinotrocha ashworthi, described by Selys-Longchamps ( 1907) belongs, I believe, to A . sabatieri: it is 0.65mm long, has about 20 tentacles with anlage of adult ones and three masses of blood corpuscles. Steuer (1933)found two larval forms one of which has the following main characteristics: i t is about 0.6 mm long; has up to 16 tentacles and three blood masses: this form seems to be related to A . sabatieri. Recent!y, the larval form considered by Zimmer (1978) as that of Phoronis architecta (which species is a synonym of P . psammophila according to Emig, 1972a, 1977c) is thought to be related to Actinotrocha branchiata (see following paragraph).

5. Larva of Phoronis muelleri: Actinotrocha branchiata

Actinotrocha branchiata was discovered near Helgoland by Muller (1846), who considered this animal to be an adult. The adult form named Phoronis was described in 1856 by Wright on the English coast. The transformation of this actinotroch into Phoronis muelleri was established by Selys-Longchamps (1903). The other main works on A . branchiata are from Selys-Longchamps (1907), S i l k (1954a), Emig (1973a), Siewing (1974a) and Herrmann (1976). Recently, Zimmer (1978) related a larva to Phoronis architecta, but this larva belongs to Actinotrocha branchiata, which confirmed the confusion introduced by Brooks and Cowles (1905), and discussed by Emig (1977c),between P . muelleri and P . psammophila which may both be mixed in the same locations. Actinotocha branchiata is a transparent larva with numerous pigmented amoebocytes; yellow pigments are located at the base of the tentacles, around the preoral lobe and near the ciliated perianal ring. This larva, the largest known in phoronids, grows to an unusual size (about 2 mm in length), and the larval tentacle number increases to 42. Paired vacuolated stomach diverticula are present, as are two ventral blood masses just above the nephridial site, lateral to the stomach. The two masses originate at about the 20-tentacle stage and usually fuse just before metamorphosis. The adult tentacles arise as independent eversions under the bases of the larval ones until the larva has usually about 22 (Figs 9,16,20).

The larva which is ready to metamorphose from about the 24tentacle stage (Fig. 20) becomes opaque, although a protruding tip called the piriform organ appears on the preoral lobe (anteriorly to the apical plate, between the latter and the ventral free margin of the lobe). Herrmann (1976) suggested that the function of the piriform organ is t o select a suitable substratum for larval settlement and then to induce the processes of metamorphosis. According to Emig (1980)

4-T

8-T l4 -T

Metasomals

30-1

PI(:16

Home drvelopinental stages of Artiriofrochn brartchznta

the piriform organ seems to be related to larval ecological behaviour and has no evolutionary relationships within the Lophophorata or with related phyla. The larva can induce metamorphosis without the piriform organ being present (Fig. 9).The length of pelagic life can be prolonged by lack of food or other unfavourable conditions which may delay development in A . branchiata; the increase of the number of tentacles could then be explained by the lengthening of the pelagic life; the same statement seems to be true of the other actinotroch species, especially Actinotrocha sabatieri. All actinotrochs collected by Browne (1895, 1900) and studied by Selys-Longchamps (1907) belong to A . branchiata; the specimens named A . brownei are o f the same species just beginning their metamorphosis. Schepotieff ( 1906) described two forms which are both probably related to A . branchiata. Similarly the first larva identified by Steuer (1933)belongs to the latter species. The form B of the species established by Ikeda (1901) could be a synonym of A . branchiata and probably also the form D which seems to be an abnormal stage in metamorphosis.

6 . Larva of Phoronis pallida: Actinotrocha pallida

The adult form of Actinotrocha pallida, a larva known since Hchneider ( 1 862), has recently been described by Silkn ( I 952), under the name Phoronispallida. Other information on the larva is given by Selys-Longchamps ( 1 903, 1907), Silkn (1954a) and Zimmer (1964). A . pallida is small, opaque, yellowish-white, provided with a considerable amount of yellowish pigment located in the apices of the epidermal cells (no pigment in the apical plate); there are no piginentiferous amoebocytes (Figs 9, 17). The stomachal diverticulum is unpaired. There is only one blood mass in a paired

u

100pm

10 - T

Ffc. 17 Drvrlopluental rtagr of Actirrotrocltu pallzdn

aggregation united in the midline in the fore ventral part of the stomach. The larva ready to metamorphose exhibits a maximum number of ten tentacles, but sometimes two additional tentacles appear just before metamorphosis. At this stage, the metasomal sac occupies virtually the whole of the trunk coelom. It seems t h a t the extensive pigmentation and the highly colourful body distinguish A . pallidu from the other actinotroch species (see discussion in A . hippocrepia ) . 7 . Larva of Phoronopsis harmeri: Actinotrocha harmeri

A . harmeri is the larva of Phoronopsis harmeri: this larva has been described by Zimmer (1964) under the name Actinotrocha A and recent unpublished observations have confirmed this parental relationship. Zimmer (1978) established t h a t there was no difference between the larvae of Phoronopsis harmeri and Phoronopsis viridis;it must be remembered t h a t both species are considered as synonyms

(Marsden, 1959; Emig, 1971a, 1979) although Zimmer’s (1978) opinion has never been further supported, A . harmeri is large, transparent, without epidermal pigmentation; only concentrated yellow pigmented amoebocytes occur in characteristic locations: margin of the preoral lobe, tentacles, metasomal sac, collar ring muscle, oesophagus and perianal ciliated ring. There are two pairs of blood masses which are located as follows: one discshaped pair in the dorso-lateral corners of the preoral lobe and one pair elongate in the collar, ventro-laterally a t the site of the third tentacles (Figs 9, 18). The larva is ready t o metamorphose a t the 20tentacle stage without the presence of adult tentacles. Zimmer (1964) Blood mass

I

@ 6’

4-T

L 6-1

10-T

20-T

FIG 18 Main developm~ntalstages of Actiiiofrochn hornwrt (four-tentacle larva to 16 T aftel Zimmer, 1964)

suggested t h a t a piriform organ could be present shortly before metamorphosis, b u t i t does not possess the remarkable extensibility of t h a t organ in Actinotrocha branchiata. The species named Actinotrocha ikedai A by Selys-Longchamps (1907) has been studied by Ikeda (1901) who considered i t t o be the larva of P. ijimai. According t o its characteristics, this larva is mostly similar t o A . hurmeri: short and thick body; 1-1-5 mm long; about 16 tentacles; metasomal sac a t about the eight-tentacle stage and the two pairs of blood masses at the 14-tentacle stage, one pair of these masses covering the stomach diverticulum, the other pair ventrolaterally in front of the septum on both sides of the stomach.

8. Larva of an unknown adult: Actinotrocha wilsoni Under this name is described the “species B” of Wilson (1881).The adult form of Actinotrocha wilsoni is presently unknown, b u t it could

be suggested that the larva belongs to Phoronopsisalbomaculutn on the basis of the similarities with Actinotrocha harmeri. Selys-Longchamps (1907)and Forneris (1959)have studied the present form B which they considered to be a distinct larva. However, a synonymy with A . harmeri cannot be excluded. The pigmentation of the body is diffuse, not in amoebocytes, Pigment spots are located in the preoral lobe, on the inferior face of the larval tentacles and in the perianal ciliated ring. The stomach protrudes usually into paired diverticula, but this is not invariable. The piriform organ is present in front of the apical plate in larvae ready to metamorphose. A t the latter stage, the actinotroch shows up to 26 larval tentacles, and definitive ones independent of the larval tentacles. Four masses of blood corpuscles occur, two dorso-lateral at the level of the oesophagus and two ventro-lateral to the stomach '(Figs 9, 19).

FIG. 19. Some developmental stages of Actinotrochm udsoni (after Forneris, 1959). The 26tentacle larva is fully developed with 20 adult tentacles, ventral view.

The characteristics of Actinotrocha menoni X , given by SelysLongchamps (1907) based on a few specimens collected by Menon (1902), are similar to those of A . wilsoni: an oval body with a large preoral lobe; about 1.40 mm long; 44 tentacles; four blood masses: two lateral to the stomach just, above the septum and two dorsolateral in the fore-part of the stomach. The same suggestion is made for Actinotrocha bella whose description by Forneris (1959) is very similar to that of A . wilsoni, but the former larva as with A . menoni X could have delayed development as indicated by the high number of tentacles.

31

9. Other described actinotroch species Several actinotroch species are insufficiently characterized; most of them have been established by Selys-Longchamps ( 1907): Actinotrocha gegenbauri, A . sheareri, A . selysi, A . dubia, A . olgae, A . henseni, A . gardineri and A . goodrichi after description of Goodrich (1903), A . spauldingi after Spaulding (1906), A . haswelli A and B according to Haswell (1893),A . ikedai C after Ikeda (1901),and A . menoni A , B, C after the description by Menon (1902).Actinotrocha chata, recently discovered by Forneris (1959), is probably the tententacle stage of a known species. C. Larval settlement an,d metamorphosis When the actinotroch is mature and ready to undergo metamorphosis, the metasomal sac is completely developed; the larva becomes opaque and negatively phototactic (Cori, 1939; Silen, 1954a).On the latter point, however, Zimmer (1964) and Herrmann (1976) do not agree. The actinotroch sinks to the bottom; this behaviour seems to be induced by bacteria or chemical substances as metamorphosis proceeds (Forneris, 1959; Herrmann, 1976). Such behaviour is known from the literature in other zoological groups. During settlement, the preoral lobe becomes round and the pirifortn organ protrudes (sometimes sharply as in Actinotrocha branchintn)

. Perional ciliated ring

(a)

,

. . . .. ..

.

(b)

FIG.20. Aetinotrochu branehiata. (a) Larva ready to metamorphose, with adult tentacles and piriform organ; (b)settlement and beginning of the metamorphosis process (eversion of t'he metasomal sac into the soft sediment).

(Fig. 20) in species possessing such a structure; in other species the preoral lobe may be pointed, the anterior tip being the apical plate. The preoral lobe enters directly into contact with a suitable bottom and metamorphosis is invariably induced. The piriform organ or apical plate have a t this time the function of selecting a favourable substratum and probably of starting the process of metamorphosis. The cilia cease to beat and through violent muscular contraction the metasomal sac is suddenly fully evaginated, passing vertically downwards into the soft sediment where it rapidly secretes a tube (Figs 20b, 21); on hard substrata the animal begins to burrow into the bottom after the secretion of a thin hyaline tube (Silkn, 1954a). -.

z

-

-_____ -

--_

the metasome evagination in horizontal plane, with the tube and adult consequently adnate.

In view of the fundamental importance of the relationship between the adult phoronid and the substratum and its associated fauna, the actinotroch can, but probably with only a small chance of success, search for a suitable bottom, and metamorphosis then seems not to be delayed for long. When the settlement of larvae occurs within adult phoronid aggregations, which seem attractive to actinotrochs, the nearest-neighbour distances are not limiting in the settlement which occurs randomly (Fig. 22) (Ronan, 1978; personal observations on Phoronis hippocrepia, P. ijimai, P. australis, P . psammophila). Close n-n distances need a stratification of the lophophores to provide a fully tentacular expansion, which is especially observed in clumps of burrowing or encrusting forms. Such a disposition in suspension feeders always requires some water currents to bring food. The turbidity of the sea water is not a factor affecting the abundance of the phoronids. Figure 22 is compamble with the curves published by Ryland (1976, Fig. 35) on Bryozoa. Similar figures would probably be established for hard-substratum species. The nearest-neighbour

33

20

-

,

w.

c

10

t

U

0

N N distance mm

42. Frequtxncy distribution of the nearest-neighbour distanres in nine intrrtidal aggregat’ions of Phorortis h r n w r i (established after the data of Table I of Ronair. 1978).

E’lc:.

distances in Phoronis psammophila were, however. never less than the space required to expand two adjacent lophophores completely 1966), even in a high density of about 18000 individuals (Emig, m ’. According to Ollivier et al. (1977),Phoronopsis harmeri may avoid locations near large deposit feeders; similar observations were made by me with filter-feeders (Emig, 1966). On the other hand, the presence of phoronid aggregations prevents the settlement of larvae and adults of the associated fauna, particularly of tube-builders.

D. Metamorphosis In all published works, metamorphosis has been studied irnperfectly (see Roule, 1900; Ikeda, 1901; Selys-Longchamps, 1907; Cori, 1939; Veillet, 1941; S i l h , 1954a; Herrmann, 1976). The actinotroch passes from a highly adapted pelagic form to a slender benthic organism organized as a tubicolous adult in a very short period of time, about 5-30 min. The adult organization arises from larval structures and only certain of the larval structures break down. I n fact, metamorphosis is “catastrophic” in regard to the rapid formation of all adult structures which begin by the rotation through about 90” of the larval axis to assume a new adult axis (parallel with the polar axis of the egg) arising by the eversion of the metasomal sac (Figs 20, 23). This sac, which is thus the wall of the adult trunk, evaginates entirely, drawing down the digestive tract attached by the ventral mesentery in the adult position. The posterior end of the sac differentiates into the ampulla. During the

Perianal ciliated ring

Adult tentacle

Autolysed larval tissue

Lophophore

h

Anus

LateralI vessel Median vessel

I-

Incipient blood plexu

FIG.23. Main metamorphosis stages in Actinotrocha branchiata (see also Fig. 20) from the evagination of the metasomal sac to a juvenile Phoronis muelleri. (a) About 1 min after settlement; (b) about 3 min; (c) about 8 min; (d) about 1 day after the beginning of the metamorphosis.

evagination mouth and anus are brought into close proximity. The preoral lobe and the larval tentacles shrink and are mostly cast off a:id ingested (they represent the first food intake of the adult). The definitive adult tentacles are elevated around the mouth in the functional position for food gathering. The circulatory system also becomes functional. Internal and some external changes are briefly considered below and particular attention is given to the transformation from the larval to the adult status of the main organs (Fig. 23). The processes of metamorphosis retain the archimeric disposition of the larval body, but the borders of the three body regions of the adult and their coelomic cavities will have other relationships owing to the axis rotation, while the dorsal body side is largely reduced (Figs 24, 25, Table 111). Such dispositions have been largely discussed by Emig (1973a, 1976a, b, 1977b). The differentiation of the epistome has been the subject of controversy; most previous investigators stated that the epistome does not arise from the larval lobe, which is refuted by Wilson (1881), Caldwell (1882), Schultz (1903), Meek (1917) and Zimmer (1964). The recent studies of Siewing (1974) and Zimmer (1978) confirm the opinion of Roule (1896) that the preoral

35

THE BIO1,OOY OF PHOKONIDA Preorol lobe

Piriform oraan Adult tentacles

Mou

PIC:.24. Diagram of the differentiation of the epistome during the metamorphosis of Actinotrocha branch,iata (after some photographs and description of Sipwing, 1974 and of Zimmer, 1978).Successive stages from the preoral fold (incipient epistome) at the first stage of the metamorphosis t o the epistome of an individual in which the rasting off of the autolieed larval tissue is complete.

Gprotocoelom /Anus

?Digestive

tube

/Metacoelom

Anus

/

\

-

Ventral

FIG.25. Diagram of the archimeric structure of a n actinotroch just before metamorphosis and of it phoronid, showing the disposition of the coelomic cavities and the ventral and dorsal sides.

lobe shrinks and is cast off, but a small bleb issuing from the internal part of the vestibule is retained as a remnant of the lobe (Figs 23,24). This fold, partly containing the protocoel, bends dorsally t o fuse with the lophophore and trunk epidermis and soon differentiates (Fig. 24) into the adult epistome. The delimitation of the epistome and of its coelomic cavity was established by Emig and Siewing (1975). The larval nervous ganglion, and the piriform organ if present, are not retained; the adult ganglion appears later in the dorsal wall of the epistome (Fig. 25) as a thickening of the adult nerve ring which probably originates from the larval collar ring nerve (Emig, 1976a). The larval tentacles or their distal portions are swallowed and ingested. The adult tentacles or the basal buds are elevated around the mouth in the lophophore; the new tentacles arise then on the dorsal side between mouth and anus. The mesocoel is horseshoeshaped with an enlargement in each tentacular bud. The trunk septum has now the status of a mesentery and becomes the adult diaphragm which separates the pro- and meso-coelom from the metacoelom (Table 111).

Phoronid

A ctinotroch 1. Prosorne

Preoral lobe Protocoel Preoral septum 2. Mesosome Collar Mesocoel Blastocoelic collar space Preseptal cavity Trunk septum 3. Metasome Trunk Metacoel

+

Epistome Protocoeloin

+

Mesome (or lophophore) Mesocoelom Lophophororal blood vessel

+

+

+

Diaphragm Metasome (or trunk) Metacoelotn

+

By strong muscular contractions (Fig. 20b) the larval trunk evaginates the metasomal sac which is then the wall of the adult trunk provided with all layers (epidermis, basiepithelial nervous plexus, basal lamina, circular and longitudinal muscle layers, peritoneum). The posterior part of the adult trunk becomes rapidly an enlargement or ampulla (Fig. 23). The metacoel is retained and now named metacoelom separated distally from the pro- and mesocoelom by the diaphragm, a complete mesentery derived from the

larval trunk septum. The trunk contains the largest coelomic cavity with the most internal adult organs. During the evagination of the metasomal sac, the whole larval digestive tract, which is attached by the single ventral mesentery, moves downwards to take the adult position, and a t the same time mouth and anus are brought into close proximity, whilst the larval walls of the collar (except definitive tentacles) and of the trunk shrink and disintegrate little by little around both openings (Fig. 23). The digestive tract is now divided into a descending branch with successively an oesophagus, and an elongate stomach which differentiates gradually into a prestomach and a stomach, and a slender ascending branch represented by the intestine. As observed by Herrmann (1976), the larval stomach diverticula degenerate in the prestomach epithelium (Fig. 24b, c). The ventral mesentery of the actinotroch becomes the oral mesentery and the anal one in the adult form, connecting the trunk wall with the U-shaped digestive tract. No description of the differentiation of the median and lateral mesenteries is given. Their ontogenesis may be compared with t h e regeneration process (Emig, 1972b, c, 1973a). The protonephridia with solenocytes are transformed into metanephridia in the adult, classified as mixonephridia by Goodrich (1945). Such a transformation needs much further attention (Emig, 1973a). A t metamorphosis, the solenocytic cells fall into the blastocoelic space and the two larval nephridal tubes narrow the anus on either side in a dorso-lateral position. Most previous investigators stated that the larval ducts are retained and the internal coelomic funnels are secondarily acquired probably from mesodermal cells. Nevertheless, Ikeda (1901), Cowles (1904b), Brooks and Cowles (1905) and Cori (1937) suggested that the larval ducts degenerate totally or partly. It is interesting to note that during regeneration the nephridia originate entirely from mesodermal cells. The differentiation of the nervous system is largely unknown. According t o Silth, (1954b) no larval nervous structures remain in the adult. SelysLongchamps (1907)stated that the giant nerve fibre is differentiated in the metasomal sac in the late actinotroch stage. Finally, the circulatory system becomes functional. The lophophoral vessel is produced from the reduction of the blastocoelic collar space, as described first by Wilson (1881) and since generally confirmed. Both afferent and efferent vessels that together form t h e lophophoral vessel differentiate about 12 h later and send capillaries into the tentacles. The blood masses break rapidly apart in t h e blastocoelic space and the erythrocytes are distributed throughout

the whole system. The median vessel (or dorsal vessel of the actinotroch) develops rapidly into a large vessel which unites the incipient stomach blood plexus to the lophophoral vessel, whilst t h e second longitudinal vessel, the lateral, originates as a splanchnic slit along the left side of the descending branch of the digestive tract (Fig. 23). The latter vessel unites the lophophoral vessel to the stomach plexus. Now, the circulatory system is of the closed type and the train of peristaltic waves begins. The differentiation of this system obviously needs to be studied in the different actinotroch species, and the regeneration processes compressed (Emig, 1973d). Then, it is probable that more information will be obtained regarding the evolution of the circulatory system. Selys-Longchamps (1907) and Emig (1973d) considered the two lateral vessels to be primitive. The right branch of the lateral vessel and the second lateral vessel on the right side of the stomach blood plexus are considered t o be remnants of the second lateral vessel. I n Phoronis ovalis the two lateral vessels are represented in the metasome (Emig, 1969). It seems that the divergent descriptions of previous authors on some processes of the metamorphosis can be explained by the probable occurrence of abnormal phenomenona during experimental metamorphosis or development under the microscope.

V. ECOLOGY All phoronids are tubicolous and free living within their tubes. They may be found singly or in masses of many individuals, embedded vertically in soft bottoms or buried in, or encrusting on, hard substrata, including the special position of Phoronis australis within tubes of cerianthids. The best investigations on ecology and sampling are obtained by means of Scuba diving, especially on hard biotopes, and by the use of suction samplers (Emig, 1971a, 1977d; Emig and Lienhart, 1966, 1971). A. Tube The tube of Phoronida is secreted by epidermal gland-cells, recently studied by Pourreau ( 1979) in Phoronis psammophila. Two cell types actively secrete the tube (Fig. 26). The acidophilic A cells secrete mucopolysaccharides which constitute the two thinner but compact peripheral layers whose thickness varies little compared to that of the central basophilic layer. This is secreted by the basophilic

B cells which produce sulphomucopolysaccharides; its thickness varies considerably along the tube and consists of numerous very thin parallel coats. A third less frequent type of cell, the B’ cell, also

FIG. 26. (a) Uistrihution of the epidermal gland cells along the hody wall of I’horoi~ia

paammophila. A: acidophilic cells-muropolysaccharides; B: hasophilic cells-arid 1nuc.opolysaccharides; C : C cells-acidophilic nature (from Pourreau. 1979).Recent observations on Phoronis hippocrepia reveal the presence of A and B cell types in the major length of the lophophore. (a’)Relative abundance of the epidermal gland cell types (in yo)along the hody wall of F . psammophila (from C. Pourreau, unpublished results); (h)some aspect ofthe tube laying in relation to the thickness of the tube. A : acidophilic coating layer; B: basophili~. coating layer (from Pourreau, 1979). The main thirkness of the tuhe is about 10pm.

basophilic in nature and secreting protein, probably participates in tube formation. On the external surface of the tube, substratum elements adhere, particularly in soft sediments (various grains, debris, detritus), which cover the whole tube in a single layer. The above description of those epidermal gland cells does not agree with the descriptions given by several previous workers (SelysLongchamps, 1907; Marcus, 1949; Lonoy, 1954; Forneris, 1959),but confirms the incomplete short work of Hyman (1958) who identified only a positive reaction on the outer layers. A fourth type (C cells), named “corps en massue” by Selys-Longchamps (1907),occurs in the anterior part of the trunk, especially just below the lophophore (Fig. 26). According to Pourreau (1979), their function is probably lubrication to permit rapid motion in and out of the tube. The distribution of all epidermal cell types is represented on a diagram (Fig. 26). In the anterior trunk where all A, B and C cells are abundant, active secretion allows repair of the upper part of the tube (about 2-3 cm),which is easily damaged during bad weather by water movement. The posterior half of the ampulla has mostly a mechanical function either in substratum burrowing or in anchorage of the animal within its tube, which explains the reduced density of gland-cells. Tube secretion cannot be restricted to any particular area of the body, all parts of which are involved in the tube building process. Observations on tube secretion in the aquarium revealed that a fragile and transparent membraneous coat generally surrounds first the base of the ampulla, then it becomes visible anterior to the ampulla and particularly in the lophophoral region. The ampulla remains uncovered for a longer period of time but is finally also covered by the coat. The phoronid moves freely within its newly formed tube. The substratum particles adhere to the viscous tube. The length of the tube varies according to the extended size of the body (about five times longer than the contracted size) of the phoronid species and to the type of substratum. The smallest occurs on hard substrata, particularly in shells, and the largest on sandy bottoms. Such differences can also be observed within the species according to the burrowing potential of the substratum, The greatest recorded length and diameter of the tube in the different species may be summarized as follows: P . ovalis: 3 cm, 0-4mm; P . hippocrepia: 8cm, 2mm; P. ijimai: 10cm, 2mm; P. australis: 18cm, 5mm; P . muelleri: 12 cm, 1.5 mm; P . psammophila: 19 cm, 2 mm; P . pallida: 14 cm, 1.5 mm; Phoronopsis albomaculata: 15 cm, 2 mm; Phoronopsis h r m e r i : 22 cm, 4 mm; Phoronopsis californica: 45 cm, 5 mm.

I n soft sediments the phoronids burrow downwards posterior end first; active burying is accomplished by the ampulla initiated by hydrostatic pressure changes of the coelomic fluid and aided by contractions of the metasomal muscles (Pig. 27). After about only 8-

FJG.27. Hydraulic pressure changes in the ampulla of Ph.orortis p.scimmwph,ila during the, burrowing process (from ('. Pourreau, unpublished observations). + Swelling movennent; A motion of hydraulic fluid; A muscle rontraction.

24 h the lophophore remains out of the substrate. On the other hand, the animal moves toward the sediment surface by a similar action in the anterior part of the body (lophophore).During both movements the tube is concomitantly being secreted, and during this sticky phase various particles of the immediate environment adhere to the tube, which therefore reflects the nature of the substratum. The coating of particles can vary from fine mud to coarse sand grains, with sometimes numerous fragments, e.g. of shells, algae, sponge spicules, test of Foraminifera, or urchins, coral, etc., and fine pebbles. There was a great amount of controversy among previous authors concerning the selection of substratum particles for the tube. Some authors argued that the phoronid selects the particles, i.e. Andrews (1890) who named P . architecta (synonym of P. psammophila) on a possible grain selection. Other authors such as Selys-Longchamps (19071, Hyman (1959), Emig (1971), say, on the contrary, t h a t the particles adhere randomly to the sticky tube without specific sorting of grains by the animal. This latter statement has been confirmed by Pourreau (unpublished data) on P. psammophila. No significant differences occur between the two grain size curves of the tube and sediment composition, which confirms the observations of Emig (1971a, 197313).In all species embedded in soft sediment the posterior end of the tube always shows a small opening (Fig. 28a). We have no information on boring species, but i t is likely that a similar method is used, namely by ampulla movements probably assisted by chemical action on the substratum. Each individual lives

42

Lophophore

Sediment particles adhering fothe tube

.

'

Cerianthid tube

*

. . . . -- . . .. ... . ,,: _ . .. (a

1

Flc. 28. Diagrams of the position of Phoronida in the substrata. (a) Embedded vertically in soft sediment; (b) burrowing in hard substrate; (c) Phoronis australis living in the tube wall of cerianthids (after Emig et al., 1972).

in its own burrow which is lined by a yellowish membrane that sometimes projects and is then covered with debris (Fig. 28b). The perforated hard substrata are of different kinds in Phoronis ovalis, P . hippocrepia and P . ijimai. They generally live in empty mollusc shells, sometimes in living ones, but also in barnacles, Caryophyllia, Lithothumnim, coral rubble and rocks (lime- sand- and calcarous stones), and logs (see Emig, 197313). The burrows in shells are generally parallel with the shell layers between the periostracum and the pearl layer; distally the tube becomes perpendicular to the shell surface and the lophophore protrudes. The posterior end of the tube is thus in a cul-de-sac. P . hippocrepia and P . ijimai can also be encrusting, forming turf-like masses composed of many intertwined individuals. Such clumps adhere to pilings, logs, ledges, soft

sediments, rocks, often in crevices, in rather quiet water. The tubes are covered by particles (mud, sand grains, faeces, debris, algae) and within the entangled tubes sometimes occur molluscs, ascidians, polychaetes, ophiuroids, actinians or other animals. Burrowing of encrusting forms may depend upon the nature of the environment (hydrodynamics, desiccation, presence of other boring animals). Brazilian specimens of Phoronis ovalis sometimes show “cuticular processes” around the posterior end of the tube about 0.50 mm long (Marcus, 1949; Lonoy, 1954; Forneris, 1959; Voigt, 1975). Such extensions are not explained, but are not related to the presence of Cliona (cf. Voigt, 1975), as supposed by Marcus (1949). Phoronis australis, in its own tube, occurs in the tube-wall of cerianthids, which must be considered as a “hard” substratum, and thus displays a unique association. I n Madagascar the tube-wall of Cerianthus maua may be divided into five distinct layers (Fig. 28c): the ampulla of P . australis is always located in the fourth, the hardest one; the phoronid tube passes across the other three layers and the lophophore projects externally (Emig et at., 1972). B. Biotopes The main results of my review on the ecology of Phoronida (Emig, 1973b) are summarized here with the addition of more recent work (cited below) and of recent unpublished observations (particularly on the west coast of Panama, vicinity of Marseille and in the Indian Ocean). All generally confirm the previous studies. The vertical zones whose nomenclature is used in the present review are given according to PBrhs (1967).

Phoronis Ovalis. The density is great owing to the small body size of this species; it reaches 150 individuals/cm2. The bathymetric

distribution is a t present from low tide mark to about 50 m. Phoronis ovalis lives usually in coraligenous and detritic communities between 20 and 50m (Fig. 29b). Recently specimens have been recorded in shallow waters by Stancyck et al. (1976),Scelzo (1976) and Emig and Bailey-Brock, (1980), and in greater depths by Emig (1977~).

Phoronis hippmrepia. Like the former species, P . hippocrepia has a patchy distribution which can reach about 57 000 individuals/m2, from the intertidal zone to 55 m. Individuals observed between 0 and 10 m live preferably in turbid or shady conditions, in harbours (Emig and Bailey-Brock, 1980),and under overhangs and in cave entrances (Fig. 29b). I n similar conditions, Ocharan (1978) recorded P.

hippocrepia (in burying form) in two locations (Baiiagues and El Puntal) on the northern Atlantic Spanish coast. The mean density is about 91 000 individuals/m2, but it reached 5.7 individuals/cm2 over 13cm’. This species occurs in shallow depths under rather strong water motion. A list of the associated fauna and flora is given by this author. Phoronis ijimai. This species is usually known in the encrusting form living in similar conditions to the former species. The density in encrusting clumps can be greater than where i t occurs as a boring form. The bathymetric distribution ranges from low tide mark to about 10 m. Emig (1977c),on the basis of recent observations in Ban Juan Island (U.S.A.), confirmed the synonymy of Phoronis vancouverensis with P. ijimai. Recently, this species has been located by Haderlie and Donat (1978)on piles on the east side of the wharf in Monterey Harbor (California) where P . ijimai covers a large area of the pile surface between 0.50 and 7 m deep; a list of the associated fauna is given by these authors. Phoronis australis. The burrowing habit of this species is characteristic. I n general i t occurs in groups of 20-50 individuals and sometimes up to 100 per tube in the tube-wall of cerianthids (Coelenterata, Anthozoa). The recorded depth range is from the low tide mark to more than 36m (Emig, 1977c; Emig et al., 1977). The known identification of the cerianthid species are Pachycerianthus Jimbriatus by McMurrich (1910), Cerianthus maua by Emig et al. (1972),C. membranaceus by Emig (1977c), C .Jiliformis by Ishikawa (1977);all other investigators indicated Cerianthus sp. Phoronis muelleri. This species is characteristic of muddy bottoms, with a sandy, sometimes a coarse, fraction, over a large recorded depth range, from 1 to 390 m (Fig. 29). Anadon and Anadon (1973) discovered a specimen in a depth of0.60 m, whilst Holthe (1973,1977) described P . muelleri as abundant all over the Borgenfjorden, from 10 to 25 m, and in Trondheimsfjorden usually from 10 to 50 m (mostly 2@25 m ) with some locations at 100 m and one at 208 m. Thomassin and Emig (1980) sampled P . muelleri during the “BenthediExpedition” from 5 to 390m, most individuals being collected between 16 and 30 m. These records confirm the common range of P . muelleri as 10-50m, usually 15-30m (Fig. 29b). The density can reach 3000 individuals/m2; this number is recorded by Barnes and Coughlan (1972) in “mud on clay” (probably related to a Macoma-

community) where P . muelleri is regarded as the indicator of this type of bottom. According to Buchanan et al. (1978),Phoronismuelleri lives in a silty area along the Northumberland coast from about 20 to 85 m: where an Amphiura$liformis-A. chiajei community occurs mainly on mud ( A . chiajei subcommunity) and muddy sand ( A . jiliformis subcommunity) (Buchanan, 1963). Phoronis muelleri is ubiquitous and a top ranked species over the entire area. The seasonal recruitment reaches a maximum between September and November, and the greatest mortality follows between November and January.

Phoronis psammophila. It occurs generally in shallower waters, from low tide mark to 8 m , with extension to 20m and, as recently reported, to 25m (Emig et al., 1977). The greatest density of this species is 18 000 individuals/m2, reached in fine well-sorted sand at 4-6 m depth, although occurrences are also mentioned from mud to shelly coarse sand with a fine sand fraction, as well as in Zostera or Cymodocea sea-meadows, and polychaete reefs. I n the Mediterranean benthic populations, P. psummophila live mostly in the biocoenoses of fine well-sorted sand and of superficial muddy sands in sheltered areas which show several facies as sea-meadows (PArGs, 1967). In the other areas i t occurs in Venus-Abra alba and in Macoma baltica communities, in the Tellina boreal lusitanic community intertidally in fine sand at densities of 6(f150individuals/m2 (Vieitez, 1977), and in several types of sandy bottom (Thomassin and Emig, 1980; Emig and Bailey-Brock, 1980). The distribution of this species often overlaps some other phoronid species, particularly P . muellezi, P . pallida and Phoronopsis albomaculata (Fig. 29a). The presence of P . muelleri in the type locality of Phoronis architecta ( = P. psammophila, synonymy established by Emig, 1972d) is a confirmation of the mixing of these two species under a single name “ P .architecta :’by Brooks and Cowles (1905). This confusion led Ernig ( 1 9 7 7 ~ )to attribute to “ P . architecta” some characteristics of P. muelleri, especially in reproductive strategy. In areas of high density (mean about 15 000 individuals/m2), the spatial patterns are a uniform general distribution, but as the individual number per square metre decreases a patchy distribution appears perhaps caused by asexual reproduction, actinotroch settlement, presence of suspension feeders, etc. The expanded lophophores can cover more than the half of the sediment surface. The nearest-neighbour distances always remain greater than the space required to provide for full expansion of adjacent lophophores; thus, an overlying of lophophores has never been observed.

46

C' EMIG

Phoronis pallida. This species occurs in fine sand to clayed sand, in shallower waters (1-14m). The greatest density is 74000 individuals/m2. Emig et al. (1977) recorded P. pallida mainly in fine sand from 2 to 25 m with the highest abundance near 15 m in Port Philip Bay, and from 3 to 8 m in the other Australian waters (Fig. 29). Thomassin and Emig (1980) described P. pa,lZida living exclusively in fine sand from 0 to 13m with the maximal density (about 20 individuals/m2) between 6 and 13 m. Phoronopsis albomaculata. The general occurrence is in soft sediments from sandy mud to clogged coarse sand with a fine fraction, from subtidal areas to about 55 m (Fig. 29). The only cited density gives 37-70 individuals/m2 (Thomassin and Emig, 1980).

c

Fine

Sediment

Coarse

>

L

0)

c +

E

u ._ 0 C

!

rnuefferi

/

P 1

0

V

Fu;. 29. (a) Diagram of a possible relationship between the different species living in soft substrate and the sediment particles and the organic matter based on our prevent knowledge of the distribution of Phoronida (modified, after Thomassin and Emig, 1980);(11) distribution of the phoronid species in relation t o the depth (after Emig. 1979, completed).

Around Madagascar (Thomassin and Emig, 1980), Phoronopsis albomaculata is generally recorded in coarse sand communities under bottom currents, characterized by Asymetron (or sometimes Branchiostma); but in the Brisbane River and in Port Philip Bay

THE: RIO1,OBY OF PHORONIUA

47

(Australia) (Emig et al., 1977) this species has been collected in silty and muddy sand from 2 to 25 m, abundant in locations from here to 15 m. Like Phoronis pallida, the present species is often mixed with Phoronis muelleri and P . psammophila which is considered to be an overlapped distribution. However in Port Philip Bay there is great resemblance between the distribution of P. psammophila and that of Phoronopsis albomaculata. More investigations are necessary on the preferrred biotopes of the latter species (Fig. 29a).

Phoronopsis harmeri lives in soft sediments, sand to mud, sometimes with a coarse fraction, and has recently been found in Zostera beds (Emig et al., 1977),from the intertidal zone to 89 m deep, with a common range from 0 to 12m (Fig. 29). Depths are cited by Emig ( 19774 from different geographical areas. Phoronis harmeri is locally abundant, up to 28 000 individuals/m2 particularly in fine sandy bottoms. The highest density in a P . harmeri population, about 68 OOCk93 000 individuals/m2, was established by Ollivier et al. (1977, Fig. 79) in spring 1972, but the population suffered high mortality by heavy predation (60000adults/m2 by June). In some Californian areas, Ph>oronopsisharmeri shows a uniform distribution over the littoral region of the flats, slightly less dense on the upper and lower edges of the intertidal zone where the sediment is coarser. The average annual population of P. harmeri probably remains rather constant. But, according to Ronan (1978), the population of P . harmeri shows density variations of 0-22 000 individuals/m2 along the intertidal zone. The phoronids are aggregated in discrete clusters which are largest in area and individual density in sand with particles less than 250pm, and rare or absent in greater sediment particles. The mean distance between two individuals in fine sediment is 5.4 mm with a range of 1-25 mm (Fig. 22) and the most common distances are between 3 and 7 mm (frequency about 74%): the mean distance is roughly half the diameter of the expanded lophophore (Ronan, 1978). Phoronopsis californica. It occurs from mud to coarse sands, from 0 to 17m in depth (Fig. 29). C. Ecological effects In general, temperature and salinity are not limiting factors since the range normally encountered in the distribution of Phoronida is such that they can be regarded as eurythermal and euryhaline

animals. Nevertheless, some species are only recorded from tropical and subtropical regions. Phoronids are able to live in water with small amounts of oxygen. During summer 1971, Simon and Dauer (1977) studied a massive outbreak of red tide and its results on a sandy intertidal habitat in Tampa Bay (Florida); they pointed out that Phoronis psammophila appeared to be completely unaffected b the red tide; before the occurrence the mean density was 3/m , 1 month afterwards l/mz and 2 years later 23/m2. The effects of the tropical storm “Agnes” led to a delayed decrease in abundance of Phoronis psammophila, living in shallow sand bottoms of Chesapeake Bay. This was probably a response to the effect of lowered salinity (below loxo)for over a week (Boesch et al., 1976). Since the hurricane passed during the reproductive period of the phoronids this may have been a recruitment failure. The population had recovered a year after the spawning period (spring to autumn), see Fig. 30a which is to be compared with the curve published by Virnstein (1979), Fig. 30b, and with my own observations on gonad maturations (Fig. 3). All phoronids are suspension feeders which demand water movement. If the currents are strong their distribution is limited, especially in or below the intertidal zone (Emig, 1971a). The spatial

f

N

E L

e n L

n e

1 0 Q i . - d 0 ’.

f

1972 N

0 c

O

I

(a1

I

M N I F

1973

E,

“ 7 :

N

E

ln 0

0

0 L

n e J

S 1973

N

1

J

M

M

J

S

N

1974

FI~:. 30. (a)Change in the mean density of Phoronisp.yarnrnophilashowing significant, changes a t the 3 m deep sand bottom stations in the lower York River, prior to and after the passage of the tropical storm Agnes (21 June 1972) (after Roesch et al., 1976);(b) monthly abundance patterns of Phoronis pmmrnophila in natural uncaged sediments (after Virnstein, 1979).

distribution can be modified by different factors: predation, associated fauna (filter-feeders; burrowing and digging animals) and asexual reproduction which occurs in all phoronids. The occurrence of the species of Phoronida in their preferred substrata and benthic communities has been described above. The depth range, the vertical distribution and their possible relationships between the different species are summarized in Fig. 29. The presence of numerous tubicolous phoronids in soft sediments reduces erosion of the substratum and thus stabilizes the sediment and its infauna. The burrowing species have a destructive action on shells; in living animals, as in Caryophyllia smithi (cf. Hiscock and Howlett, 1975), the phoronids may cause death by structural damage.

D. Predators of Phoronida The predators of Phoronida are not well known, yet phoronids do form a significant proportion of the diet in more animals than is generally supposed. They may provide an abundant food supply in areas of high density. Emig (197313) estimated that the biomass of Phoronis psammophila was about 45 g/mZ wet-weight (15 g dry weight) for 15000 individuals, the anterior part of the body being about 3 g and 1 g (wet and dry weights respectively) for this number of animals. The defence against many predators is a very rapid retraction into the inner part of the tube where the animal is anchored by means of its ampulla.However, the response to a slight disturbance (Emig, 1966) is only a folding of the lophophore near the tube opening. The anterior part of the body with the lophophore (which protrudes over the surface, sometimes up to 3cm, particularly in burrowing or encrusting species) is the most vulnerable to attack by predators. Phoronids can, however, rapidly regenerate the lost part in 2 or 3 days (Emig, 1972c; unpublished observations on Phoronis ijimai and Phoronopsis harmeri). The tube should offer some protection from predators (Virnstein, 1979). Phoronids occur in the guts of fishes which are able to remove an animal from its tube. Other observations seem to indicate predation on small animals by nematodes. Of predators on phoronids, the gastropods are the best known. Attacks by the nudibranch Hermissenda crassicornis removed only a few of the tentacles but sometimes the entire lophophore (Ronan, 1978). Ollivier et al. (1977) indicated high mortality rates in Phoronopsis harmeri populations

50

('. (1. EMIG

which decrease by about 60% after June, probably due to predation by Hermissenda crassicornis which is then very abundant: the adult phoronid abundance patterns may be strongly influenced by predators. According to Virnstein (1979), the abundance of Phoronis psammophila is apparently not significantly regulated by predators such as crabs or fishes. The opposite observations of those authors suggest that the predation on phoronids obviously needs further study, particularly that by fish and gastropods, probably the most important consumers. My opinion is supported by my own incomplete observations on predation of P. psammophila in a high density area. E . Geographical distribution The geographical distribution of the Phoronida is well known in several substantial areas of all oceans and seas (Fig. 31), and our knowledge has increased recently as benthic studies have developed. Since the exhaustive review of Emig (197313, completed in 1977c), new localities have been cited by Barnes and Coughlan (1972),Voigt (1975), Holthe (1973, 1977),Scelzo (1976),Vieitez (1977), Emig et a2. (1977),Vieitez and Emig (1979), and Emig and Bailey-Brock (1980), Thomassin and Emig (1980); and recently new locations have been recorded from China (Tsingtoa: P. ijimai, P. australis); the latter species occurs also on the west coast of Australia (off Geraldton) and in Florida (St Lucie); from Panama (especially Naos, Culebra, Perlos Islands: Phoronis muelleri, P. psammophila, P . hippocrepia, Phoronopsis albomaculata and P . harmeri) ; from Mexico (Vera-Cruz: P . hippoerepia) and this species occurs also near Marseille (Gulf of Fos; Calanque of Morgiou); from the U.S.A.(Humboldt Bay, Morro bay, Santa Barbara: Phoronopsis harmeri; Port Aransas: P . muelleri). Phoronis psammophila also occurs in the Bahamas, Bermuda and in France (Cortiou near Marseille; Gulf of Morbihan). In general, phoronids have a world-wide distribution as indicated in Fig. 31, and most of the species can be considered as cosmopolitan, particularly P. muelleri, P. hippocrepia and P. psammophila. Some species, such as P . australis, Phoronopsis albomaculata and P . californica, seem to be restricted to the tropics. VI. FOSSIL PHORONIDA Several authors suggested that tubes or tubicolous burrows in fossil records belong to Phoronida (Fenton and Fenton, 1934;

* Phoronis sp

Flu. 31. Geographical distribution of Phoronida (after Emig, 1979. completed)

Avnimelech, 1955; Josey, 1959; Voigt, 1972; Mackinon and Biernnt, 1970). Recently, Voigt (1975)has proved the identity of species of the genus Phoronis with the ichnogenus Talpina v. Hagenow, 1840, confirmed in 1978. The Talpina burrowed in such diverse calcareous substrata as calcareous algae, echinids, mollusc shells and rostra of Belemnites (Fig. 32). The fossil phoronid burrows seem to have been present since Devonian times. Voigt gives criteria used for the discrimination of the phoronid burrows from other similar ones such as those of Thallophytes, sponges, Bryozoa or “worms”. The frequent presence of agglutinating Foraminifera surrounding the opening of

FIG.32. (a) Talpinu gruberi Mayer, in diagenetically destroyed aragonitic layer of shells from Muschelkalk (Trias) ( x 5 5 ) (photo. Prof, Dr E. Voigt, reproduced with his permission):( h ) Talpinu rumom v. Hagenow, in Belemnellu lunceolatu (Maastrichtian Luneburg) ( x 5.2 ) (from E. Voigt, 1972, Table 3, Fig. 1).

THE BIOLOGY OF PHORONI1)A

53

the tube of a worm-like fossil animal provisionally determined as Phoronopsis and suggesting commensalism between both fossil organisms (Voigt, 1970) in Upper Maastrichtian, has never been confirmed in recent observations on Phoronida. Tubes of the ichnogenus Talpina ramoaa which occur frequently within the guards of Belernnella and Belemnitella (cf. Voigt, 1972) are described within the cavities probably originating from diagenetically destroyed aragonitic corals of the Maastrichtian chalk-tuff (Voigt, 1978). VII. FEEDING Phoronida are suspension feeders capturing particulate matter, detritus or small organisms from the water by means of the lophophore, but it is not yet known if they show feeding preferences. However, food is also available in the form of dissolved organic matter. It is obvious that our knowledge is fragmentary and that the feeding biology of Phoronida needs research in a variety of ways.

A. Lophophore and epistome The definition of a lophophore given by Hyman (1959) and recently completed by Emig ( 1 9 7 6 ~is ) as follows: “A lophophore is defined as a tentaculated extension of the mesosome (and its cavity the mesocoelom) that embraces the mouth, but not the anus, and its main functions are feeding, respiratory and protective”. * The latter author pointed out that the term lophophore is limited to the phylum Lophophorata (Brachiopoda, Bryozoa, Phoronida: Emig, 1977a), and cannot be replaced by any other terms. The general form of the lophophore of Phoronida is well known (Emig, 1971a, 1976c, 1979). Briefly, it is bilaterally symmetrical, supported by a collagenous “skeleton” which is an enlargement of the basal lamina. The lophophore shape assumes a greater complexity with increase of the tentacle number in relation to the general size of the species (Table IV; Figs 1, 5 , 7), which suggests a dimensional relation between food-gathering, respiratory capacities and metabolic requirements. The arrangement of the cilia of the tentacles is shown in Fig. 33. Mucous gland cells (B cells: see Section V, A ) occur in the laterofrontal surfaces of the tentacles. There is no evidence that mucus * I n the present section only the feeding function will be described whilst the respiratory one will he considered in the following section on the circulatory system.

Lophophore shape

Maximal number of tentacles

Phoronis ovalis

oval

28

P . hippocrepia

Horseshoe

190

Species

1’. mu.elleri P . psammophila P. pallida Phoronopsis albomaculata Phoronis ijimai Phoronopsis harmeri Phoronis australis Phoronopsis californica

Horseshoe Horseshoe Horseshoe -Or sometimes slightly coiled Up to 2 coils 3.5 coils Helicoidal u p t o 5 coils

100

130 140 160

230 400 1600 1500

plays any role in food capture (Bullivant, 1968b; Strathmann, 1973; Emig, 1976c), but it is associated in the particle rejection mechanism when the particles are bound into strands by mucus. According to the observations of Emig and RBchBrini (1970),Emig ( 1 9 7 6 ~and ) Gilmour (1978), it could be suggested that the feeding position of the lophophore is somewhat different in species with a horseshoe-shaped lophophore than in those with a coiled one, as shown in Fig. 33, and the water currents seem to be deflected in different ways. Nevertheless, the adult phoronids always orient their lophophores into the prevailing water current, and when currents change direction, the phoronids can rapidly re-orientate to maintain the food-catching surface of the lophophore in the water flow. The mouth, located at the bottom of the lophophoral cavity, is covered on the dorsal side by a lip extended along the inner tentacle row, with its own coelomic cavity the protocoelom. This lip is called the epistome (or protosome, first of the three body regions; Figs 1, 25). The anal side of the epistome is sparsely ciliated whereas the oral side shows a dense ciliation (Fig. 33b). The epistome is involved in feeding (Selys-Longchamps, 1907; Pross, 1978). Gilmour (1978) supporting Masterman (1896,1900)speculates that the function of the epistome is “involved in the rejection of inedible material and acceptance of food particles during suspension feeding”, which is similar to the function of the labial palps in bivalvesand the gill slits of chordates (Gilmour, 1979).

Lophophore

Inner tentacle

Mouth

Oesophagus

1.3mm Preoral lobe Nervous layer

Peritoieum

‘Perianol ciliated ring

,Frontal

cilia

Muscles

(91

( f )

Diagrams of the suspension feeding in Phoronida. (a) Feeding position of I’horouis psummophila showing the water currents according to the studies of Emig and HGchGrini (1970; Pig. 7) and Emig (1976c, Fig. 4b); (b) section of the lophophore of I’horcmis psummophilu with feeding currents past tentacles: (a)capture of food particles on frontal surface, (b) rejection of inedible particles through the lophophore: the diagram represent)s also the tentacle flicking; (c)water flow in the lophophore of Phormis ij’imai(from Gilniour, 1978) (a), (b), see above); (d) feeding position of Phoronqpsis hurmeri with water currents (after Gilmour, 1978, modified); ( e )longitudinal section of the lophophore of I’horonop~~s hurmeri, with the particle motion (from Gilmour, 1978) (a),( b ) ,see above); (f) longit,udinitl section of an actinotroch with currents (interpretedfrom Figs 1 and 3 ofGilmour, 1978)and motion of the ingested food particles (from own unpublished observations); (g)cross-section of a tentacle of an adult phoronid.

F I G . 33.

B. Mechanisms of feeding After several studies on food-gathering methods, it has been found that the manner in which phoronids capture particles from suspension is difficult to determine; the various modes of food selection which have been suggested for Phoronida are summarized below. The first observations were made by Masterman (1900) who suggested that the cilia of the inner surface of the tentacles cause currents of food and water to pass downwards towards the mouth, the outer surface of the tentacles having a non-ciliated epithelium. Bullivant (1968a,b ) pointed out that the mechanism of feeding of Phoronida and other Lophophorata can be termed as “impingement feeding” in analogy with the impingement particle separation used in industrial processes: the sharp deflection of the water currents through the lophophore causes particles to be thrown towards the mouth. Strathmann (1973) believed that impingement would not be effective for particle capture considering the slow speed of the feeding currents and the high density of the food particles in the sea water. The author suggested that food-gathering occurs by an induced feeding current of the lateral cilia of the tentacles. transporting the particles along the frontal surface of the tentacles to the mouth, and by means of local reversals of the lateral cilia as a result of contact of the cilia with passing particles (Fig. 33b). But according to Ryland (1976)the co-ordination of such local reversal must be fairly complex in Bryozoa, and the same problem seems to exist in Phoronida. The inward flicking of the distal part of the tentacles may serve sometimes to transfer particles into the central feeding current and can also be considered as a feeding mode (Fig. 33b). The particles, sometimes nearly loo%, are trapped within the lophophoral concavity; however, some particles are allowed to pass between the tentacles and expelled in the out-flowing current (Strathmann, 1973; Emig, 1 9 7 6 ~ ) . Gilmour (1978) who has examined Phoronis ijimai (as have Strathmann (1973) and Pross (1974)), and Phoronopsis harmeri, observed a different particle selection and rejection method: the edible material is carried towards the mouth onto the epistome by an incoming water current created by the lateral cilia of the tentacles which beat with a dexioplectic metachronal rhythm. The particles may be transported down to the bases of the tentacles where they are collected by the cilia of the oral surface of the epistome and by the oral groove, to converge in two lateral streams in the midline of the epistome to pass into the mouth. The epistome, particularly visible in

species with a coiled lophophore, shows a midline oral groove (own unpublished observations) directed towards the mouth. Heavy inedible particles are expelled by the beat of the frontal cilia to the tips of the tentacles (Fig. 33c, e) in the out-going current. Gilmour (1978) assumed that Phoronida are able simultaneously to accept food particles by a filtration mechanism and reject inert ones by an impingement mechanism; such a theory contradicts the previous models. Particle selection is probably purely mechanical based on both weight and size (Emig, 1976c) and depends on impingement or inertial impaction (Gilmour, 1978). According to the speculations of Pross (1974), recapitulated in 1978, the epistome has the function of raising the lophophore which leads to closure of the oesophagus. The epistome concentrates the particles which are carried by the feeding current as a food-filtrate which passes into the digestive tube by contraction of trunk muscles. Such theoretical statements obviously need experimental corroboration, and Gilmour (1978) noted that the epistome is simultaneously involved “in acceptance of food and rejection of solid waste material and allows the escape of excess water travelling towards the mouth with food particles”. The opinions of the two former authors were put forward earlier by Selys-Longchamps (1907, p. 90) and Cori (1939). Gilmour (1978) also made observations on the food selection of the actinotroch larvae: the tentacles (larval ones?) of old larvae are provided with ciliated epidermal cells which are similar to those of adult phoronids as well as to those described in Rhabdopleura by Dilly (1972).The lateral cilia beat with a dexioplectic metachronal rhythm and so assist the cilia of the perianal ring in driving water (Fig. 33f): the flow in the ventral region is deflected in the vestibule, observed at first by Lebour (1922), where it swirls and the particles are collected and swept to the oesophagus by the cilia of the inner surface of the preoral lobe, while the flow in dorsal and lateral regions is weaker. The frontal cilia beat an antiplectic metachronal rhythm to discharge the heavy particles into the out-going currents. Thus, based on Gilmour’s observations, it can be suggested that the (larval)tentacles and those of the adult lophophore have similar functions in food selection and rejection of inedible material, although the disposition in larva and adult is opposite (Fig. 33).

C. The alimentary canal The digestive tract of Phoronida is U-shaped and its four component parts are represented in the Figs 1 and 25.

58

(‘.

C . EMIG

1. Oesophagus The oesophagal epithelium originates from the ectoderm in the actinotroch, whilst during asexual reproduction the regeneration of the oesophagus occurs by “metaplasia” of the prestomach cells t o assume ectodermal status (Emig, 1 9 7 3 ~ )The . oesophagus, lying immediately internal to the mouth, is lined by a highly ciliated columnar epithelium with numerous gland cells of acid mucopolysaccharide secretion. This epithelium is very similar to and continous with that surrounding the mouth, i.e. the basal epithelium of the outer tentacle row and of the oral side of the epistome (see also Fig. 26). I n Phoronis ijimai, the histochemical results published by Vandermeulen and Reid (1969) disclose an activity of phosphatases and esterases in the oesophagus, which are associated with the digestive processes. These authors showed also the occurrence of lipid droplets surrounded by digestive enzyme films. I agree with the statements of Vandermeulen (1970)that the oesophagus aids in food ingestion and conduction of the foodmass. This is due to a dense ciliation, mucus secretion, and also to numerous longitudinal muscle fibres and some circular ones, together with enzyme synthesis indicated by phosphatase and esterase activity. 2. Prestomach The oesophagus passes rapidly into the prestomach, sometimes called the proventriculus. This is the longest part of the descending branch of the digestive tract, characterized by a broader tube and a simple small and weakly ciliated epithelium, except along the median blood vessel where there is a deep ciliated groove. The prestomach has a weak musculature with circular fibres and some longitudinal ones. Emig (1968) described under the electron microscope t w o cell types in Phoronis psammophila. The first one is laterally ciliated, its apical cell membrane often showing microvillous projections. Small granules (up to 0 3 pm in diameter) bounded by a single membrane are confined in the apical cell region and small rounded vesicles (up to 0.1 pm) frequently associated with the apical membrane have a pinocytotic function. The mitochondria occur always in the supranuclear position. Lipid bodies sometimes appear in the basal region of the cells. The apical membrane of the cells of type 2, which are club-shaped, shows no cilia or microvilli. These cells are characterized by rough endoplasmic reticulum (or ergastoplasm) throughout the whole

THE BIOLOGY O F PHORON1I)A

59

cytoplasm, and by granules, up to 0.8pm in diameter (probably zymogen granules), surrounded by a single membrane, frequently associated with the ergastoplasm. The hypothesis that the two types are distinct or that they are the same is discussed by Emig (1968).In the prestomach of Phoronis ijimai, Vandermeulen ( 1970) described one cell type which shows the characteristics of both previous types, in which the small PAS-positive granules (up to 0 5 pm in diameter), located in the apical region of the cells, show an intense activity of phosphatases and esterase (Vandermeulen and Reid, 1969). According to the two latter authors, the histochemical staining discloses also the distribution of lipid already discribed in the oesophagus. Often, another cell type occurs in the basal region of the prestomachal cells, as well as in the intestine (Emig, 1968; Vandermeulen, 1970; unpublished observations on all phoronid species), generally near the attachment of the mesenteries where these cells can be numerous. The cells contain numerous very electron-dense granules (up to 1 pm in diameter), chromaffin-like ones, which are bounded by a single limiting membrane and separated from this membrane by a clear space. According to Cori (1890), Becker (1938), Hyman (1959), Emig (1968) and Vandermeulen (1970), the prestomach has a primordial role in digestion by secretion of enzymes supported by secretion of zymogen granules and phosphatase and esterase activity; the enzymes are mixed with the foodmass by the beating of the cilia of the prestomach groove. This seems to be its main function, as well as the conduction of this mass towards the stomach, aided by the wall muscles of the prestomach. Its second role is in absorption and storage of lipid and digestive products suggested by the lipid bodies within the cells and by esterase activity, and by the presence of pinocytotic vesicles.

3. Stomach The prestomach passes imperceptibly into the stomach in which the ciliated groove extends over a long part. This latter structure exists in all phoronid species and is not, as stated by Vandermeulen (1970), due to a misinterpretation. The stomach is much larger relative to the other regions of the digestive tube. It appears to lack muscular investment and externally is surrounded by a blood-plexus with a wall structure of type 1 described in the following section. The stomach continues to the ampulla where it ends in a sphincter, the

60

C’ EMI(:

pylorus, and passes into the intestine. The food mass is moved in the stomach by ciliary action. The stomach epithelium in Phoronis psammophila, described by Emig (1967), consists of very tall columnar cells, with an apical microvillous border and paired cilia. From the apical part to the basal one, different regions can be distinguished in each cell: the apical zone contains numerous excretory vesicles (up to 0.5 pm in diameter), zymogen-like granules (up to 1.5pm in diameter) and pinocytic vesicles; then, a mitochondria1 zone with the Golgi apparatus and lysosomes; a zone with lipid bodies (up to 1 pm in diamater), interrupted by the nucleus; sometimes mitochondria occur basally. I n about the middle part of the stomach occur one or more horizontal girdles of gland-cells (of mucus secretion). I n Phoronis ijimai, Vandermeulen (1970) pointed out that prestomach and stomach epithelia “do not differ significantly from each other ultrastructurally and histochemically”, which does not agree with the above observations on Phoronis psammophila. Vandermeulen and Reid (1969) suggested that there is extracellular digestion in the stomach as well as in the prestomach owing to phosphatase activity and to a lesser extent esterase activity. The presence of a single bulging, vacuolated syncytium or of a paired syncytial mass, which appeared capable of phagocytosis and intracellular digestion, has been cited by different authors (Becker, 1938; Cori, 1939; Marcus, 1949; Silkn, 1952; Lonoy, 1954; Emig, 1967); many recent observations lead to the conclusion that such structures occur only in animals in a poor state of preservation and were never seen in phoronids fixed immediately after sampling. The main functions of the stomach in Phoronis psammophila are in absorption and storage of products of digestion which pass into the blood at this level, and secondarily in enzyme production. The absorptive function is provided by the microvilli which increase the absorptive surface by 30-60 times (Emig, 1967). In Phoronis ijimai, Vandermeulen (1970) considered the above second function as the primary one. Both intracellular and extracellular digestion occur in Phoronida. It seems that extracellular digestion takes place particularly in the prestomach and intracellular digestion in the stomach. However, in Phoronis ijimai, Vandermeulen ( 1970) indicated that ingested food particles travel in less than a minute down to the stomach where they remain for up to half an hour; thus, extracellular digestion occurs in prestomach and stomach, and absorption in the intestine. Is the rate of food passage different between the phoronid species, as well as the

length of time necessary for particle passage through the alimentary system Z 4. Intestine The intestine, which is separated from the stomach by the pylorus, represents the whole ascending limb of the digestive tube; its proximal part, located in the ampulla, has a greater diameter and its epithelium is taller, surrounded by longitudinal muscle fibres, while the major part of the intestine is a slender tube with a similar longitudinal muscle arrangement. The intestine is ciliated throughout its length. In Phoronis psammophila, Emig (1968) described one type of intestinal cell; a second type occurs basally, similar to that mentioned in the paragraph on the prestomach. The intestinal cells are ciliated, with a microvillous apical border. Single membranebound secretory granules (0.24.3pm in diameter), numerous vesicles (up to 0-2pm in diameter) particularly abundant in the apical region and in the microvilli, and mitochondria are present in the supranuclear region of the cells. Lysosomes occur in all intestinal cells. The nucleus is located basally. Lipid bodies are more frequent in the proximal hindgut part. In Phoronis ijimai, Vandermeulen ( 1970) defined two intestinal cell types in the proximal part of the intestine: gland cells with sulphated mucopolysaccharide secretion. Gland cells are also described in the intestine of Phoronis australis by Ohuye (1943); however, they were never observed in P. psammophila. No phosphatase activity nor presence of lipid has been demonstrated, only traces of esterase activity in the intestine (Vandermeulen and Reid, 1969). The intestine has a primarily absorptive function (Vandermeulen, 1970),which, except in the proxinial part, seenis.less important in elaboration of faecal pellets, which are rejected by a combination of ciliary and muscular action through the anus, situated on an anal papilla. The anus is encircled by a tall, probably glandular, epidermis. The function of the basal cell type in the prestomach and in the intestine may, according to Vandermeulen (1970), be a neurosecretory control of digestive processes.

D. Food particles ingested by Phoronida Possible food types, nature and abundance, available to Phoronida are largely unknown. Various authors suggest algae,

diatoms, flagellates, peridinians, small invertebrate larvae, detritus, etc. Future studies are needed on the food sources of phoronid species. E . Uptake of dissolved organic matter The direct uptake of dissolved organic compounds, through the epidermis, by marine invertebrates, including suspension feeders, is now generally considered a normal process of nutrition (cf. Emig and Thouveny, 1976).However, the importance of this food source is not yet known for Phoronida, except in Phoronis psarnmophila, which is able t o remove actively amino-acids from natural concentrations in the sea water. The experiments on I4C-valine uptake by Phoronis psammophila, reported by Emig and Thouveny (1976), show polyphasic kinetics of uptake at external concentrations higher than 1.5 pM (Fig. 34). This process through the trunk wall and leads t o an

Min

F I ~34. : . Kinetics of ‘‘C:-valine uptake at two external concentrations in I’horonis psummophila in winter (after results of Emig and Thouvmy, 1976)

internal accumulation of amino-acids. The internal concentration is twice in winter and 7-20 times in summer t h a t of the external medium. The seasonal variations suggest the intervention of an active transport which occurs only during the summer time, which is confirmed by temperature experiments and the action of ATP inhibitors on amino-acid absorption. The tube of phoronid species is not a significant barrier t o uptake of dissolved organic matter from the surrounding medium. The cilia of the epidermal cells produce an exchange of water within the tube.

63

THE I3IOLOUY OF PHOROSI1)A

I n species embedded in soft sediments the rear end of the tube has a small opening (Fig. 28a). VIII. CIRCULATORY SYSTEM Knowledge from earlier sources has been summarized by SelysLongchamps (1907), Cori (1939) and Hyman (1959). Interesting recent work and observations will be discussed in the following account.

A. General structure The circulatory system of Phoronida is of the closed type, with red blood corpuscles (erythrocytes). In all species except Phoronis ovalis, there are two longitudinal trunk vessels (Fig. 35), a median Tentacular capillary

n/

\n

Lophophoral vessel

Accessory vessel

j

Left lateral vessel

1 1

Right lateralvessel

Medianvessel

-Lateral

vessel

Captllary caeca \

;;

/1 ,I

Stomach

A

d

$ 1 , I

, I

(a)

FIG.35. Circulatory system 1979).

in (a)Phoronis

ovalis, and (h) other phoronid species (froln Emig.

vessel, arising from the extensive stomach blood plexus, and a lateral vessel, originating from the lophophoral vessel by two branches (a left and a right one) which unite at the oesophageal level. Both longitudinal vessels communicate through the blood plexus and the lophophoral vessel. The lateral vessel gives rise along most of its length to numerous blind-ending diverticula called capillary caeca. A t the level of the blood plexus, where the caeca are somewhat larger, they are particularly well developed along the lateral vessel, the secondary lateral vessel, which occurs only in the plexus (Fig. 35), and on the posterior end of the plexus (Figs 1,3,35).In Phoronis ovalis (Fig. 35), three longitudinal vessels occur, a median vessel and two lateral ones. The left lateral vessel has an oral branch called the “accessory” vessel, along the oral mesentery. The lophophoral vessel, following the shape of the lophophore, is composed of closely applied afferent and efferent branches; at the junction of both branches a tentacular capillary arises and ascends in each tentacle (Figs 1, 35). More details on the general structure of the circulatory system and of its regeneration are given by Selys-Longchamps (1907) and by Emig (1971a, 197213, c). B. Circulation and function The median afferent vessel contains venous blood and is the main blood vessel (Figs 35, 37). The flow is produced by peristaltic waves along the vessel. Regular rhythmic contractions, about 4-1 6/min, are performed by the strong muscular layer of the blood vessel wall (Wright, 1856; Selys-Longchamps, 1907; Bethe, 1927). Bethe (1927) pointed out that the waves originate at a point where the median vessel arises from the blood plexus. The blood passes from the median vessel through the T-shaped vessel into the afferent branch of the lophophoral vessel, and then fills each highly contractile tentacular capillary. Up and down movement of the erythrocytes within t h e tentacular capillary occurs in “strings” which undergo very rapid acceleration caused by the blood flow; each “string” is always preceded by a single blood corpuscle, 2CL30pni in front of the “string”. The erythrocytes are often distorted by the small diameter of the capillaries (Fig. 36). The blood remains stationary for some seconds before the capillary empties by a strong muscular contraction, from the tip basally, into the efferent branch through a valve located in the proximal part of the capillary at the level of the

65

/

Nucleus

\ Blood corpuscle

36 Distortlon ofan erythrocyte by friction along the capillary wall during blood flou (fr
junction between the afferent and efferent branches of the lophophoral vessel. The frequency of contractions in the tentacles is 7-13/min (Bethe, 1927; Emig, 1966). In general, no coordination can be seen between the flow in neighbouring capillaries which contract autonomously. The respiratory gas exchange takes place while the erythrocytes are in the tentacles (Fig. 37). There is no doubt as to the respiratory

7 1 ,

Lophophoral vessel

Lateral

--4Nutrients

FI(Z37 Diagram ofthe main function of the circulator) In the future

SJ

stem The gaps have to be cotnplrted

function of the lophophore in Phoronida (Emig, 1 9 7 6 ~ ) .This important function, like that of nutrition, is implied in the rapid regeneration of this organ and its circulatory apparatus after autotomy or removal by predators. Such process have been observed in several phoronid species (Emig, 1972b, c, and unpublished studies on Phoronis ijimai, P. hippocrepia and Phoronqsis harmeri).

66

('

ISMIC:

From the lophophoral vessel, the blood descends into the lateral vessel. In both vessels, the muscular contractions are weaker and less regular than in the median vessel. The lateral vessel is efferent and filled by arterial blood. Along most of its course this vessel gives off numerous generally simple capillary caeca, extended particularly in the oral coelomic compartments, and sometimes in the anal ones, of the metacoelom. The caeca show, like the tentacular capillaries, autonomous contractions and they empty by a sudden contraction with a frequency of 3-l0/min, each contraction is followed by a longer period of relaxation during which the caeca fill again. Bethe (1927)suggested that the function of the caeca is similar to that of the capillaries of the vertebrates. My opinion is that the gas distribution to the different organs is provided by means of the coelomic fluid after gas-exchange between this fluid and the blood of the caeca, the vessels and the plexus (Fig. 37). The circulation of the coelomic fluid is induced by the muscular contraction of the body wall. This opinion was expressed by SelysLongchamps in 1907. The circulatory system also distributes the digestive products, the uptake occurring between the digestive tube and its blood plexus. The nutrients would be distributed like gases, with the coelomic fluid assisting (Fig. 37). Nutrients reach the germ cells for use during gametogenesis by means of the blood caeca to which these cells lie adjacent but probably the coelomic fluid serves also as a means of distribution of dissolved nutrients to the gonads. Selys-Longchamps indicated a third function, namely storage of lipids. A reverse blood flow or to and fro movements may occur during disturbance, and always during the first regeneration stages before the junction of the two longitudinal trunk vessels.

C. Walt structure of the circulatory apparatus The results of recent studies on the wall structure of the vessels and capillaries by Kawaguti and Nakamichi (1973) on Phoronis australis, by Emig (1977e) on P. psammophila and by Storch and Herrmann (1978) on P . muelleri are summarized below. 1. Type 1

The walls of the capillaries and the blood plexus consist, from the interior to the exterior, of endothelial cells, a thin basal lamina and peritoneal cells (Fig. 38). However, the blood is frequently in direct

67 Coelomic fluid Mvofilaments

Peritoneal cell

Circular muscle fibre!S

3 Muscular cells

Longitudinal miJscle fibres

Basal lamina --Myofilaments ---Endothelial

cell

4 Flu. 38. Diagram of the wall types of the circulatory apparatus in l’horvnis psammoph,ila.

contact with the basal lamina, the endothelium being sparsely distributed. In some areas, the peritoneal cells, or both endothelium and peritoneum, are lacking and then the basal lamina directly faces the lumen of the capillary. Such regions are especially favourable for exchanges between blood and coelomic fluid. The thickness of the walls of Type 1 is about 1 4 p m , that the basal lamina is of about 0.1 pm ( 0 2 4 5 p m in the plexus at the bases of the stomach cells).

In peritoneal cells there are often bundles of myofilamen ts, located near the basal lamina. The myofilaments run mostly longitudinally in extended capillaries; they are of 10-35nm in diameter. According to Kawaguti and Nakamichi (1973),the bundles of myofilaments occur in various positions in concentrated capillaries. These authors discriminate also between two kinds of cells, muscle cells and peritoneal ones, but they frequently observed peritoneal cells containing muscle fibres. 2. Type 2

Type 2 (Fig. 38), the most common in the circulatory system, occurs particularly in the broader parts of the longitudinal trunk vessels. The wall is thicker than in the former type owing to a higher external layer of muscle cells, but no peritoneum lining is observed. The muscle fibres are almost longitudinal; in Phoronis psammophila circular muscle fibres are rarely observed covering the former cell layer. The thickness of the basal lamina is of about 0.1-0.3 pm. The endothelium is sometimes wanting. Small bundles of myofilanients appear in some endothelial cells; they are very thin and almost scattered, and circular.

3. Type 3 The walls of the longitudinal trunk vessels show important muscular layers in some regions which are responsible for the blood circulation. Such a structure occurs in the distal part of the median vessel. The wall has a thick outer muscular layer (Fig. 38). The longitudinal muscle fibres are intimately connected with the basal lamina. I n Phoronis australis, Kawaguti and Nakamichi (1973) describe only a longitudinal muscle layer with cross-striated fibres, while in P. psammophila two smooth muscle layers occur, an outer circular one and a longitudinal one, which together reach 6pm in thickness (Emig, 1977e). The basal lamina shows a folded belt of 0.3-1 pm in thickness. The endothelium covering is continuous, of about 3 pm thick. The endothelial cells often contain bundles of my ofilaments.

4. Podocytes in Phoronis muelleri Podocytes which show radiating processes ramify in many pedicels interconnected by slit membranes (Fig. 39); they are often

encountered in the coelomic lining of different metasomal blood vessels in Phoronis muelleri (cf. Storch and Herrmann, 1978). These cells contain myofilaments near the basal lamina, and they are Coelomic fluid Pedicels and slit membranes

Basal lamina

Blood vessel

’w . PIC.39 Diagram of podocyte lining o f a blood vessel of P h o r o i ~muellrrc i~ The dotted line shoxh an area with infoldings of the wall.

involved in the formation of the fenestration of the vessels. The wall structure appears to be related t o the type 1 described above. The presence of such structures very similar to podocytes, only known in P . muelleri, is confirmed by the unpublished observations of Mattisson (personal communication). The capillary caeca are lacking on the major part of the lateral vessel of P. muelleri, which suggests that their exchange function is here performed by the podocytes, which are known from different excretory organs as sites of ultrafiltration.

D. Blood corpuscles The blood of phoronid species is composed of a colourless plasma containing solitary leucocytes (which have never been seen in the blood of Phoronis psammophila and their occurrence is considered by the present author as uncertain) and red corpuscles, the pigment of which is a kind of haemoglobin, originally established by Lankester and confirmed since by several authors (Cori, 1890, 1937; Ohuye, 1943; Lonoy, 1954; Emig, 1966). Recently, Garlick et al. (1979) studied the intracellular haemoglobin of Phoronopsis harmeri which is composed of four unique polypeptide chains, two of which can combine t o form hetero- and homodimers and two of which do not associate at all. The four chains all have molecular weights of about 16 900 and they have been characterized by amino-acid composition, tryptic peptide patterns and the amino-acid sequence of the NH, terminal segment. The authors have determined the oxygen

equilibrium of a dimeric fraction at pH 7.5 and 20°C, and the pressure of half saturation which is 2.3 mm Hg. The erythrocytes are generally approximately spherical, up to 12 pm in diameter. Their nuclei, about 3 pm in diameter, are located rather eccentrically; the dense coloured cytoplasm contains haemoglobin, Golgi apparatus, lysosome-like granules whose numbers increase in old cells, vesicles, some mitochondria which often surround the nucleus, and a poor ergastroplasm (Fig. 40). Mattisson

FIo. 40. General view of a n erythrocyte of Phoronispsammqphilu. The Golgi apparatus lies near the nucleus, which shows some nuclear pores. Near the outer membrane occur mirrovesicles, which apparently coaleme to form larger ones; they are probably pinocytotic vesicles.

(unpublished results) has shown the occurrence of a marked endocytotic activity indicating a function of defence like that of white blood cells of higher organisms. I n his ultrastructural studies, Mattisson has found a transport of mitochondria into the nucleus of the erythroaytes. Such an engulfing of mitochondria by the nucleus has also been. reported to occur during spermiogenesis in the crustacean family Trogulidae by Juberthie and Manier (1977). Cori (1890) pointed out that the red corpuscles possibly arise from the endothelial cells of the blood vessels. This origin was corroborated

THE I%IOI,OUY OF PHOR0NII)A

71

by Ohuye (1942) in Phoronis australis: the endothelial gives rise to the erythrocytes and to the leucocytes. However, the observations of a similar origin of the blood cells, arising from the peritoneal lining and from the vasoperitoneal tissue (particularly during regeneration: Ohuye, 1943), are untrustworthy and need more detailed studies in the future. It is of great interest to compare the structures of the wall and blood corpuscles of the Phoronida with those of other invertebrate groups, whose vascular systems up to now have been little studied, to establish possible relationships and probable phylogenetical evidence. It is suggested that the wall structure of type l is for low blood pressure which allows uptake and exchange of gases or food, while in type 3 strong muscular contractions pump blood around the whole system.

Ix. PHYLOGENETIC RELATIONSHIPS OF PHORONIDA The phylogenetic position and relationships of the Phoronida since their discovery have been the subjects of extensive controversial speculations. In the light of recent interpretations based on new information, particularly on developmental patterns and larval morphology, it is now generally agreed that the Lophophorata, to include Brachiopoda, Bryozoa and Phoronida, should have phylum status (Emig, 1977a). The position of the Phoronida within the Lophophorata, one of the most interesting groups in the phylogeny of the animal kingdom, as well as the relationships of the Lophophorata within the Chordata trend, and particularly in the Archimerata assemblage, will be discussed in relation to phylogenetic arguments expressed in recent studies (Valentine, 1975; Siewing, 1976; Emig, 197613, 1977a; Farmer, 1977; Nielsen, 1977; Zimmer, 1978; Gilmour, 1979).Immediately, it may be suggested that the Phoronida share a larger proportion of evolved features than in the other lophophorate groups, and that the Lophophorata form a close-knit group with strong similarities deserving reassessment as possible ancestors from which the other Archimerata groups may have evolved. A. Archimeric subdivisions, morphological adaptations and phylogenetic relationships In general, Archimery ( = Oligomery, Paurometamery, Trimery, ... ) is defined by three body regions, prosome, mesosome and metasome, each with its own paired or unpaired coelomic cavity.

Recent findings (see Section 111)lead to a general agreement that the fundamental pattern in Phoronida is the regionalization of the larval and adult body into three archimeric parts, each with an unpaired coelom (Fig. 25); their disposition is somewhat different in actinotroch and adult (Emig, 1976a). Archimery in Brachiopoda is fully anticipated in the larvae of Inarticulates, while in adult Articulates only the meso- and metasome are already developed, as well as in Bryozoa, in which only the Phylactolema show the presence of an epistome. The archimeric subdivision in the Lophophorata is one of the characteristics that no longer justifies their larvae being regarded as a modified trochophore; instead these larvae, especially actinotrochs, are to be related to the Dipleurula-t,ype, which corroborates my previous opinion (Emig, 1976b). 1. Prosome The preoral lobe of the actinotroch is homologous with the epistome, both structures being equivalent to the prosome, and homologous with similar structures in Brachiopoda and Bryozoa. The origin of the coelomic cavity of the preoral lobe is entirely independent from that of the other coeloms in the actinotroch; in the juvenile adult, a discrete protocoel is retained throughout metamorphosis as the protocoelom, the cavity of the epistome (Fig. 24). A small communication is formed between the protocoelom and the mesocoelom in adult forms (Emig and Siewing, 1975). Zimmer (1978) suggested that a remarkable parallel exists between the fates of the preoral lobe of phoronids and asteroids. The preoral lobe and epistome of lophophorates function in food-collecting and escape of excess water during feeding (see Section VII). Gilmour (1979)pointed out that the epistome is similar in internal structure to the proboscis of Hemichordata: the gill-slits originate from an epistome-like structure of a lophophorate ancestor, which is compatible with the present speculations. Such a structure, according to the same author, has become elaborated into a system of folds and grooves which may become fused to form a series of water pores, precursors of the gillslits of chordates, as was first suggested by Masterman (1896). The actinotroch, like the other lophophorate larvae, retains the apical plate (larval nervous centre) which is lost during metamorphosis (Fig. 24) and does not serve as the primordium of the adult nervous ganglion, which is also located in the prosome (Emig, 1976a, b); this is corroborated by recent findings (Siewing, 1974; Zimmer, 1978; see also Section IV).

In actinotrochs the selection of a suitable substrate during settlement is accomplished by the piriform organ, or by the apical plate in species which lack this structure (Figs 20, 21). In the bryozoan cyphonaute larva the similar function is performed by the vibratile plume at the anterior end of the piriform organ, a complex of glandular cells whose function seems to be enigmatic (Farmer, 1977). In both larval types the piriform organ and the apical plate are united by a cord. Emig (1980)suggested that the piriform organ seems only to be associated with the ecological behaviour of the actinotroch without having any true evolutionary relationship. 2. Mesosome The mesosome of the adult phoronids, surrounding the prosome (Figs 1, 25, 33), is represented by the lophophore, whose definition and functions have been discussed in Sections 11, VII and VIII. Through its definition, the lophophore has an important phylogenetic significance (Emig, 1 9 7 6 ~ )The . mesosome contains also the mouth, the ring nerve, the lophophoral vessel and accessory sex glands. The mesocoel differentiates in the collar region of the actinotroch, generally in the late larval stage, the remaining collar space being blastocoelic (Figs 9, 10, 25; Table 11). Clark (1964) and Trueman (1968)argued that the tentacular apparatus or proboscis are secondary modifications in the acquisition of a burrowing habit, toward a sessile mode of life, and in this way, the retractable lophophore of the Bryozoa is interpreted by Farmer et al. (1973) as a coadaptation to colonial life. The evolutionary model of the lophophore, convincingly discussed by Cowen ( 1974),should only be applied to the Chordate trend. Emig (1974a) pointed out that phoronid brooding species with a lecithotrophic larval stage are derived secondarily from forms which shed their eggs directly into the sea water, involving a long pelagic stage. The former pattern is considered as a more adapted reproductive strategy in Phoronida; brooding tends to reduce the pelagic phase which minimizes losses from larval predation and reduces larval mortality. The present argument needs further investigation in relation to spermatophore production and habitat (all burrowing species have brooding patterns). A similar view on Bryozoa has been expressed by several authors (e.g. Emig, 1976b; Farmer, 1977), considering the planktonic larva cyphonaute as an archaic and presumably primitive type, while, in contrast, SilBn (1944)believed brooding forms to be primitive in Bryozoa. Recently,

74

(’. c‘. EMIU

Strathmann (1978) pointed out that Oligomera do not re-acquire a planktotrophic larval phase once i t is lost, the loss being accompanied by an extensive loss of larval structures used in feeding, and that a secondarily acquired larval planktotrophy may not be as effective in permitting reduced egg size and hence greater fecundity. These arguments confirm our former statement. Phoronida in their lack of cephalization, and as regards the disposition of their nervous system (Emig, 1976b),may be considered as “Epineuriens”: the nervous system is basiepithelial and has shared similarities with those of the Hemichordates and Echinoderms, while relationships with the Spiralia have been disproved. I n the genus Phoronopsis, the epidermal collar fold at the basis of the lophophore (Fig. 1) leads to a primitive neurulation and may be considered as an evolutionary process which has t o be compared with the collar nervous cord of Hemichordata and with the radial nerve of Echinodermata. 3. Metasome

The metasome is the main locomotory organ of the Phoronida, adapted t o a larval pelagic existence (or to a creeping life in t h e larva of Phoronis ovalis) and in the adult to a sedentary or sessile mode of life (see Sections I1 and VlI). The adult trunk, and particularly its ampulla, is used t o form a living burrow and may be homologous with similar structures, such as adhesive organ and pedicle which occur in other Archimerata groups. It shows a secondary internal subdivision into four cavities separated by mesenteries (one or two are lacking respectively in Phoronis muelleri and P . ovalis).Similar subdivisions occur in the Hemichordate Cephalodiscus and Ptychoderidae. A convincing correlation between the archimeric condition and the adult tubicolous habit has been put forward by several authors, which is corroborated by the fact that the evolution of a coelomic cavity (primitively a hydrostatic skeleton) is closely tied to the acquisition of a burrowing habit (Clark, 1964; Trueman, 1968). Recent phylogenetic arguments refuted the opinion of Remane (1950) that the oligomerous condition represents a primary adaptation of early coelomates; this leads to the rejection of the term Archicoelomata introduced by Masterman (1898) and used by German workers. Structures contained in the metasome have phylogenetic implications. The adult digestive tract of Phoronida is U-shaped, bringing the anus anteriorly to lie close to the mouth and reducing to

T H E BIOLOGY OF PHOIL0SII)A

75

a remnant the dorsal side of the trunk (Fig. 2 5 ) .The mouth orginally occupied an antero-ventral position and the blastopore has at no time the function of a mouth, both structures evolving independently (Williams, 1960). The Protostomia-Deuterostomia theory, initially proposed by Grobben (1908), and used for classification by phylogenetic relationships has been refuted by many previous zoologists (e.g. Lovtrup, 1975; Siewing, 1976; Emig, 197613);this old concept is not in conformity with modern embryological findings and can no longer be upheld. Protonephridia with solenocytic cells occur in the actinotroch, metanephridia in the adult phoronid (Figs 1, 11); t h e relationship of both structures has been discussed by Emig (1973a, 197613) and future investigators have t o try to reach conclusions on nephridial evolution in Phoronida. I n this way, the arguments of Wilson and Webster (1974), that the evolution of the blood vascular systems of moderate pressure provides an alternative muscle-powered filtration force, and other forms of kidney have evolved to replace the protonephridium, may be accepted in Phoronida (see Section VIII). The apparent structural similarities in protonephridia would be the result of convergent evolution imposing a conformity based on functional requirements. Protonephridial terminal organs are divided into three or four different groups which are probably not interrelated, and the protonephridia are assumed to be homologous throughout the groups in which they occur (Wilson and Webster, 1974). As in actinotrochs, the lancelot Branchiostma (Cephalochordata) possesses protonephridia with a number of solenocytes. However, in general very little is known of the way in which solenocytic cells function. Wilson and Webster (1974) give evidence to support the early notion of Goodrich (1945) that protonephridia in Kamptozoa (Endoprocts) conformed to the general platyhelminth pattern. This is another argument to refute Nielsen’s (1977) conclusion that Bryozoa and Kamptozoa stem from a common ancestor.

B. Other phylogenetic expression In Phoronida, as in the other Archimerata groups, the egg cleavage is of the radial type, total and subequal. The differentiation of the mesoderm occurs by a derived enterocoelous method, considered as a primitive stage of enterocoely (Emig, 1976b, 1980), which may be a general pattern in the Chordate lineages. In recent studies on Bryozoa, Zimmer and Wollacott (1977) described the

arising of mesenchymal cells from ectodermal tissue, specifically the epidermal blastema (apical disc). Such cells occur throughout the blastocoel of the bryozoan larvae, and the adult coelomic cavities become defined through the rearrangement of these cells, while the mesodermal lining of the basal part or of all ectodermal derivatives become peritoneum after metamorphosis. The ribosomal RNA of Phoronis australis, Lingula anatina and Terebratalia coreana studied by Ishikawa (1977) may be of the “protostomiate” type with respect to thermal stablity and molecular size. However, the opinion of Ishikawa (1977) that the ribosomal RKA’s molecular characteristics support the peculiarity of the embryology has to be refuted since he was not aware of recent embryological findings on Lophophorata. That the lophophorate ribosomal RNA is distinct from that of Echinodermata is not an adequate reason for opposing the combination of the Archimerata groups. Phylogenetic speculation based on this biochemical aspect needs more information. I n relation to the intracellular haemoglobin, Garlick et al. (1979) concluded with caution that the Phoronopsis Hb-Ia sequence appears to be evolutionarily far from Protostomia and Deuterostomia and perhaps about equidistant from each group.

C. Relation of the Phoronida to the other Lophophorata There is general agreement that the Brachiopoda, Bryozoa and Phoronida share a large proportion of structures, with strong similarities in morphology and embryology, so as to constitute a closely related group, the Lophophorata. Although they have often been regarded as independent phyla, there is no reason to detach the three groups from one another. They constitute not only a phylogenetic assemblage but also a natural systematic unit which, as a taxon, deserves the status of a phylum (Emig, 1977a), the Brachiopoda, Bryozoa and Phoronida being classes (Fig. 41). The Lophophorata have the same well-defined characteristics as the other Archimerata phyla (Fig. 42). The strong similarities shown by the three lophophorate classes are indicative of common ancestry and the relation of the Phoronida to the Brachiopoda and Bryozoa will be considered in the light of recent proposals, and particularly the question whether the Phoronida may be placed nearest to the ancestral stock (e.g. Hyman, 1959; Farmer et al., 1973; Valentine, 1975; Farmer, 1977) or be

Class Phoronida Solitary, t,ubicolous animals; U-shaped digestive tract wit.h mouth and anus: nervous aystem with ganglion between mouth and anus, ring nerve. one or two giant fibres; paired metanephridia (also acting as gonoductx): closed circulatory system with erythrocytes.

Class Bryozoa Sessile. colonial animals living in e x o skeletal cases or gelatinous matrrial; l1shaped digestive tube with mouth and nus: retractable lophophore: nervous system with ganglion between mouth a.nd anus, nerve ring; laek nephridia and circulatory SJ ' Tstem.

/": Class Brachiopoda Solitary bivalved animals, bilaterally symmetrical. attached to the substrate by a pediclt. o r directly cemented or free; ventral and dorsal valves lined by mantle extensions (epidermal) and surrounding the lophophore; digestive tract with or without anus; nervous system with principal centre below the oesophagus; one, rarely two, pairs of metanephridia (also act,ing as gonoducts); circulatory system open.

\

Phylum Lophophorata into three archimeric regions, each with its own coelomic cavit,y (prosonie representd by the preoral lobe and epistome, the mesosome by the larval collar region and the lophophore; metasome contains the main organ systems); nervous system basiepithelial, lack of cephalization; exoskeleton of their own secretion; egg cleavage radial and total; coelorn issued from archenteric cell proliferation by a modified enterocoelic method (unknown in Bryozoa); larva related to Dipleurula type.

FIG.41. The taxonomic and systematic relations between the Lophophorata, established on the basis of a phylogenetic reconstruction.

78 Chordoto Phylum Hemtchordata

-

m

1

Phylum Echinodermato

Phylum Lophophorato

F~I:. 12. Phylogenetic classification of the Archimerata phyla anti their relationuhips, and their position in the animal kingdom.

regarded. as the most evolved within the Lophophorata (Emig, 197613). Clark (1964)argued that the primitive function of the coelom was to act as a hydrostatic skeleton in burrowing and in this way the Lophophorata were derived from an infaunal ancestor of tubicolous habit to which the coelom is a functional necessity. Valentine (1975) estimated that this adaptive radiation probably occurred in the late Precambrian time. Another interesting argument concerns the modes of respiration (Cowen, 1974) and feeding (Webb, 1969) which become adaptive in the Lophophorata by development of a lophophore. Thus, speculation on the ancestral form suggests a softbodied infaunal vermiform burrow-dwelling animal, with a tentacle crown for feeding and respiration, and a fluid-filled coelomic space in trunk and tentacles (probably meso- and metasome) to produce an archimeric body plan. This ancestor lacked a skeleton and the time of its origin is unknown and perhaps unknowable.

THE HIOISWY 01"PHORONIDA

79

Speculation as to radiation from the ancestral morphological form can only be based on the structure of existing animals so leading to consistent phylogenetic reconstructions. The prolophophorate form evolving from the Precambrian fauna would be of a small size in the absence of highly evolved organ systems. Presumably under environmental changes it developed a semi-epifaunal and epifaunal mode of life, or kept an infaunal one, with increase in size and in morphological complexity. This population gave rise to three successive evolutionary branches: the Brachiopoda arose first, then the Bryozoa and the Phoronida appeared successively, the Lophophorata being considered as monophyletic (Fig. 41). In other words, the common ancestral stock gave rise to the Brachiopoda and the Bryozoa as blind branches, then continued along its main line of evolution to the Phoronida. Valentine (1975) proposed that inarticulate brachiopods developed on soft substrata, subsequently radiating onto solid ones (craniids) and back into infaunal lingulids, while Articulates evolved from an ancestry which lacked a pedicle on hard substrata which later led to species which reinvaded soft sediments. The second lineage, the Bryozoa, may have followed a pathway of miniaturization and colony formation leading to the loss of some organ systems (nephridia, circulatory system) by adaptation and evolutionary divergence (Jebram, 1973; Farmer, 1977). According to morphological and behavioural features, Phoronis ovalis appears as the phoronid approaching nearest to the bryozoan lineage (SilBn,1952; Emig, 1969,197613;Farmer, 1977).This species is also the most primitive of the Phoronida and the oldest known phoronid fossil (see Section VI). On similarities with a phylogenetic basis the Phoronida may appear to show the highest degree of evolution, especially in fulfilling their life cycle and by their reproductive strategy, which emphasize the advantage of a rapid metamorphosis and regeneration (of the anterior body part and in asexual propagation). Phoronida also stand nearest to the Pterobranchia (Hemichordata). Brachiopoda may have evolved in the latest Precambriam times; the Bryozoa are known since the Ordovician and Phoronida since the Devonian. Each lophophorate class has assumed a selection of the optimal reproductive method (planktotrophic, pelagic lecithotrophic, nonpelagic lecithotrophic) for the opportunistic use of the habitat by the adult. Controversy exists in tracing the evolutionary trend of the actinotroch from a cyphonaute ancestor (Jagersten, 1972). Conversely the actinotroch may be the ancestral form for the

Bryozoa, particularly of the cyphonaute from which the other bryozoan larvae are modified or altered forms (Farmer, 1977). The fact that the cyphonaute larva is highly adapted for an extended pelagic existence (up to 2 months) appears to be a more primitive condition compared with actinotroch types (see foregoing pages). On the other hand, the larva of Phoronis ovalis approaches in many respects the morphology of some bryozoan larvae. The study of the development and anatomy of this larva will probably resolve the present controversy. However, it seems reasonable to imagine an intermediate form from which both actinotroch and cyphonaute were derived. In any event the lophophorate evolution proposed in this paragraph is incompatible with Nielsen’s (1977) arguments that the Bryozoa originated from the Kamptozoa ( = Endoprocta), arguments which have been refuted by Emig (197613,c) and Farmer (1977) and which cannot be retained in the Archimerata concept.

D. Relation of the Lophophorata to the other related Phyla The following phylogenetic reconstruction of the relationships of the Lophophorata in the animal kingdom is based on adaptive transformations brought t o light in the present review on the biology of Phoronida and by several investigators during the last decade (for bibliography refer to, e.g. Jollie, 1973; Valentine, 1975; Emig, 1976b; Siewing, 1976). The phyla with archimeric body regionalization have several homologous features; they may be considered as a monophyletic assemblage, called Archimerata by Emig (1976b). The term Archicoelomata, initially proposed by Masterman ( 1898), has been rejected earlier in the present review on the basis of recently published arguments. One major homologous character appears to be the mesosome and its coelomic cavity, whose derivatives take the form of tentacular outgrowths and arms to the ambulacral system. Stephenson (1976) agreed with Nichol’s hypothesis (1967) that the echinoderms originate from a lophophorate animal but he stated that the tentacles of the lophophore have to be considered homologous with the echinoderm tube feet. The Archimerata assemblage brings together the Lophophorata, the Hemichordata and the Echinodermata, which all stemmed from a common lophophorate ancestor. I n other words, it is suggested that the Lophophorata may be a possible ancestor of the Chordata (Fig. 42),which gave rise, on the one hand, to the Echinodermata, which evolved as a blind branch, and, on the other hand, to the Hemichordata and Chordata; the latter is a

hypothesis which is generally agreed upon; this is in contrast to the speculative hypothesis of Gutmann and Bonik (1978).The Phoronida show similarities to the Pterobranchia, which are generally admitted to be more primitive than Enteropneusta. A close relationship between Hemichordata and Chordata has been generally acknowledged, but the ideas of the different authors on the prochordate ancestor vary. Recently Welsch and Welsch (1978) investigated the suggestion of homology of the preoral ciliary groove of Enteropneusta (Hemichordata) and the Hatscheck’s pit of Branchiostoma (Acrania) with the vertebrate adenohypophysis, and this was neatly confirmed by them. I n accordance with Ulrich (1972) and Siewing (1976), the Archimerata group should not be considered only as a phylogenetic concept but also as a natural systematic unit. It does not represent a central group in the Coelomata classification as was proposed by Siewing (1976, Fig. 13). This corroborates my former opionin (Kmig, 1976b) that the Archimerata lie at the base of the Chordata stern of which they are primitive precursors and that they show very few, even no, similarities with the Spiralia. To summarize these speculations, the Lophophorata, and particularly the Phoronida, are allied with the Chordata stem of which they represent an early evolutionary offshoot. It appears beyond doubt that they are allied neither to the Spiralia, nor to some central position between the Spiralia and Chordates.

X . REFERENCES Anadon, R. and Anadon, E. (1973). Primera cita para Espana del phylum Phoronida. Boletim dela Real Sociedad espa6ola de Historia natural (Biologica)71, 193-200. Andrews, E A . (1980). On a new American species of the remarkable animal Phoronis. Annals and Magazine of Natural History 5 , 445-449. Avnimelech, M. (1955). Occurrence of fossil Phoronidea-like tubes in several geological formations in Israel. Bulletin of Research Council of Israel 5, 174 177. Barnes, R . S. K . and Coughlan, J. (1972). The bottom fauna of the Blackwater estuary: the macrofauna of the area adjacent t o Bradwell nuclear generating station. Essex Naturalist 33, 15-31. Becker, G. (1938). Untersuchung uber den Darm und die Verdauung von Kamptozoen, Bryozoen und Phoroniden. Zeitschrift f u r Morphologie und Okologze d w Tiere 33, 72-127. Benham, W . B (1889). The anatomy of Phoronis australis. Quarterly Journal of Microscopical Sciences 30, 125-158.

82

c‘.

EMIC

Bethe, A. (1927).Eigenthiimliche Formen und Mittel der Blutbewegung (Phoronis, T m q t e r i s , Squilla). Zeitschrift f u r vergleichende Physiologie 5, 555-576. Boesh, D. F., Diaz, R. J. and Virnstein, R . W. (1976).Effects of tropical storm Agnes on soft-bottom macrobenthic communities of the James and York estuaries and the lower Chesapeake Bay. Chesapeake Science 17, 246-259. Brooks, W. K . and Cowles, R. P. (1905).Phoronisarchitecta: its life history, anatomy and breeding habits. Memoirs of the National Academy of Sciences, Washington 10, 72-113. Browne, E. T. (1895). On the change in the pelagic fauna of Plymouth during September, 1893 and 1895. Journal of the Marine Biological Association U . K . 4, 168- 173. Browne, E. T. (1900).The fauna and flora of Valencia Harbor on the west coast of Ireland. Proceedings of the Royal Irish Academy 5, 667-693. Buchanan, J. B. (1963). The bottom fauna communities and their sediment relationships off the coast of Northumberland. Oikos 14, 154-175. Buchanan, J . B., Sheader, M . and Kingston, P. F. (1978).Sources of variability in the benthic macrofauna off the South Northumberland coast, 1971-1976. Journal of the Marine Biological Association, U . K . 58. 191-209. Bullivant, J. S. (1968a). The method of feeding of lophophorates (Bryozoa, Phoronida, Brachiopoda). New Zealand Journal of Marine and FreshwatPr Research 2, 135-146. Bullivant, J. S. (1968b).A revised classification of suspension feeders. Tuatara 16, 1 5 1- 160. Caldwell, W . H. (1882). Preliminary note on the structure, development and affinities of Phoronis. Proceedings of the Royal Society of London 34, 37 1-383. Clark, R. B. (1964).“Dynamics in Metazoan Evolution”. Clarendon Press, Oxford, 313 pp. Cori, C. J . (1889). Beitrag zur Anatomie der Phoronis. Inaugural Dissertation, Prague (Universitat Leipzig), 48 pp. Cori, C. J. (1890). Untersuchung uber die Anatomie und Histologie der Gattung Phoronis. Zeitschrift f u r wissenschaftliche Zoologie 51, 486568. Cori, C. J. (1937). “Phoronidea. Handbuch der Zoologie” (Kiikenthal and Krumbach, eds), Vol. 3, pp. 71-135. Cori, C. J. (1939).Phoronidea. Bronn’s Klassen und Ordnungen des Tierreiches 4 (4). 1-183. Cowen, R. (1974). Respiration in metazoan evolution. Evolution 27. 69G701. CowIes, R. P. (19044. Origin and fate of the blood vessels and blood corpuscles of’tht. Actinotrocha. Zoologischer Anzeiger 27, 598-606. Cowles, R. P. (1904b).Origin and fate of the body-cavities and the nephridia of the Actinotrocha. Annals and Magazine of Natural History 14, 69-78. Dilly, P. N. (1972). The structure of the tentacles of Rhabdopleura compacta (Hemichordata) with special reference to neurociliary control. Zeitxhrift fiir Zellforschung und mikroskopische Anatomie 129, 2 6 3 9 . Dyster, F. D. (1859). Note on Phoronis hippocrepia. Transactions of the Linnean Society of London 22, 251-256. Emig, C. C. (1966). Anatomie et Bcologie de Phoronis psammophila Cori (Golfe de Marseille et environs; Etang de Berre). Recueil des Travaux de la Station marine d’Endoume 40, 161-248. Emig, C. C. (1967). Ultrastructure du tube digestif de Phoronis psammqhila Cori. I . Estomac (cellules cili6es). Archives de Zoologie expirimentale et ginhale 108, 617632.

THE 13IOLOGY OF PHORON1T)A

83

Emig, C. C. (1968).Ultrastructure du tube digestifde Phoronispsammophila Cori. 11. Pr&-estomacet intestin. Archives de Zoologie expirimentale et ginirale 109, 145155. Emig, C. C. (1969). Considerations sur la systematique des phoronidiens. VIII. Phoronis pallida (Schneider) Silen, 1952. IX. Phoronis ovalis Wright, 1856. Bulletin du Muse‘um d’Histoire naturelle de Paris 41, 1531-1542. Emig, C. C. (1971a). Taxonomie e t systematique des Phoronidiens. Bulktin, du Musium d’Histoire naturelle de Paris (Zoologie)8, 469-568. Emig, C. C. (1971b). Remarques sur la systematique des Phoronidea. X . Notes sur I’Bcologie, la morphologie e t la taxonomie de Phoronis ijimai et Phoronis vancouverensis. Marine Biology 8 , 154-159. Emig, C. C. (1972a). Phoronidiens r6colt6s lors de la campagne “Biapores” du N/O Jean Charcot ( 3 octobre-20 novembre 1971). Te‘thys 4, 423428. Emig, C. C. (1972b).Reproduction asexuee chez Phoronispsammoph,ila Cori. Marine Biology 13, 247-258. Emig, C. C. (19724. R&g&n&ration de la r6gion antkrieure de Phoronia psammophilu Cori (Phoronida). Zeitschrift fiir Morphologie der Tiere 73, 117-144. Emig, C. C. (1973a). Les processus de I’ontogenGse, compares B ceux de la rBg6nBration des Phoronida. Zeitschrift f u r Morphologie der Tiere 75, 329350. Emig, C. C. (197310). Ecologie des Phoronidiens. Bulletin d’Ecologie 4, 339-364. Emig, C . C. (1973~).L’histogenGse reg&n&ratricechez les Phoronidiens. Wilhelvn Roux Archiv 173, 235-248. Emig, C. C. (1974a). The systematics and evolution of the phylum Phoronida. Zeifschrift fur zoologische Systematik und Evolutionsforschung 12, 128-151. Emig, C. C. (197413).Observations et discussions sur le dbveloppement embryonnaire des Phoronida. Zeitschrift f u r Morphologie der Tiere 77, 317-335. Emig, C. C. (1975). Terminology in Phoronida. Zoological Scripta 4, 37-40. Emig, C. C. (1976a). Disposition e t delimination des cavitBs coelomiques chez les Phoronidiens (Lophophorates). Comptes Rendus de l’dcadimie des Sciences dp Paris 282, 1445-1447. Emig, C . C . (1976b). Phylogenkse des Phoronida, Les Lophophorates et le concept des Archimerata. Zeitschrift f u r zoologische Systematik und Evolutionsforschung 14, 1@24. Emig, C. C. (1976~).Le lophophore-structure significative des Lophophorates (Brachiopoda, Bryozoa, Phoronida). Zoological Scripta 5 , 133-137. Emig, C. C. (1977a). Un nouvel embranchement: les Lophophorates. Bulletin dP la Socidtl zoologique de France 102, 341-344. Emig, C. C. (1977b). Embryology of Phoronida. American Zoologist 17, 21-38. Emig, C. C. (1977~).Notes sur la localisation, 1’6cologie et la taxonomie des Phoronidiens. Tithys 7, 357-364. Emig, C. C. (1977d). Un nouvel aspirateur sous-marin, B air cornprime. Marine Biology 43, 37S380. Emig, C. C. (1977e). Sur la structure des parois de I’appareil circulatoire de Phoronis psammophila Cori (Phoronida, Lophophorata). Zoomorphologie 87, 147-1 53. Emig, C. C. (1979). “British and other Phoronids. Synopses of the British Fauna,” No. 13 (D. M. Kermack and R. S. K . Barnes, eds), 57 pp. Academic Press, London and New York. Emig, C. C. (1980). Phoronida. Fertilization, development and parental care. I n “Reproductive Biology of Invertebrates” (R. K. Adiyodi, ed.), Vol. 4, in press. Wiley; London.

Emig, C. C. and Bailey-Brock, J. H . (1980).Phylum Phoronida. I n “Reef and Shore Fauna of Hawaii” (in press). Emig, C. C. and Becherini, F . (1970). Influence des courants dans 1’6thologie alimentaire des phoronidiens. Etude par s6ries de photographies cycliques. Marine Biology 5, 239-244. Emig, C. C. and Emig, A. (1975).Etude des variations du nombre e t de la rkpartition des muscles longitudinaux chez Phoronis psammophila Cori. Jou.rnal of Experimental Marine Biology and Ecology 17, 23-31. Emig, C. C. and Lienhart, R. (1966).Un nouveau moyen de ri?colte pour les substrats meubles infralittoraux: I‘aspirateur sous-marin. Recueil des Travaux de la Station marine d’Endoume 42, 115-120. Emig, C. C. and Lienhart, R . (1971). Principe de l’aspirateur sous-marin automatique pour s6diments meubles. Eur. mar. biol. symp. Arcachon, 1968. Vie el Milieu suppl. 22, 573-578. Emig, C. C. and Siewing, R. (1975). The epistome of Phoronis psammcphila (Phoronida). Zoologischer Anzeiger 194,47-54. Emig, C. C. and Thouveny, Y. (1976). Absorption directe d’acide amin6 chez Phoronis psammophila. Marine Biology 37,69-73. Emig, C. C., Herberts, C. and Thomassin, B. A. (1972).Sur I‘association de Phoronis u.ustralis (Phoronida) avec Cerianthus maua (Ceriantharia) dans les zones recifales de Madagascar. Marine Biology 15,304-315. Emig, C. C., Boesch, D. F. and Rainer, S. (1977).Phoronidafrom Australia. Records of Australian Museum 30 (16), 455-474. Farmer, J. D. (1977).An adaptive model for the evolution of the ectoproct life cycle. In “Biology of Bryozoa” (R. M. Wollacott and R. L. Zinimer, eds), pp. 487-517. Academic Press, London and New York. Farmer, J. D., Valentine, J . W. and Cowen, R . (1973).Adaptive strategies leading to the ectoproct ground plan. Systematic Zoology 22, 233-239. Fenton, M. A. and Fenton, C. L. (1934). Scolithus as a fossil phoronid. American Geologist 61,341-348. Forneris, L.(1959). Phoronidea from Brazil. Boletim do Instituto OcCanographico, Sao Paul0 10(2), 1-105. Franzen, A. (1956). On spermiogenesis, morphology of the spermatozoon and biology of fertilization among invertebrates. Zoologiska Bidrag, Uppsala 31, 439-44 1. Franzen, A. (1977). Sperm structure with regard to fertilization, biology and phylogenetics. Verhandlungsbericht der deutschen zoologischen Gesellschaft 70, 123- 128. Ganguly, D. N. and Majumdar, N. (1967).On a new species of Phoronis (Phoronida) from India. Journal of the Zoological Society of India 19,7-1 1. Garlick, R. L., Williams, B. J. and Riggs, A. F. (1979). The hemoglobins of Phoronopsis viridis, of the primitive invertebrate phylum phoronida: characterization and subunit structure. Archives of Biochemistry and Physik lN,13-23. Gilchrist, J. D. (1907). New forms of the Hemichordata from South Africa. Transactions of the South African Philosophical Society 17,151-176. Gilmour, T. H . J. (1978). Ciliation and function of the food-collection and wasterejection organs of lophophorates. Canadian Journal of Zoology 56,2142-2155. Gilmour, T.H . J . (1979).Feeding in pterobranch hemichordates and the evolution of gill slits. Canadian Journal of Zoology 57, 1 1 3 6 1 142. Goodrich, E . S. (1903).On the body-cavities and nephridia of the actinotrocha larva. Quarterly Journal of Microscopical Science 47, 103-121.

Goodrich, E. S. (1945).The study of nephridia and genital ducts since 1895. Quarterly Journal of Microscopical Science 86, 113-392. Grobben, C. (1908).Die systematische Einteilung des Tierreiches. Verhandlungen drr zoologischen und botanischen Gesellschaft Wien 58. 491 -51 1 . Gutmann, W. F. and Bonik, K. (1978). Detaillierung der Acranier- und Enteropneusten- Modells. Senckenbergiana biologica 58, 325-363. Haderlie, E. C. and Donat 111,W. (1978).Wharf piling fauna and flora in Monterey Harbor, California. The Veliger 21, 45-69. Haswell, W. A. (1883). Preliminary note on an Aust,ralian species of Phoronis (Gephyrea tubicola). Proceedings of the Linnean Society of New South Wales 7. 606-608. Haswell, W. A. (1893). The occurrence of a second species of Phoronis in Port Jackson. Proceedings of the Linnean Society of New South Wales 7, 340-341. Hatschek, B. (1888). “Lehrbuch der Zoologie”. Jena. Herrmann, K . (1976).Untersuchungen uber Morphologie, Physiologie und Okologie der Metamorphose von Phoronis muelleri (Phoronida). Zoologischen Jahrbuchu, Anatomie 95, 354-426. Herrmann, K . (1977). Vergleichende Untersuchungen der postembryonalen Entwicklung hei Phoronis muelleri und Ph. psammophila. Verhandlungsbericht der deutschen zoologischen Gesellschaft 318. Herrmann, J . ( 1979). Larvalentwicklung und Metamorphose von Phoronis psammophila (Phoronida, Tentaculata). Helgolander wissenschaftliche Meeresuntersuchungen 32, 550-581. Hilton, W. A. (1930). A new Phoronopsis from California. Transactions of the American Microscopical Society 49, 154-159. Hiscock, K . and Howlett, R . M. (1975).The ecology of Caryophyllia smithi Stockes and Broderip on south-western coasts of the British Isles. I n “Underwater Research” (E. A. Drew, J. N. Lythgoe and J. D. Woods, eds), pp. 319-334. Academic Press, London and New York. Holthe, T. (1973).Borgensfjordens Polychaeta, Echiurida, Sipunculida, Priapulida, Phoronida og Enteropneusta. Unpublished th,esis, University of Trondh,eim, 190 PP. Holthe, T. (1977).A quantitative investigation of the level-bottom macrofauna of Trondheimsfjorden, Norway. Gunneria 28, 1-20. Hyman, L. H. (1958).The occurrence of chitin in the Lophophoratephyla. Biological Bulletin Marine Biological Laboratory Woods Hole 114, 106-1 12. Hyman, L. H. (1959).“The Invertebrates, Vol. 5 . Smaller Coelomate Groups.,” 783 pp. New York, McGraw-Hill. Ikeda, I. (1901).Observations on the development, structure and metamorphosis of Actinotrochu, Journal of the College of Science, Imperial University, Tokyo 13, 507-59 1 . Ikeda, I. (1903). On the development of the sexual organs and their products in Phoronis. Annotationes zoologicae Japonenses 3, 141-153. Ishikawa, H. (1977). Comparative studies on the thermal stability of animal ribosomal RNA’s. V. Tentaculata (Phoronids, moss-animals and lamp-shells). Comparative Biochemistry and Physiology 57B,9-14. Jagersten, G. (1972). “Evolution of the Metazoan Life Cycle. A Comprehensive Theory’, 283 pp. Academic Press, London and New York. Jebram, D. (1973).Ecological aspects of the phylogeny of the Bryozoa. Zeitschrift fiir zoologische Systematik und Evolutionforschung 11, 275-283. Jollie, M. (1973). The origin of the Chordates. Acta Zoologica 54, 81-100.

Josey, K. A. (1959). Probable cirripede, phoronid and echinoid burrows within a Cretaceous echinoid test. Paleontology 1, 397-400. Jubertie, C. and Manier, J . F. (1977).Etude ultrastructurale de la spermiogenke de Trogulus nepaeformis (Scopoli) Opilion, Palpatores. Annales des Sciencea naturelles (Zoologie) 19, 247- 260. Kawaguti, 8. and Nakamichi, A. (1973). Electron microscopy on muscle fibers in blood vessels of a phoronid worm. Biological Journal of Okanyama University 16, 73-82. Kume, N. (1953). Some observations on the fertilization and early development of Phoronis australis. Natural Science Report of Ochanomizu University 4, 253-256. Lebour, M. V . (1922). The food of plankton organisms. Journal of the Marine Biological Association, lJ.K. 12, 644-677. Lonoy, N. (1954).Comparative anatomical study on Phoronis ovalis Wright from Norwegian, Swedish and Brazilian waters. Universitetet Z Bergen Jrbok (naturvitenskapelig rekke) 2, 1-23. LpPvtrup, S. (1975).Validity of the Protostomia-Deuterostomia theory. Systematic Zoology 2496-108. McIntosh, W. C. (1888).Report on Phoronis buskii n. sp., dredged during the voyage of HMS Challenger, 1873-1876. Voyage of H M S Challenger (Zoology) 27 (75), 1-27. Mackinnon, D. J. and B i e n a t , G. (1970).The probable affinities of the trace fossil Diorygma atrypophilia. Lethaea 3, 163-172. McMurrich, J. P. (1910).Actinaria. I. Ceriantharia. Siboga Expedition 15 (53), 1-48. Marcus, E. (1949). Phoronis ovalis from Brazil. Zoologia, Sao Paulo 14, 157-170. Marsden, J. C. (1959). Phoronidea from the Pacific coast of North America. Canadian Journal of Zoology 37, 87-11 1 . Masterman, A. T. (1896). Preliminary note on the structure and affinities of Phoronis. Proceedings of the Royal Society of Edinburgh 21, 59-71. Masterman, A. T. (1898).On the theory of archimeric segmentation and its bearing upon the phyletic classification of the coelomata. Proceedings of the Royal Society of Edinburgh 22, 27Ck310. Masterman, A. T. (1900). On the Diplochorda. 111. The early development and anatomy of Phoronis buskii, McIntosh. Quarterly Journal of Microscopical Science 43, 373- 418. Meek, A. (1917).On the Phoronidea. Report of Dove Marine Laboratory, Cullercoats 6, 33-48. Menon, K . R . (1902). Notes on Actinotrocha. Quarterly Journal of Microscopical Science 45, 473-484. Metschnikoff, E. (1869). Ueber die Metamorphose einiger Seetiere (Cyphonautes. Mitraria, Actinotrocha). Nachrichten der Gesellschaft der Wissenschaft, Giittingerb. 232 -233. Metschnikoff, E . (1871). Ueber die Metamorphose einiger Seetiere. 111. LJeber Actinotrocha. Zeitschrift fur wissenschaftliche Zoologie 21, 244-252. Miiller, J . (1846).Bericht iiber einige Tierformen der Nordsee. Archiv f u r Anatomic und Physioloyie 13, 101-104. Nichols, D. (1967).The origin of Echinoderms. Symposium of the Zoological Society of London 20, 209-229. Nielsen, C. (1977).Phylogenetic considerations: the protostomian relationships. I I L “Biology of Bryozoans” ( R . M. Wollacott and R . L. Zimmer, eds), pp. 519-534. Academic Press, London and New York.

THE UIOLOGY OF PHOK.ONTDA

87

Ocharan, F. J. (1978).Primera cita para la Peninsula Iberica de Phoronis hippocrepia Wright 1856 (Phoronida).Rev. Facultad de Ciencias Universidad de Oviedo 17-19. 393-397. Ohuye, T. (1943). On the blood corpuscles and the blood formation in Phoronis. Scientifc Report of the University of Tokyo (Biology) 17, 167-185. Oka, A. (1897). Sur une nouvelle espkce japonaise de genre Phoronis. Annotation,rs zoologicae Japonenses, Tokyo 1, 147-150. Ollivier, J . S., Slattery, P. N., Hulberg, L. W. and Nybakken, J. W. (1977).Patterns of succession in benthic infaunal communities following dredging and dredged material disposal in Monterey Bay. U.S. Army waterways experiment station. Dredged material Research program. Technical report D, 77-27, 1-186. Peres, J. M. (1967). The Mediterranean benthos. Oceanography and M a r i m Biology, an Annual Review 5, 4 4 S 5 3 3 . Pixell, H . L. (1912). Two new species of Phoronidea from Vancouver Island. Quarterly Journal of Microscopical Science 58, 257-284. Pross, A. (1974).Untersuchungen uber die Muskulatur des Epistoms der Phoroniden (Phoronis ijimai Oka (Phoronidea)).Zoologischen Jahrbiicher, A n a t o m i ~92,391403. Pross, A. (1978). Motion of epistom in Phoronis ijimai (Phoronida). Zoologischrn Jahrbiicher, Anatomie 99, 54-58. Pourreau, C. (1979).Morphology, distribution and role of the epidermal gland cells in Phoronis pfiammophila Cori. Te‘thys 9, 133-136. Rattenbury, J. C. (1953). Reproduction in Phoronopsis viridis. The annual cycle in the gonads, maturation and fertilization of the ovum. Biological Bulletin Marine Biological Laboratory Woods Hole 104, 182-196. Remane, A. (1950). Die Enstehung der Metamerie der Wirbellosen, Zoologischer Anzeiger Suppl. 42, 1G23. Ronan, T. E. J r (1978).Food resources and the influence of spatial pattern on feeding in the phoronid Phoronopsis viridis. Biological Bulletin Marine Bioloyical Laboratory Woods Hole 154, 472- 484. Roule, L. (1896). Sur les mbtamorphoses larvaires de Ph,oronis sabatieri. Comptes Rendus de l’Acad6mie des Sciences de Paris 122, 1343-1345. Roule, L. (1900).Sur le dkveloppement embryonnaire des Phoronidiens. Annales dc Sciences naturelles 11, 51-247. Ryland, J. S. (1976). Physiology and ecology of marine Bryozoans. Advanzes in Marine Biology 14, 285-443. Scelzo, M. A. (1976). Sobre la presencia de Foronidos en aguas marinas argentinas (phylum Phoronida),. Physics, Seccion A , Buenos Aires 35 (91), 255-256. Schepotieff, A. (1906). Uber einige Actinotrochen der norwegischen Pjorde. Zeitschrift fur wissenschaftliche Zoologie 84, 7 S 9 4 . Schneider, A. (1862). Ueber die metamorphose der Actinotrocha branchiata. Archiv f u r Anatomie und Physiologie, 47-65. Schultz, E. (1903). Aus dem Gebiete der Regeneration. IV. Uher Regenerationserscheinungen bei Actinotrocha branchiate Muller. Zcit f u r wissenschaftliche Zoologie 75, 473-494. Selys-Longchamps, M. de (1903). Ueber Phoronis und Actinotrocha bei Hdgoland. Wissenschaftliche Meeresuntersuchungen, Helgoland 6,1-55. Selys-Longchamps, M. de (1904).D6veloppement post-embryonnaire et affinitks des Phoronis. Mdmoires (Sciences) de l’dcaddmie royale de Relgique 1, 1-150. Selys-Longchamps, M. de (1907). Phoronis. Fauna und Flora des Golfes von Neapd 30, 1-280.

Siewing, R. (1974).Morphologische Untersuchungen zum Archicoelomatenproblem. 2. Die Kijrpergliederung bei Phoronis muelleri de Selys-Longchamps (Phoronida). OntogeneseLarva-Metamorphose-Adultus. Zoologischen, Jahrbueher, Anatomie 93,275-318. Siewing, R. (1976).Probleme und neuere Erkenntnisse in der Gross-systematik der Wirbellosen. Verhandlungsbericht der deutschen zoologischen Gesellschaff 1976,5983. SilBn, L. (1944). The anatomy of Labiostom.eZla gisleni (B~yozoa Protocheilostomata), with special regard to the embryo chambers. KungZ. Svenska Vetenskaps Akademien Handl. 21, 1-1 1 1 . SilBn, L. (1952).Researches on Phoronidea of the Gullmar Fjord area (West coast, of Sweden). Archiv for Zoologi 4 (4),95-140. SilBn, L. (1954a).Developmental biology of Phoronidea of the Gullmar Fjord area (West coast of Sweden). Acta Zoologica 35,215-257. SilBn, L. (1954b). On the nervous system of Phoronis. Archiv for Zoologi 6, 1-40. Simon, J. L. and Dauer, D. M. (1977). Reestablishment of the benthic comrnunit,y following natural defamation. I n “Ecology of Marine Benthos” (B.C. Coull, ed.), pp. 139-154. Belle W. Baruch Library in Marine Science 6. Spaulding, M. H.(1906).Note on the occurrence of Phoronis larvae (Actinotrocha)a t Monterey Bay, California. Zoologischer Arbzeiger 30,184-185. Stancyk, S. E., Maturo, F. J. S. J r and Heard, R. W. J r (1976).Phoronids from the East coast of the United States. Bulletin of Marine Science 26.576584. Stephenson, D. G. (1976). On the origin of the five-fold pattern of echinoderms. Thalassia Jugoslavica 12,337-346. Steuer, A. (1933).Zur Fauna des Canal di Leine bei Rovigno. Thalassia Jugoslavica 1(4), 1-40. Storch, V. and Herrmann, K. (1978). Podocytes in the blood vessel linings of Phoronis muelleri (Phoronida, Tentaculata). Cell and Tissue Research 190,553556. Strathmann, R. (1973). Function of lateral cilia in suspension feeding of Lophophorates (Brachiopoda, Phoronida, Ectoprocta). Marine Biology 23, 12!+136. Strathmann, R . R. (1978).The evolution and loss of feeding larval stages of marine invertebrates. Evolution 32,894-906. Thomassin, B. A. and Emig, C . C . (1980). In preparation. Torrey, H. B. (1901). On Phoronis pacijica n. sp. Biological Bulletin, Marine Biological Laboratory Woods Hole 2,282-288. Trueman, E. R. (1968). The burrowing activities of Bivalves. Symposium of the Zoological Society of London 22, 167-186. Ulrich, W. (1972).Die Geschichte des Archicoelomatenbegriffs und die Archicoelomatennatur der Pogonophoren. Zeitschrift f u r .zoologische Systematik und Evolutionsforschung 10,301-320. Valentine, J. W. (1975). Adaptive strategy and the origin of grades and groundplans. American ZooZogist 15,391404. Vandermeulen, J. H. (1970). Functional morphology of the digestive tract epithelium in Phoronis vancouverensis Pixell: an ultrastructural and histochemical study. Journal of Morphology 130,271-286. Vandermeulen, J.H. and Reid, R. G. (1969).Digestive tract enzymes in Phoronida. Comparative Biochemistry and Physiology 28,443448.

Veillet, A. (1941). Description e t mecanismes de la m6taniorphose de la l a r w actinotroque de Phoronis sabatieri Roule. Bulletin dQl’lnstitut oceanographiqur dr Monaco 810, 1-1 1 . Vieitez, J. M. (1977). Primera cita para EspaAa de la especie Phoroiiispsa,mrriophila Cori. Boletin dela real Sociedad espaizola de Historia naLural (Biologics) 75,243-248. Vieitez, J. M. and Emig, C. C. (1979).Presencia de Phoronispallida (Schneidrr, 1862) (Phoronida, Lophophorata) en la costa atlantica espaiiola. TP’hys 9, 91-92. Virnstein, R. N. (1979). Predation on estuarine fauna. Response patterns of component species. Estuaries 2 , 69-86. Voigt, E. (1970). Foraminiferen und ( 1 ) Phoronidea als Kommensalen auf den Hartgriinden der Maastrichter Tuffireide. Paliio?Ltotogisch,eZeitschrift 44,86-92. Voigt, E. (1972). Uber Talpina ramosa v. Hagenov 1840, ein wahrscheinlich zu den Phoronidea gehoriger Bohrorganismus aus der Oberen Kreide, nebst Bemerkungen zu den ubrigen bisher beschribenen kretazischen “Talpina” Artjen. Nachrichten der Akademie der Wissenschaften in Gottinyen, I I . mathrwiatischphysikalische Klasse 7, 93-126. Voigt, E. (1975). Tunnelbaue rezenter und fossiler Phoronidea. Palaon.tologischr Zeitschrift 49, 135-167. Voigt, E . (1978). Phoronidenbaue (Talpina ramosa v. Hagenow), i ~ u s der Maastrichter Tufikreide. Publicaties van Het natuurhistorisch Genootschap it1 Limburg 28, 3-6. Webb, M. (1969). An evolutionary concept of some sessile and tubicolous animals, Sarsia 38, 1-8. Welsch, L. T. and Welsch, U. (1978). Histologische und electronenmikroskopische Untersuchungen an der praoralen Wimpergrube von Saccoglossus horsti (Hemichordata) und der Hatschekschen Grube von Branchiostoma lanceola,tum (Acrania).Ein Beitrag zur phylogenetischen Entwicklung der Adenohypophyse. Zoologischen Jahrbiicher, Ana.tomie 100, 564-578. Williams, J. B. (1960). Mouth and blastopore. Nature, London 187, 1132. Wilson, E. B. (1881). The origin and significance of the metamorphosis o f Actinotrocha. Quarterly Journal of Microscopical Science 21, 202-21 8. Wilson, R . A. and Webster, L. A. (1974). Protonephridia. Biological Reviews 49, 127 -160. Wright, T. 8. (1866). Description of two tubicular animals. Edinburgh New Philosophical Journal 4, 313. Zimmer, R. L. (1964). Reproductive biology and development of Phoronida. University Microfilm, Ann Arbor, 416 pp, Zimmer, R. L. (1967). The morphology and function of accessory reproductive glands in the lophophores of Phoron.is vaneouverensis and Phoronopsis harmrri. Journal of Morphology 121, 159-178. Zimmer, R. L. (1972). Structure and transfer of spermatozoa in Phoronopsis viridis. 30th annual Proceedings of the Electron Microscope Society of America, Los Arigeles Arceneaux, 2 pp. Zimmer, R. L. (1978). The comparative structure of the preoral hood coelorn in Phoronida and the fate of this cavity during and after metatnorphosis. In “Settlement and Metamorphosis of Marine Invertebrate Larvae” (F. S. Chia and M. E. Rice, eds), pp. 2 3 4 0 . Elsevier, New York. Zimmer, R. L. and Wollacott, R. M. (1977). Structure and classification of Gymnocoelomate larvae. In “Biology of Bryozoans” (R. M. Wollacott and R. L. Zimmer, eds), pp. 57-89. Academic Press, London and New York.

This Page Intentionally Left Blank

Ado. Mar. Biol., Vol. 19, 1982, pp. 91-132

CORAL COMMUNITIES AND THEIR MODIFICATIONS RELATIVE TO PAST AND PROSPECTIVE CENTRAL AMERICAN SEAWAYS

P. W. GLYNN Smithsonian Tropical Research Institute, A PO Miami 34002, U.S.A. I.

Introduction

..

,

.

11. Panamic Isthmian Setting

.

,

, ,

.. , . .. ., .,

., ., ..

..

.. ..

.. .. .. .. ..

.. .. ..

, . .. .. .. , .

, ,

., ..

A. Paleoecological background B. Character of extant reefs . .. .. , . , , C. Availability of colonists , . , . ., .. .. D. Access through the Panama Canal and the proposed inter-ocean seaway 111. Theoretical Considerations .. .. .. .. .. .. .. .. A. Attributes of good colonists . . .. .. .. , . .. .. B. Establishment in relation t o the biotic community , .. .. .. IV. Speculations on Some Potential Ecological Interactions . . .. .. .. A. Feeding relations , . , . ., , . ., , . , . .. ., B. Competition. .. .. .. .. .. .. .. .. .. C. Symbiosis . . , . .. .. ., .. , . ., .. , . D. Diseased organisms , . .. .. .. ., .. .. .. E. Biotic disturbance . ., ., .. .. .. , . .. .. V. Conclusions. , . , . ., ., , , , . .. .. , . , , VI. Acknowledgements .. ., .. .. .. , . ., ., . . VII. References . .. , . ., .. ., .. .. .. .. ..

The risk of adverse ecological consequences stemming from construction and operation of a sea-level Isthmian canal appears to be acceptable. (Atlantic-Pacific Interoceanic Canal Study Commission, Interoceanic Canal Studies, 1970, p. 62.) As an example of the worst thing that biologists might let slip by them, consider the possibility t h a t the Atlantic and Pacific biotas could be mingled by migration through the new Panamanian sealevel canal proposed for construction in the 1980s. (E. 0. Wilson and E . 0. Willis, Ecology and Evolution of Communities, 1975, p. 522.)

92 93 93 94 100 102 103 103 105 106 107 113 118 119 120 121 122 122

92

I. INTRODUCTION During the past century man’s activities have resulted in the accidental or intentional introduction of numerous exotic species in marine, brackish and freshwater environments. Some of these introductions have been beneficial, some neutral (as presently understood) and others clearly undesirable. Commercially important fin fish and shell fish species have been transplanted on a grand scale over many parts of the world. Many of these introductions have resulted in highly successful fisheries, providing oysters, salmonids, shad, etc. to new areas (Elton, 1958; Bardach et al., 1972). Exotic fin fish have also migrated to the Mediterranean through the Suez Canal and now contribute importantly to the eastern Mediterranean fisheries (Ben-Tuvia, 1966, 1978; Ben-Yami and Glaser, 1974). Unfortunately, such redistributions may also lead to the establishment of undesirable pest species and, thus, result in serious disruptions to assemblages of native species. The slipper limpet and oyster drill, gastropod molluscs native to the North American east coast, were accidentally introduced with oyster transplants to Europe and the Pacific coast of North America (Elton, 1958; Yonge, 1960). I n British and other northern European waters, the slipper limpet became a serious competitor for space in oyster beds and reduced the population size and even replaced native oysters. The American oyster drill is a serious predator of oysters, and, because of its relatively greater tolerance to cold winters than native oyster drills, has achieved a dominant status in English oyster communities. Construction of the Erie and Welland Canals permitted the establishment of the alewife and sea lamprey in the Great Lakes (Aron and Smith, 1971). These exotic species brought about a major disruption in the native Great Lakes fish fauna. The species interactions that have occurred since these introductions are complex and have at times been influenced by over-exploitation and pollution (Christie, 1974). One unfortunate result was a serious decline of large piscivores, such as the Atlantic salmon and the lake trout, brought about in part by the predatory sea lamprey. This reduction in salmon and trout has allowed an explosive increase in the alewife, a migratory marine herring that has largely replaced the lake herring. The lake herring in the past provided forage for desirable predators and was also a valuable commercial species. If sufficient attention had been focused on the possible consequences of the introductions noted above, these ecological problems might have been anticipated and avoided. The application

C'ORAI, C'OMMI'SITI ES

93

of quarantine and cleansing procedures can greatly restrict the movement of problematical species. Such screening measures would surely have prevented the introduction of the slipper limpet and oyster drill into European oyster communities. Measures t o restrict the movements of marine species (alewife and sea lamprey) into the Great Lakes could have prevented the serious problems t h a t occurred there as well. Although i t would be naive t o pretend t h a t all problem species can be identified and their movements prevented, this article will offer a tentative identification of some potentially troublesome species in American coral reef ecosystems if these Atlantic and Pacific biotas are allowed t o mingle through a sea-level canal. 11. PANAMIC ISTHMIAN SETTING A. Paleoecological background The isthmian* region of Central America is a significant biogeographic focal point because i t is here t h a t the last portal existed through which Atlantic and Pacific coral reef biotas mingled (Woodring, 1954; Durham and Allison, 1960; Newell, 1971). When the isthmus emerged i t isolated the eastern Pacific region from t h e Caribbean Sea and the Tethyan realm, the great tropical seaway in existence since the Triassic. A restriction of flow across Centrd America occurred by the early t o mid Miocene (Holcombe and Moore, 1977; Mullins and Neumann, 1978). It has generally been held that communication of marine species, via the Panama-Costa Rica and Bolivar Troughs, ceased sometime during the Pliocene epoch, between 1 million (Olsson, 1972) and 5.7 million (Emiliani et al., 1972) years ago. Changes in the direction of coiling of planktonic foraminifera from the Atlantic and the Indo-Pacific Oceans (Saito, 1976) and the major interchange of mammalian faunas between North and South America (Webb, 1976), indicate the emergence of the Isthmus of Panama, and complete intercontinental terrestrial connection 3.5 and 3.0 million years ago, respectively. The latest cohabitation of Pacific and Caribbean reef coral biotas presently known on the isthmus is from the Panama Formation of early Miocene age (Woodring, 1957). The latest outpost of a former panneotropical (Caribbean and eastern Pacific) coral fauna probably occurred in the Pliocene, as evidenced by the Caribbean genera present in the Imperial Formation at the head of the Gulf of *Isthmian, when used alone or otherwise unqualified, refers to both t h e t'aribbean and Pacific marine ecosystems of Panama

94

P.

w. ULY"

California (Vaughan, 1919; Durham and Allison, 1960). The existence of this assemblage was ephemeral for these corals became extinct shortly thereafter. Although post-Tethyan events are geologically recent, paleontological reconstructions in this region are limited by a meagre fossil record (Durham, 1966). Therefore, any interpretation of the evolutionary development of coral reef communities in Central America must] be viewed cautiously. During the Pliocene and Quaternary evidence indicates a general deterioration of the environmental conditions favouring reef corals, such as warm water and gradual eustatic sea-level changes (Wells, 1956; Durham and Allison, 1960; Newell, 1971). Contrasted with the Caribbean, which supported extensive coral reef development during the interstadial periods in the Pleistocene (Vaughan, 1919; Mesolella et al., 1970; James et al., 1971), and probably significant reef accretion during glacial periods (Macintyre, 1972), very few fossil reef deposits of comparable age are known from the eastern Pacific (Durham, 1980). It has been postulated that the eastern Pacific reef coral biota is a relict assemblage, surviving since its separation from the Caribbean Province after closure of the isthmian portal (McCoy and Heck, 1976; Heck and McCoy, 1978). However, Hubbs (1952, 1960) has marshalled evidence suggesting that tropical environments were severely reduced in size, if not entirely eliminated, in the eastern Pacific during Pleistocene glacial advances. Dana (1975) has suggested that eastern Pacific reef corals became extinct at such times, and extant corals and reefs in the eastern Pacific are derived largely from recent colonists from the central Pacific (Line Islands). The faunal affinity of eastern Pacific and central Pacific corals, the virtual absence of fossil Quaternary reefs on American Pacific shores, and the age of formation of extant eastern Pacific reefs (4000-6000 years B.P.; Glynn and Macintyre, 1977) are consistent with this view.

B. Character of extant reefs Many differences are evident in the general appearance of coral reefs on opposite shores of Central America. Extant Pacific reefs, which have formed during Holocene time (over the past 6000 years), are small (one to a few hectares) compared with Caribbean reefs which often cover tens to hundreds of hectares. I n addition, reef development in the eastern Pacific is attenuated in upwelling areas (Gulfs of Panama, Papagayo, Tehuantepec), along stretches of sand beach (southern Mexico to El Salvador; Springer, 1958) and near

CORAL COMMLINITIES

95

large river mouths and coastal areas of high rainfall (e.g. northwestern Colombia; Glynn, 1974a). A corollary of this fact is that Pacific coral reef habitats tend to be discontinuous and very limited in extent. Pacific coral reefs also show limited vertical framework construction (11-12 m maximum; Glynn and Macintyre, 1977) compared with Caribbean reefs, the largest of the latter having attained 33 m in thickness (Macintyre et al., 1977). Additionally, Pacific reefs are usually confined to protected (wave-sheltered) habitats and are restricted to shallow depths ( 1 H 5 m, as opposed to Caribbean reefs which show constructional activity to 60 m; Goreau and Goreau, 1973; Lang, 1974), particularly along Pacific continental coastlines (Porter, 1972; Dana, 1975; Glynn, 1976). As regards habitat diversity, zonation is limited on Pacific reefs, with only three to four zones present (Glynn, 1976); on well-developed Caribbean reefs, as many as 11 habitat zones are recognized (Goreau and Goreau, 1973). Submerged bank reefs, often present on insular and continental shelves in the Caribbean (Macintyre, 1968, 1972), have not yet been observed in the eastern Pacific. Thus, Caribbean reefs are much more varied and are formed under a greater range of conditions (i.e. on protected and exposed coasts and in shallow and deep environments) than those in the Pacific. Eastern Pacific reefs contain few frame-building species, usually ten or fewer scleractinian and hydrozoan corals, per reef (Fig. 1). Caribbean reefs often contain 30-50 hermatypic corals (and as high as 70 species in some areas) on a single reef (Fig. 2). Mangrove and sea grass communities, which commonly intermingle with Caribbean reefs, do not generally occur in reef habitats in the eastern Pacific. Except for Clipperton Island, the constructional contribution of crustose coralline algae appears to be less in the case of Pacific relative t o Caribbean reefs. This difference may be due, in part, to the fact that eastern Pacific reefs are generally absent from environments of high wave energy where coralline algae flourish (Adey, 1978; Dawson, 1966). A rich crustose coralline algal flora is present in the eastern Pacific (Earle, 1972; Silva, 1966; Taylor, 1945), and has undergone significant accretionary development in some areas, e.g. at Clipperton Island (Sachet, 1962), in the Galapagos Islands (Wellington, 1975), and in Colombia and Costa Rica (Glynn, personal observations): however, coral framework construction and algal buildups do not generally occur together. I n addition to the significant difference in number of reef-building corals, there is also a greater species representation in many diverse taxa on Caribbean as opposed to Pacific reefs. Absent or rare in

F I ~ 1. Windward upper slope zone ( 3 m d e p t h ) o f a n actiwl? atcreting Pacific coral reef corals are the only discernible macrobenthic species

Secas Islands Gulf of Chiriqui (24J u n e

1978) Pocilloporid

FK 2 Windward upper slope zone (7 m depth) of an activel) accreting Caribbean coral reef. Korbiski Island, San Blas Islands (19 August 1978) Scleractinian corals, hydrocorals And gorgonacean coelenterates are vivible

Eastern Pacific Physical Thermocline and high nutrient environment levels occur a t shallow depths. Upwelling areas and cool water currents are widespread. Tidal amplitude large (up to 6 m j and predictable. Turbidity high near coast with increasing clarity on offshore islands. Nature of reefs

Reefs small, one to a few ha in area. Reefs are discontinuous and relatively isolated (island-likej. Maximum Holocene framework thickness 12 m. Mean reef frame accumulation rates = 1.3-75 m 1000/yr Reefs form in sheltered habitats and are not well-consolidated. Framework construction occurs a t shallow depth, usually no deeper than 10-15 m; submerged reefs on insular and continental shelves are unknown.

-

Habitat diversity is low with three to four vertical zones per reef. Number of reef-building corals is low with eight to ten species per reef. Sea grass and mangrove communities are not present near coral reefs. Crustose coralline algae rarely form an algal ridge.

Caribbean Thermocline and high nutrient levels occur to lower depths. Upwelling areas and cool currents are limited in extent. Tidal amplitude small (c. 1 in j and unpredictable. Light penetration moderate to high.

Reefs large, from one to hundreds of ha in area. Reefs are widespread and often occur continuously over large geographic areas. Maximum Holocene framework thickness -33m. Mean reef frame accumulat,ion rates = 06-39 m 1000/yr Reefs form in low to high energy environments and are well-consolidated. Framework construction occurs in shallow and deep water, to 60 m depth; submerged bank or barrier reefs on insular and continental shelves are common. Habitat diversity is high, with up to 11 vertical zones per reef. Number of reef-building corals is high, with 30-50 species per reef. Sea grass and mangrove communities overlap extensively with coral communities. Crustose coralline algae commonly serve as cementing agents and often construct algal ridge features.

TARLF: 1 (cmt.) Eastern Pacific Ecological attributes

Impoverished with respect t o numerous sedentary taxa, i.e. calcareous green algae, large fleshy algae, sea grasses, large fleshy sponges, hermatypic corals, sclerosponges, gorgonaceans," actiniarians, zoanthids,b crinoids, colonial tunicates. Number of individuals per species is generally high. Corallivores are abundant with significant effects on coral growth and relative abundances of corals. Predation on motile, invertebrate corallivores is high. Sponge predators (cowries, asteroids, fishes) abundant and possibly responsible for low sponge cover on reefs. Competition for space among corals is intense. Branching corals contain protective symbiotic crustaceans. Pathogens are presently unknown in eastern Pacific coral communities. Organisms capable of binding calcareous skeletal remains are generally rare

Bioerosion is intense.

Caribbean With relatively rich and diversified sedentary biota.

Number of individuals per species is low to moderately high. Corallivores relatively insignificant. Predation on motile, invertebrate corallivores is low. Sponge predation is limited.

Competition for space between corals and other benthic taxa is intense. Branching corals without ( ? ) protective symbiotic crustaceans. Pathogenic blue-green algae and bacteria are reported in reef corals and associated fauna. Biotic binding agents (e.g., encrusting red and calcareous green algae, sea grasses, sponges, zoanthids, gorgonaceans) are abundant. Bioerosion is moderate.

"Gorgonians are abundant on Pulmo Reef in the Gulf of California (Squires, 1959; Brusca and Thomson, 1977). 'An extensive (several hundred m2) carpet of zoanthids was observed on a pocilloporid reef near Machalilla, Ecuador (1" 28's; 80"47'W) (Glynn, personal observations). Carlgren (1951) also noted abundant zoanthid (Palythm)coverage in the Gulf of California.

100

P. W (:LYNN

eastern Pacific reef environments are calcareous green algae, sea grasses, large fleshy sponges, and gorgonacean (horny corals), actinarian (sea anemones) and zoanthidean coelenterates, all of which are prominent on most Caribbean reefs. Whereas Pacific reefs can be fairly categorized as paucispecific, they are not sub-normal (relative to Caribbean and Indo-west Pacific reefs) in growth rate (Glynn, 1977; Glynn and Macintyre, 1977), in live hermatypic surface coverage or in the biomass of their associated biota (Glynn et al., 1972; Porter, 1974; Glynn, 1976). Consequently, the relatively few species often occur at higher population densities than do their Caribbean counterparts. The generally greater abundance of reef fishes in the Pacific, compared with the Caribbean, is probably related to the greater productivity resulting from the shallow distribution of nutrients in the upper layers of the sea and periodic upwelling in certain areas. These differences between isthmian coral reefs, notwithstanding the inevitable exceptions that will come to light, are summarized in Table I . Additional attributes that are discussed below are also included in Table I . C. Availability of colonists The distribution of extant coral reefs, and some noteworthy reef associates on opposite sides of Panama, were presented in synoptic form in 1972 by Glynn (1972) and Porter (1972). A t that time, structural coral reefs were indicated near the Caribbean entrance to the Panama Canal and in the Gulf of Chiriqui, about 360km southwest of the Pacific entrance (Fig. 3). More recent findings have shown that flourishing coral reefs are also present in several areas in the Gulf of Panama (Glynn and Stewart, 1973; Glynn and Macintyre, 1977). Small patch reefs occur in the Taboga Island group, 15 km south of the Pacific entrance to the canal. Porter (1972) lists 36 species of scleractinian and hydrozoan reef builders near the Caribbean canal terminus and eight species at Taboga Island on the Pacific side. The movement of coral reef species through the Suez Canal is not favoured because of the absence of coral reefs at either end of that canal (Por, 1971). It is obvious from the above that the situation in Panama is otherwise. Although live reef communities occur near the present canal, these communities are relatively impoverished when compared with reef assemblages elsewhere in Panama. For example, reefs in the San Blas area on the Caribbean coast comprise up to 50 species of coelenterate hermatypes (Porter, 1972), and in the Gulf ofchiriqui 15

C A NA L

C A R IB B E A N

E NT R A NC E

i:;;:;o;r;

PANAMA C I T Y

I

0

I

2

3

4KM

SCALE

PANAMA BAY

PACIFIC

C A N A L ENTRANCE

FIG.3. Location of coral reefs, volcanic rock substrata and mangroves in relation to Pacific a n d Caribbean entrances of present locks canal and proposed sea level canal routes 10 and 14. Distance scale applies to both maps.

102

P. W. GLYNN

species (Glynn et al., 1972; Porter, 1972). This trend is evident in other, but not all (Abele, 1976) taxa. Reef fishes, for example, are also represented by more species on coral reefs in the Gulf of Chiriqui than in the Gulf of Panama (Rosenblatt et al., 1972).

D. Access through the Panama Canal and the Proposed Inter-ocean Seaway Pacific and Caribbean coral reefs are presently separated by less than 100 km across the Central American isthmus. Nonetheless, Gatun Lake has served as an effective freshwater barrier since the Panama Canal opened in 1914 (Rubinoff, 1970; Jones and Dawson, 1973). Few documented cases of marine species moving through t h e Panama Canal waterway are known-nine fishes had transited the canal by 1971 (McCosker and Dawson, 1975).The saline waterway of the Suez Canal has permitted the migration of nearly 200 marine species in the 100-year period 1869-1970 (Aron and Smith, 1971). The colonization of species via ship’s bottoms (Menzies, 1968) is also a likely possibility. Following drought conditions on the isthmus in the early 1970s, a plan was considered to pump sea water into Gatun Lake to increase the supply of lockage water. Even minimal salinization (to 2-3%,) would greatly reduce the effectiveness of the Gatun barrier (McCoskerand Dawson, 1975).Another way the Gatun barrier might be circumvented is by transport of marine organisms (including their spores, cysts, larvae, etc.) in the ballast water of tankers (Chesher, 1968; Dawson, 1973). Increasing numbers of oil tankers are now dumping clean (i.e. chemically, but not biologically, clean), Caribbean water, up to 20-30000 tons each, in Parita Bay, Gulf of Panama (N. Smythe, personal communication). Given an inter-oceanic seaway across Central America and an array of potential coral reef colonists, there still remain numerous circumstances that could hinder or promote migrations. While no single set of physico-chemical parameters is optimal for all species, reef-dwelling organisms generally do best in shallow, warm, sun-lit waters of high (oceanic) salinity and adequate circulation (Wells, 1957; Stoddart, 1969a).A firm substratum and low sediment load are additional factors that favour survival of corals and associated species. The impoverished character of coral communities in the Gulf of Panama is probably due in large part to the seasonal upwelling regimen in this area. The increased turbidity in the dry season, due largely to the resuspension of mud-rich sediments, is believed to be a

CORAL COMMUNITIES

103

significant limiting condition for reef growth around the Caribbean canal entrance (Macintyre and Glynn, 1976). Certain conditions are likely to occur across a Panamanian seaway: (1) a mean Pacific sea level elevation 30cm higher on the Pacific than the Caribbean would result in net water movement into the Caribbean; (2) canal water will oscillate in response to changing tides; (3) most water transport would be from the Pacific to the Caribbean, but some small amounts of Caribbean water would probably also reach the Pacific (Sheffey, 1972). If mean volume transport from the Pacific to the Caribbean is 9.6 x lo7m3 (78000 acre-feet) per day, unrestricted flow (Sheffey, 1972), it is likely that, the temperature-salinity characteristics in a sea-level canal would approximate those surrounding the canal entrances. Sea surface temperatures of 23-28°C and salinities of 28-35%0, which delimit the mean long-term ranges on either side of the canal (Glynn, 1972), are not expected to interfere with the movement of most marine species. On the other hand, oscillating tidal currents may produce unfavourable conditions-high sediment transport and resuspension regimes, reduced light levels in the water column-for many coral reef species. Harleman (1972) calculated that a parcel of water would move from the Pacific to the Caribbean side of a sea-level canal in 2.5 days. Since the majority (80-85%) of tropical, shallow-water benthonic species produce planktotrophic larvae with pelagic phases of from 1 to 26 weeks (Thorson, 1950, 1961; Mileikovsky, 1971; Scheltema, 1977), it is reasonably certain that a mass transport of young stages will take place. The likelihood that adult benthonic reef species can establish “stepping-stone” populations in the proposed sea-level waterway is probably considerably less than the movement of species via currents. While numerous adult coral reef species (including Acanthaster and hermatypic corals, inter alia) can be maintained in aquaria throughout the year a t the present Pacific terminus, surviving seasonally low water temperatures (20-22°C) and salinities (l8-20%,) (Glynn, 1974a), it seems unlikely to me that they could establish viable populations under the high sedimentary regimes envisioned in or near a sea-level canal.

111. THEORETICAL CONSIDERATIONS

A. Attributes of good colonists Successful colonists often possess one or more of the following attributes: (1) a greater dispersal ability, (2) ability to produce a

104

P. W GLYNN

propagule, i.e. the minimal number of individuals of a species necessary to allow for continued reproduction (MacArthur and Wilson, 1967), ( 3 ) a high intrinsic rate of natural increase (e.g. early age at first reproduction, repeated reproduction, high fecundity and rapid development, Steams, 1976) and (4) a generalized behaviour and physiology (sensu Slobodkin, 1968) allowing for adjustments to new environmental circumstances. Many of these characteristics are shared by opportunists, i.e. species that prosper in temporary and unpredictable environments (MacArthur, 1972; Pianka, 1974). Vermeij ( 1 978) summarizes evidence indicating that many long-lived marine organisms, which often occupy constant and predictable environments, also have exceptional powers of dispersal. Because both opportunists and long-lived species are probably involved in colonization events, Vermeij suggested that the attributes of potential isthmian migrants can be best understood from an analysis of how migrating species from a donor source (e.g. Red Sea to Mediterranean or central Pacific to eastern Pacific) differ from species that have not migrated. With respect to propagule size (2, above), certain larval traits deserve further notice. A t least some planktonic larvae test surfaces and delay settlement and metamorphosis until an appropriate substratum is found (Crisp, 1974; Wilson, 1960). Further, larval settlement is often gregarious (aggregated), occurring among previously settled larvae or adults of the same species (Wilson, 1968). Such larval behaviours will allow freely dispersing species to settle on suitable surfaces in close enough proximity to ensure reproductive success. The capacity of invading species to tolerate new environmental conditions (an aspect of 4, above) was suggested by Por (1975),t o play an important role in the Lessepsian migration through the Suez Canal. Por believes that the dominant movement of species from the Red Sea to the eastern Mediterranean, is due in large part to a greater tolerance to physical stress of the northern Red Sea biota which evolved under harsh glacial episodes in the Pleistocene. I n contrast, the eastern Mediterranean biota, which is largely a product of mild interglacial periods, is not preadapted to cope with t h e physical rigors of the Suez waterway. In Panama, Pacific species inhabiting rigorous environments, e.g. the intertidal zone and upwelling areas, may have a greater tolerance and potential for establishment than Caribbean species. However, stressful conditions, such as unpredictable tidal exposures and periods of severe wave action, also occur frequently on the Caribbean reefs of Panama.

CORAL CQMh1I’SITlES

105

Organisms reproducing by means of planktonic larvae-probably the bulk of coral reef species-commonly possess adaptations correlated with good colonists ( 1 , 2 and 3, above). Therefore, i t seems reasonable to postulate that isthmian coral reef communities of both coasts offer a great potential for reciprocal colonization. If the eastern Pacific biota contains many elements that arrived by longdistance transport from the central Pacific (and I believe the evidence for this, referred to in part below, is very good), then one may assume that such species are predisposed to invade accessible and suitable Caribbean environments. Invasion, however, does not guarantee the successful establishment of a species. An invading population must also overcome a multitude of obstacles to insure colonization.

B. Establishment in relation to the biotic community Evidence, based on the colonizing and replacement-success t h a t has occurred between mammal and bird assemblages of different regions (Simpson, 1947; Darlington, 1957, 1965; Mayr, 1965; Patterson and Pascual, 1972), and marine fish faunas and other groups (Briggs, 1970, 1974), has been synthesized into a general zoogeographic theory, namely t h a t concerning the evolutionary centres of origin. This concept maintains that there exist evolutionary centres of high species diversity; from these, advanced, competitively dominant species disperse and tend to displace older established species occupying marginal areas. The general competitive ability of a biota is positively correlated with provincial diversity. Briggs (1969)applied this view to a reciprocal interoceanic migration scenario in Panama and predicted the mass extinction (one to 5000 species) in the relatively species-poor eastern Pacific region by the richer western Atlantic tropical biota. One must not lose sight of the important contribution of IndoWest Pacific species to east Pacific coral communities, however. Accordingly, from the aforegoing argument, many species of corals, hydrocorals, molluscs, crustaceans, echinoderms and fishes (Wells, 1978; Glynn et al., 1972; Emerson, 1978; Garth, 1974; Chesher, 1972; Rosenblatt et al., 1972 respectively), for example, might be expected t o immigrate and some possibly displace their competitively inferior counterparts in the western Atlantic region. Of course, this prediction must be tempered with the possibility that many IndoWest Pacific migrants became established in the eastern Pacific because of the impoverished nature of the coral communities in this region.

106

P.

w.

GI,Y?;h;

However, i t is by no means certain that community structure (species composition and relative abundances) is, indeed, largely controlled by competitive processes. I n reviewing mostly experimental evidence, Connell (1975, 1978) concluded that population densities are seldom great enough t o allow competitive displacements between species. Rather, a strong case is made for predation (including eating plants) and disturbance (physical and biotic) as important processes in controlling community structure. Environments subject to natural perturbations or disturbances by man often allow the entry of opportunistic species, i.e. species which share many of the characteristics of successful colonists. Wilson and Willis (1975) suggested that the variable Pacific environment, subject to seasonal upwelling events, would favour the evolution of a high proportion of opportunistic species that could insert themselves into the Caribbean biota. Birkeland’s (1977) analysis of Caribbean and Pacific colonizing species in Panama has, indeed, demonstrated that opportunists predominate in later stages of community development in the Gulf of Panama. Thus, one may envisage a plume of upwelled water extending through a sea-level canal into the Caribbean, and providing suitable conditions for the transit and at least temporary presence of exotic opportunistic species. This hypothesis argues for a biotic migration opposite t o that suggested by Briggs (1969). But Caribbean coral communities also have a multitude of opportunists that invade storm-ravaged reefs (Stoddart, 1969b), reef flat habitats experiencing unpredictable exposures and mass mortalities (Hendler, 1977),and other disturbed habitats. Therefore, it is probable that some opportunistic species would move and colonize in both directions with a somewhat stronger migratory component arising from the Pacific side.

IV. SPECULATIONS ON SOME POTENTIAL ECOLOGICAL INTERACTIONS I t has been argued thus far that potential coral reef colonists occur on both sides of the isthmus near the preferred sea-level canal routes, that many (if not the majority) of these species could transit a sea-level waterway lacking an effective biotic barrier, and that isthmian coral reef communities differ on a variety of levels, e.g. taxonomic affinity, relative abundances of major guilds and ecological processes influencing community structure. These postulates form the basis of a challenging task: to predict the possible

CORAL COMMUNITIES

107

outcome of the reunion of reef biotas that have been separated for millions of years. A range of interactions and effects of certain coral reef species have been identified, at least in general terms, during the past two decades. From this largely qualitative knowledge, some speculations are offered on possible outcomes of a variety of potential Caribbeanpacific species interactions.

A. Feeding relations Clearly one of the more immediate results that can be expected, in the event of open access between the two oceans, will involve feeding interactions between invading and native species. Predation alone, recalling the dramatic effects of the Crown-of-Thorns sea star, Acanthaster, has been demonstrated to disrupt coral reef ecosystems over great areas (Endean and Stablum, 1973; Vine, 1973; Ormond and Campbell, 1974). Acanthaster has caused local extinctions of corals on some reefs with severe repercussions on the variety of life depending directly or indirectly on live coral. Several animals are now known to feed directly on the tissues of live, reef-building scleractinian and hydrozoan corals (Robertson, 1970; Glynn, 1973; Randall, 1974). Animals having apparently no or only slightly harmful effects on their coral hosts, such as species engaged in capturing food from corals or feeding on mucus and detritus (Patton, 1976), are not considered here. An inventory of the presently known corallivores of American coral reefs is presented in Table 11. Note that the majority of these corallivores (those with asterisks), 26 out of a total of 38 species, have been observed on Pacific or Caribbean reefs in Panama. Many of these corallivores, e.g. Hermodice, Coralliophila abbreviata and Acanthaster, feed on a variety of coral species, or, in the case of Mithrax, Diadema, damselfishes and parrotfishes, on other kinds of organisms as well. These species, with their generalist-type diets (an important attribute of successful colonists), could probably feed on a variety of novel prey in natural habitats [cf. next paragraph]. Preliminary observations on individual corals in aquaria and on coral patches maintained in large laboratory tanks have shown that all corallivores, thus far tested, will consume novel coral prey. (Novel species are here defined as organisms that are not presently members of a particular biota.) The ability of a Pacific corallivore, the gastropod Jenneria, to feed and reproduce on a diet of Caribbean corals was first demonstrated by D'Asaro (1969) and corroborated

TABLE 11. INVENTORY OF AMERICAN CORAI.I.IVORES Ksows TO FEED ON SCLERACTISIAN AND HYDROZOAS HER MA TYPE^ The coral species involved follow the enumeration scheme given at the bottom of the table. Corallivores observed in the isthmian faunas are indicated by asterisks. Note that key numbers of Pacific species are in boldface. Pammic Pam& Taxonmic group

Corallivore species

Coral prey

Polychaetous annelid

Gastropod molluscs

*Jenneria pustulata (Lightfoot)"

8, 13 (D'Asaro, 1969)b 4, 12, 14 (personal observation) 2 (Glynn et al., 1972)

Muricopsrs zeteka Hertlein and Strong 4 (Wellington, 1975)

Lutiaxis (Babelcnnurex) hindsii Carpenter

Caribbean- West Indian Corallivore species

Coral prey

*Hermodice earunculata (Pallas)"

10, 11 (Marsden, 1962, 1963) 2 (Glgnn, 1962) 10, 11, 14, 15, 16, 20, 27 (Ott and Lewis, 1972) 25 (D. R. Robertson, personal communication) 3 (Antonius, 1976; Shinn, 1976) 2*, 12 (personal observation)

*Corallawphila abbreointa (Lamarck)

20 (Ward, 1965) 2,3,4,7,9,10,11, 14, 15, 16, 17, 20 (Miller, 1970)' 2, 3, 4 , 8, 15, 16, 20, 24 (Ott and Lewis, 1972) 18 (personal observation)

*Coralliqphilo (caribbaea)Abbott

2, 5, 9, 10, 1 1 . 24 (Miller, 1970)' 3 (J. C. Lang. penonal communication)

*Quoyula madreporarum (Sowerby) Aeolid nudibranch

Crustaceans

Echinoderms

*Trimpagurus magnifeus (Bouvier) *Aniculus elegans Stimpson *Acanthaster plnnci (LiMaeUS)d

*Pharia pyramiduta (Gray) *Nidorellia a r m t a (Gray) *Eueidaris thomrsii (Valenciennes)

2 (Glynn et al.,

1972) * Phestilla sp. 8 . 10 (R. C . highsmith, personal communication)

Calliostoma javanieum Lamarck

6,23' (Lang, 1970) 4b (Miller, 1970)

*Mithrax sculptus (Lamarck)

11, 12 (Glynn,1975)

Oreaster reticulatus (Linnaeus)

3,20 (L.Buss, personal communication)b

*Diadems antillarum Philippi

1, 2,3,4,20,27 (Bak and van Eys, 1975) 19 (Dana, 1970)

2 (Glynn et al., 1972) Variety of Pacific hermatypic scleractinians and hy drocorals' (Barham et al., 1973;Dana and Wolfson, 1970; Glynn, 1974a,1976; Porter, 1972) 4, 10, 12 (cited in Porter, 1972)b 3 (Dana and Wolfson, 1970) 4 (Galapagos Iss., personal observation) 2. 4 (Glynn eta!., 1979)

TABLE I1 (cont.) Panamic Pacijic

Caribbean- West Indian

Taxonomic P U P

Corallivore species

Coral prey

Corallivore species

Coral prey

FISHES Spadefish Butterfly fishes and Angelfishes

Damselfishes

*Chaetcdipterus faber (Broussonet) *Chaetodm eapistratus Linnaeus

*Microspathodm chrysurus (Cuvier and Valenciennes)

* E u p m e n t r u s planifrons (Cuvier and Valenciennes)

*Eupornncentrus dmsqpunaeans (Poey)

22 (Randall, 1967) 8, 20 ( L . Buss, personal communication)

Bohlke and Chaplin ( 1968) Note that western Atlantic Chaetodontidae browse on reef- building corals 25 eaten by juveniles (Ciardelli, 1967; Glynn, 1973) 2, 13 eaten by juveniles (D. R. Robertson, personal communication) 8, 25 eaten by adults (Glynn, 1973) 3, 1 1 , 18, 20 (Kaufman, 1977) 25 eaten by juveniles (D. R. Robertson, personal communication) 25 eaten by juveniles (D. R. Robertson, personal communica.tion)

Parrotfishes

*Scarus ghobban Forsskll

6 (Glynn et al., 1972)

*Scarus perrico Jordan and Gilbert

3 (Glynn et al., 1972)

Unidentified fragments (Randall, 1967)

Scarus coelestinus Cuvier and Valenciennes *Scarus gwmmaia Cuvier

*Scarus croicensis Bloch Scarus vetula Bloch and Schneider S p a r i s m viride (Bonnaterre) Scarus taeniopterus Desrnarest Sparisoma aurofrenatum (Cuvier and Valenciennes) Alutera scripta (Osbeck) Cantherhines macrocerus (Hollard) Cantherhines pullus (Ranzani)

Filefishes

Puffers

*Arothrm hispidus (Linnaeus) *Arothron meleagris (LacBpide) *Canthigaster amboinemis (Bleeker)

II

8 and probably 2, 3, 4, 12, 20 (Bakus, 1969; Glynn, 1973) 3 (J.Ogden, personal communication) 1, 4, 8, 10, 11, 20, 21, 27 (Frydl, 1977) 1, 4, 8, 10, 11, 20, 21, 27 (Frydl, 1977) 10 (Gygi, 1975)

1, 11 (Frydl, 1977)

26 (Randall, 1967) Unidentified scleractinians (Randall, 1967)

1 , 2 (Glynn et al., 1972, unpublished data) 1 , 5 , 7 (Glynn et al., 1972, unpublished data) Unidentified scleractinians (Hobson, 1974)

Pacific corals: 1-Psammoeora (Stephanuria)stellata (Verrill): 2-Pocilloporid species, including Pocilloporu damicornis (Linnaeus), P. capitata Verrill, P . lacera Verrill, P . robusta Verrill; *Poeillopora sp.; 4--Patma clavzu Dana; 5-Pavma sp.; G P o r i t e s panamensis Venill; 7-Porites californica Verrill: 8-Porites lobata Dana.

TABLE 11 (cont.) Caribbean corals and hydrocorals: 1-Madracis mirabilis (Duchassaingand Michelotti); 2-Acrqpma palmata (Lamarck);3-Acrqpora cervicornis (Lamarck);4-Agaricia agaricites (Linnaeus);5-Agaricia sp.; 6-Agaricia spp.; 7-Helioseris cucullata (Ellis and Solander);8-~’iderastrea siderea (Ellis and Solander); 9-8iderastrea radians (Pallas);1 G P m i t e s astreoides Lesueur; 1 1-Porites porites (Pallas); 12-Porites furcata Lamarck; 13Purites sp.; l4-Favia fragum (Esper); 15-Diploria divosa (Ellis and Solander); 1 6 D i p l m i a strigosa (Dana); 17-Diplmia labyrinthiformis (Linnaeus); 18- Colpqphyllia natans (Miiller); lZ)--Colpqphyllia sp.; 2&Mmstrea annulan’s (Ellis and Solander); 21-Montastrea cuvernosa (Linnaeus);2 2 4 c d i m diffusa Larnarck; 23-Mussa angulosa (Pallas);24-Mycetophyllia lamarckana Milne-Edwardsand Haime; 25-Millepora wmplanatu (Lamarck); 2 6 M i l l e p o r a alcicmnis (Linnaeus): 27-Millepora sp. “Species with asterisks observed in Panama. Laboratory observations. Suggested to play a parasitic role. *Acunthaster ellisii (Gray) is believed to be a junior synonym of A . planci. See Glynn (1974a). ‘Species lists are available in Glynn et al. (1972) and Porter (1972).

CORAL ('OMM~'N11'IES

113

and extended to include additional novel coral species (personal observations). Acanthaster has been shown to feed on all Caribbean corals thus far offered (five species).On a small coral patch containing both Pacific and Caribbean corals, Acanthaster fed indiscriminately on all species except the Caribbean coral Porites furcata, which was eaten last; the sea star remained on the patch until every colony was consumed (Fig. 4 ) .The Caribbean polychaete Hermodice, which often feeds on thick (up to 1-5em), branching Caribbean corals (Porites furcata, P. porites and Acropora cervicornis), feeds avidly on Pacific species of Pocillopora, whose colonies are formed of branches less than 1.5cm thick (Fig. 5). In addition, the Caribbean Coralliophila abbreviata is being maintained in our laboratory on a diet of Pacific Pocillopora. The damage caused by many of these corallivores in their native surroundings does not appear to be excessive. However, in the face of an invasion, where native corals may lack adequate defensive mechanisms to cope with exotic corallivores or where predation or competition on corallivores is relaxed, i t is possible that their effects could take on a new destructive dimension. For example, the low rates of predation by fishes on the echinoid Eucidaris in the Galapagos Islands, compared with mainland eastern Pacific populations, has been hypothesized to explain the high population densities of the sea urchin in the Galapagos and its limiting effect on reef development there (Glynn et al., 1979). While it is not possible to consider here other aspects of feeding ecology in coral communities, it must be emphasized that interactions of this genre can have a wide and complex range of effects. Disruptions in coevolved antipredatory adaptations such as those involving skeletal architecture (Vermeij, 1974, 1977), the utilization of noxious and toxic substances (Bakus, 1969; Bakus and Green, 1974; Fenical, 1975), crypsis, aposematism and mimicry (Wickler, 1968; Rubinoff and Kropach, 1970;Edmunds, 1974) are all conceivable effects that could result from a biotic interchange. B. Competition Interspecific competition, a mutual interference interaction between species due to the utilization of common and limiting resources, is an omnipresent and potent ecological factor. The mechanisms of competition are diverse and may be broadly classified into exploitation and interference modes (Miller, 1967). Interference competition, exemplified in sedentary species by allelochemical

FIG.4. Acanthasterplanci feeding in a mixed patch ofCaribbean and Pacific corals. These animals were maintained in isolation in a 115 kl(30 400gal.) tank a t Naos Island. Arrows denote Pacific (Pocillvpora damicornis, (1); Puwona gigantea, ( 2 ) )and Caribbean (Agaricia agaricites, ( 3 ) )corals eaten by the sea stan.

('ORAL ('OMMUNITIES

115

FIG.5. Hermodice

mruncuZatu, a Caribbean polychaete worm, feeding (in isolation) on the Pacific coral Pocilbpora dumicornis. Arrow points to head end of worm over a branch-tip of the coral. Tissues have been stripped by the worm from the three branch-tips at topcentre.

effects, extracoelenteric digestion and overgrowth responses, in'volves a direct interaction whereby one species denies another access to a requisite limiting resource. Exploitative competition is an indirect interaction whose outcome depends on the relative efficiencies of species in utilizing a mutually accessible but limiting resource. A fast-growing species that can gain access quickly to some required minimum amount of space, sunlight or plankton-rich currents may successfully exploit these resources at the expense of its slower growing neighbours. Reef-building corals compete for space by rapid growth and overtopping (exploitative competition) and by extracoelenteric digestion and overgrowth through direct contact (interference competition), Pocilloporid corals, in part because of a relatively high growth rate ( 3 . M . O cm/yr; Glynn, 1977), predominate in the eastern Pacific where they form monogeneric reefs in many areas (Galapagos Islands, mainland Ecuador, Colombia, Panama, Costa Rica and Mexico). Caribbean acroporid corals, which occupy extensive areas in shallow reef zones, are faster growing species than eastern Pacific

116

P. W. GLYNN

corals. Acropora palmata and A . cervicornis grow linearly from 6 to 10 cm/yr (Gladfelter and Monahan, 1977) and 14-27 cm/yr (Lewis et al., 1968), respectively. If linear growth differences should prove to be important, then it is possible that Acropora corals would, to a large extent, replace pocilloporid corals on Pacific reefs in the event of interoceanic access. I n addition, two acroporid species, Acropora palmata and A . prolifera, have a spreading growth habit which frequently leads to overtopping and the death of adjacent slowgrowing or prostrate corals. Another relevant factor involves extracoelenteric feeding, whereby corals growing in close proximity extrude mesenterial filaments and digest and kill their heterospecific neighbours. Such interactions in the Caribbean, where they were studied extensively by Lang (1973), are hierarchical with small, slow-growing species ranking highest in ability to injure neighbouring corals. Thus, if dominant and subordinate juvenile corals grow side by side, the subordinate species could be eliminated in spite of an advantage in growth rate or pattern. Recent observations by Sheppard (1979) in the Indian Ocean indicate that predictions involving the competitive outcomes between Pacific and Caribbean corals will not be a straightforward task. Contrary to the situation in the Caribbean, where dominant members of a hierarchy are present in all reef habitats, but constitute relatively minor components of the reef, many of the highest ranking corals studied in the Indian Ocean are the dominant members of certain reef zones (see also Connell, 1976). Additionally, the capacity of Indo-Pacific corals to damage adjacent species through extracoelenteric feeding is not clearly related to their morphology or taxonomic position (Sheppard, 1979). A further complication arises from the discovery of Wellington (1980) that previously established hierarchies in the eastern Pacific (Glynn, 1974b) can be reversed over longer periods of time. For example, eastern Pacific Pocillopora spp. typically develop elongated sweeper tentacles that eventually (7-60 days) kill initially dominant pavonid species. The recent discovery of Pocillopora in Caribbean fossil deposits of late Pleistocene age (60000-120 000 years B . P . ) indicates that this coral genus survived on Caribbean reefs until relatively recently (Geister, 1977). Paleo-ecological analysis of former habitat conditions (shallow, protected, backreef or lagoonal environments) and species composition of the scleractinian assemblage led Geister to conclude that Pocillopora was competitively inferior to contemporaneous Caribbean coral taxa. This conclusion is compatible with some

CORAL COMMLTNITIES

117

of the known competitive attributes of Pocillopora compared with Caribbean species, i.e. (1) lower growth rate (with respect to Acropora), (2) non-overtopping colony form and (3) poor space competitor, at least in parts of the western Pacific (Maragos, 1972; Connell, 1976). On the other hand, the development of sweeper tentacles in Pocillopora might give these corals an advantage in close encounters. (Some Caribbean corals have sweeper tentacles also; see Richardson et al., 1979). Finally, considering the apparent low abundance of Pocillopora before its extinction in the Caribbean, one cannot discount the possible important effects of predation or disease on remnant populations. While corals often predominate in shallow, well-illuminated zones, reef surfaces typically contain diverse taxa that frequently compete directly with corals for space under these conditions. For example, reef-building corals are commonly overgrown by algae (Birkeland, 1977; Connell, 1973; Glynn, 1973),foraminiferans (Bak et aE., 1977), sponges (Riitzler, 1971, 1972; Glynn, 1973), other coelenterates, such as sea anemones (Ott, 1975; Sebens, 1976), zoanthideans (Glynn, 1973) and gorgonaceans (Kinzie, 1970; Glynn, 1973),and tunicates (Bak et al., 1977). Recalling the low abundances or virtual absence of such groups on eastern Pacific coral reefs, it is possible that non-coral, benthic Caribbean elements could acquire significant space in such habitats if allowed access to them. Competitive outcomes can also be influenced by the feeding activities of animals not directly involved in competition a t that trophic level. Experimental studies have shown that when browsers and grazers are excluded, the community composition of reef surfaces often changes dramatically. The constant cropping of algae by herbivorous fishes allows other benthic organisms to compete and occupy surfaces that would not otherwise be available to them (Stephenson and Searles, 1960; Randall, 1965; Bakus, 1969; Wanders, 1977). Many fishes selectively graze on algae, in some cases avoiding juvenile corals (Birkeland, 1977), and, thereby, prevent rapidly growing plants from monopolizing available substrata. Areas supporting high population densities of sea urchins, which efficiently crop fleshy algae, often contain sparse fleshy algal cover (Sammarco et al., 1974; Benayahu and Loya, 1977),have a high surface coverage of crustose coralline algae (Van den Hoek et al., 1975; Adey and Vassar, 1975), and are probably better suited for the recruitment of corals and other animals that compete with fleshy algae (Dart, 1972). There are also animal feeding behaviours that encourage algal growth, such as the algal gardens defended by damselfishes (Brawley

118

P. W. GLYNN

and Adey, 1977).Kaufman (1977)hypothesized that the competitive outcome between two coral species in the Caribbean (Montastrea annularis and Acropora cervicornis) can depend on the territorial feeding behaviour of a damselfish. I n killing coral to extend their algal gardens, the damselfish-induced mortality of corals has a greater effect on Montastrea than on Acropora. Montastrea regenerates slowly and is subject to a high rate of invasion by boring organisms, thus allowing Acropora to increase and monopolize certain reef zones. It is, therefore, necessary to consider possible competitive interactions between Pacific and Caribbean species in relation to predation and, probably, other ecologic processes (see below). It is conceivable that the migration of exotic corallivores and sponge feeders, abundant in the eastern Pacific compared with the Caribbean (Glynn, 1972), into the Caribbean could reduce coral and sponge cover and alter competitive interactions to the advantage of presently minor groups of solitary organisms (Jackson, 1977). C. Symbiosis Symbiosis, here defined as an intimate and mutually beneficial association between species, is a commonly encountered relationship on coral reefs. Crabs and shrimps living as obligate symbionts on Pacific pocilloporid corals, from which they obtain shelter and nutriment, offer protection to their coral hosts by repulsing the corallivore Acanthaster (Pearson and Endean, 1969; Weber and Woodhead, 1970; Glynn, 1976). Large coral colonies containing agonistic crustacean symbionts are virtually immune from predation because the crabs pinch and pluck at the tube feet and the shrimps grip spines and snap explosively at sea stars attempting to mount the coral. Some Caribbean corals also harbour symbiotic crustaceans (e.g. the crab Domecia on Acropora palmata), but it is not known if these act to protect their hosts from novel corallivores. In the case of pocilloporid corals and their crustacean symbionts, which have evolved in the presence of Acanthuster, the selective advantage in protecting an essential resource is obvious. Caribbean species of Acropora, however, are not attacked by corallivores that destroy the entire colony in a natural situation, and it is, therefore, unlikely that a strong defensive behaviour has evolved in this particular coral-crab partnership. Thus, if invading pocilloporid symbionts cannot adapt to Caribbean acroporid corals (and they do not live on sympatric Indo-Pacific acroporids),the latter would be vulnerable to attacks by

CORAL COMMUNITIES

119

Acanthaster. I n fact, Indo-Pacific Acropora species are a preferred food of Acunthaster (Goreau et al., 1972; Endean and Stablum, 1973; Laxton, 1974; Ormond et al., 1976). Laxton (1974) speculated that Acanthaster could seriously damage western Atlantic reefs where Acropora is an important frame-building species. Cleaning symbiosis, in which certain reef fishes allow other fishes or shrimps to clean their bodies of unwanted food, parasites or diseased tissues, is a highly coevolved relationship on coral reefs (Feder, 1966;Ehrlich, 1975).Cleaner stations, focal points of high fish abundance and diversity on reefs (Slobodkin and Fishelson, 1974), are visited by a variety of fish hosts which recognize cleaners by their colour patterns (usually conspicuous) and invitational displays. It is possible that invading naive hosts may not recognize new cleaner species, or may prey intensively on native cleaners in non-cleaning situations (Hobson, 197 1 ), thus disrupting this highly coevolved interaction. However, aquarium observations have demonstrated that some naive hosts can quickly adjust their behaviours to accommodate novel cleaners (G. Barlow in Feder, 1966). Fish predators also interfere with cleaning symbiosis by frightening away the cleanees (Potts, 1973). The introduction of novel predators that may eat cleaners, such as sharks, barracuda, jacks, groupers and snappers, could conceivably affect this system. D . Diseased organisms Considering the significant effect of pathogenic organisms on species populations and communities, it is unfortunate that our knowledge of such interactions in the marine environment is so limited. While two serious epidemics in the sea have been well documented-the devastation of eel grass communities on both sides of the Atlantic in the 1930s (Hopkins, 1957), and of commercial sponges in the Gulf of Mexico and Caribbean in the 1930s and 1940s (Storr, 1964)- the ultimate processes responsible for these upheavals have never been satisfactorily explained. Since the mid- 1970s, several reports have appeared on the incidence of diseases in reef organisms (in sponges, Antonius, 1977; in alcyonarian coelenterates, Antonius, 1977; Morse et al., 1977), especially in reef-building corals. These diseases are widespread in the western Atlantic, having been observed in corals in Bermuda (Garrett and Ducklow, 1975), Florida (Antonius, 1976,1977;Dustan, 1977),the Virgin Islands (Gladfelter, personal communication) and Panama (Glynn, personal observations). The pathogens thus far identified are blue-green algae

120

P. W. GLYNN

(Antonius, 1976, 1977) and bacteria (Garrett and Ducklow, 1975). The diseased corals are often, but not invariably, living under disturbed or stressful conditions. In spite of a recent emphasis on reef studies in this region, no diseased organisms have yet been reported from eastern Pacific reefs. If Pacific species have not been exposed to the disease organisms of the Caribbean, and do not possess protective mechanisms against such forms, the spread of pathogens through a sea-level canal could be disastrous to eastern Pacific coral communities.

E. Biotic disturbance Just as there exists a suite of organisms that bind, cement and otherwise stabilize the reef frame and surrounding sediments, reef communities also contain a variety of burrowing and boring organisms whose activities are destructive to various degrees. Some important binding organisms present in the Caribbean, e.g. calcareous green algae, sea grasses, sponges, zoanthids and gorgonians, are absent or rare in the eastern Pacific. This may explain in part the fragility of Pacific pocilloporid reefs and their usual development in sheltered areas, even outside of the hurricane belt. The insertion of some of these stabilizing Caribbean forms (e.g. see Wulff and Buss, 1979 for evidence on the binding capacity of sponges) into eastern Pacific coral reef communities could conceivably enhance their integrity and allow their expansion into a greater range of habitats. Bioerosion is intense in isthmian coral communities with endolithic algae, sponges, polychaetous annelids, sipunculans, cirripeds, molluscs and echinoids contributing substantially. Highsmith ( 1980) has demonstrated a significant relationship between intensity of boring and primary productivity, which was greater in the eastern Pacific than in the Caribbean. Fishes also appear to have a greater effect on Pacific coral communities as compared with those of the Caribbean (Glynn, 1972, 1973; Bakus, 1969). Pacific puffers and parrotfishes feed directly on live corals, a triggerfish causes extensive damage to massive corals (in the process of extracting boring bivalves) and a jack overturns rubble and corals along the edge of reefs (in search of crustaceans). While much coral is killed in this way, these activities also enhance the asexual propagation and lateral extension of reef communities. Toxopnuestes roseus, a sea urchin present only on Pacific reefs, also plays a significant role in moving coral rubble. The die1

CORAL COMMVNITIES

121

movements (nightly surfacing and daily burial) of large populations of this urchin through rubble sediments topples young corals and commonly churns them into the bottom where they die (personal observations). Moreover, coral recruits are rasped from rubble surfaces by these foraging urchins. These activities of Pacific urchins and fishes, in combination with their cropping of potential binding agents (algae, sponges, etc.), could generate a de-stabilizing effect in Caribbean reef communities. These, and other properties related to species interactions on Pacific and Caribbean reefs, are contrasted in Table I.

V. CONCLUSIONS It is evident that present knowledge permitting insight into the kinds of ecologic interactions that might occur, should isthmian coral reef biotas merge, is decidedly fragmentary. It is also clear that the possible levels and variety of effects are enormous. This creates a dilemma as to the kinds, extent and emphasis of research that should be devoted to this problem. From the standpoint of ecological effects, and without consideration for effort and priorities, I believe it is now* time to initiate studies on two general themes regarding the possible mingling of coral reef biotas across the Isthmus of Panama: ( 1) identification and study of potential colonizing species and (2) assessment (in isolation, but simulating natural systems as closely as possible) of the nature and intensity of the interactions of novel species. The first line of inquiry would provide (a)a n inventory of local coral reef biotas and (b)information on the life history tactics of diverse taxa relevant to colonization. Of particular interest are population growth characteristics, including egg and larval production and development (e.g. frequency and intensity of reproduction, seasonal timing, time spent in water column), the settling behaviour of planktonic larvae and morphological plasticity. Information obtained on species interactions, such as feeding ecology (generalist versus specialist diets, switching behaviour, prey palatability, etc.), competition, symbiosis, parasites and disease organisms, inter alia, could be used to help identify potentially undesirable species. The identification of such forms could also serve as a point of focus for further analysis of the *The main problem areas outlined here were already identified and recommended for study in 1970 by the Committee of Ecological Research for the Interoceanic Canal (CERIC),National Academy of Sciences (Newman, 1972).

122

P.

w.

(:LYNN

attributes of colonizing species. I would recommend, as has Vermeij (1978), that such studies also include comparisons with migrations that have occurred elsewhere in the world. Finally, every effort should be made to maintain effective biotic barriers across the Panama waterway. This applies to the existing locks canal-in the light of recent freshwater shortages and proposed salinization of Gatun Lake, and the discharge of increasing amounts of seawater ballast (from the Caribbean into the tropical eastern Pacific) in the trans-shipment of Alaskan oil-as well as to a possible sea-level canal. The uncertainties inherent in the outcome of species introductions demand that the utmost precaution be taken to safeguard the integrity of tropical marine ecosystems. Our present meagre understanding of coral reef ecosystems raises many possibilities, but allows no failsafe judgements to be passed, on the relative risks of potential ecological and environmental effects resulting from transisthmian migrations.

VI. ACKNOWLEDGEMENTS I wish t o thank the following for the suggestions they offered to help improve this essay: L. Buss, J . Cubit, C. E. Dawson, J . W. Durham, M. L. Jones, H. Lessios, W. A. Newman, R . M. Overstreet and G. J . Vermeij. I gratefully acknowledge the following for permission t o include in Table I1 their unpublished feeding observations: L. Buss, R. C. Highsmith, J. C. Lang, J. Ogden and D. R. Robertson. New data were obtained from research supported by the Smithsonian Research Foundation. I owe a special debt to R. W. Grigg, and D. R. Stoddart, who provided the initial stimulus to examine the problem of Panamic marine migrations. It is also a pleasure to acknowledge the encouragement and support provided by I. Rubinoff, director. Smithsonian Tropical Research Institute.

VII. REFERENCES Abele, L. G. (1976). Comparative species richness in fluctuating and constant environments: coral-associated decapod crustaceans. Science, Washington, D.C. 192, 461463. Adey, W. H. (1978). Coral reef morphogenesis: a multidimensional model. Science, Washington, D.C. 202, 831-837. Adey, W. H. and Vassar, J. M. (1975). Colonization, succession and growth rates of tropical crustose coralline algae (Rhodophyta, Cryptonemiales). Phycologia 14, 5569.

CORAL COMMLWITIES

123

Antonius, A. (1976). New observations on coral destruction in reefs. Abstract, Proceedings, Association of Island Marine Laboratories of the Caribbean, 10th Meeting, p. 17. Antonius, A. (1977).Coral mortality in reefs: a problem for science and management. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. 11, pp. 617-623. University of Miami, Florida. Aron, W. I. and Smith, S. H. (1971). Ship canals and aquatic ecosystems. Science, Washington, D.C. 174, 13-20. Atlantic-Pacific Interoceanic Canal Study Commission ( 1970). Report of the Atlantic-Pacific interoceanic canal study commission. Interoceanic Canal Studies 1970, 129 pp. + 5 annexes. Bak, R. P. M. and van Eys, G. (1975). Predation of the sea urchin Diadema antillarum Philippi on living coral. Oecologia, Berlin 20, 1 1 1-1 15. Bak, R. P. M., Brouns, J. J. W. M. and Heys, F. M. L., (1977). Regeneration and aspects of spatial competition in the scleractinian corals Agaricia agaricites and Montastrea annularis. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 143-148. University of Miami, Florida. Bakus, G. J. (1969). Energetics and feeding in shallow marine waters. International Review of General and Experimental Zoology 4, 275-369. Bakus, G. J. and Green, G. (1974). Toxicity in sponges and holothurians a geographic pattern. Science, Washington, D.C. 185, 951-953. Bardach, J. E., Ryther, J. H. and McLarney, W. 0. (1972). “Aquaculture. The Farming and Husbandry of Freshwater and Marine Organisms”, 868 pp. Wiley-Interscience, New York. Barham, E . G., Gowdy, R . W. and Wolfson, F. H. (1973). Acanthaster (Echinodermata, Asteroidea) in the Gulf of California. Fishery Bulletin, Seattle, Washington 71, 927-942. Benayahu, Y. and Loya, Y. (1977). Seasonal occurrence of benthic-algae communities and grazing regulation by sea urchins a t the coral reefs of Eilat, Red Sea. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 383-389. University of Miami, Florida. Ben-Tuvia, A. (1966). Red Sea fishes recently found in the Mediterranean. Copeia 2, 254-275. Ben-Tuvia, A. (1978). Immigration of fishes through the Suez Canal. Fishery Bulletin, Seattle, Washington 76, 249-255. Ben-Yami, M. and Glaser, T. (1974). The invasion of Saurida undosquamis (Richardson) into the Levant Basin-an example of biological effect of interoceanic canals. Fishery Bulletin, Seattle, Washington 72, 35S373. Birkeland, C. (1977). The importance of rate of biomass accumulation in early successional stages of benthic communities to the survival of coral recruits. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 15-21. University of Miami, Florida. Bohlke, J. E. and Chaplin, C. C. G. (1968). “Fishes of the Bahamas and Adjacent Tropical Waters”, 77 1 pp. Livingston, Wynnewood, Pennsylvania. Brawley, S. H. and Adey, W. H. (1977). Territorial behavior of threespot damselfish (Eupomacenlrus planijrons) increases reef algal biomass and productivity. Environmental Biology of Fishes, The Hague 2, 45-51. Briggs, J . C. (1969). The sea-level Panama Canal: potential biological catastrophe. Bioscience, Washington, D.C. 19, 4 4 4 7 .

124

P. W. OLYNN

Briggs, J . C. (1970). Tropical shelf zoogeography. Proceedings of the California Academy of Sciences, Sun Francisco 38, 131-137. Briggs, J. C. (1974). “Marine Zoogeography”, 475 pp. McGraw-Hill, New York. Brusca, R. C. and Thomson, D. A. (1977).The Pulmo Reefs o f Baja California-true coral reef formation in the Gulf of California. Ciencias Marinas 1, 37-53. Carlgren, 0.(1951).The actinian fauna of the Gulf of California. Proceedings of the United States National Museum 101,415449. Chesher, R. H . (1968).Transport of marine plankton through the Panama Canal. Limnology and Oceanography, Lawrence, Kansas 13, 387-388. Chesher, R. H. (1972).The status of knowledge o f Panamanian echinoids, 1971, with comments on other echinoderms. In “The Panamic Biota. Some Observations Prior to a Sea-Level Canal” (M. L. Jones, ed.). Bulletin ofthe Biological Society of Washington 2,13S158. Christie, W. J . (1974).Changes in the fish species composition of the Great Lakes. Journal of the Fisheries Research Board of Canada 31,827-854. Ciardelli, A. (1967).The anatomy of the feeding mechanism and the food habits of Microspathodon chrysurus (Pisces : Pomacentridae). Bulletin of Marine Science 17,845-883. Connell, J. H. (1973).Population ecology o f reef-building corals. In “Biology and Geology ofCora1 Reefs” (0.A. Jones and R. Endean, eds), Vol. 11, Biology 1 , pp. 205-245. Academic Press, London and New York. Connell, J. H. (1975). Some mechanisms producing structure in natural communities: a model and evidence from field experiments. In “Ecology and Evolution of Communities” (M. L. Cody and J. M. Diamond, eds), pp. 46&490. Belknap Press, Harvard University Press, Cambridge, Massachusetts. Connell, J. H. (1976).Competitive interactions and the species diversity of corals. In “Coelenterate Ecology and Behavior” (G. 0. Mackie, ed.), pp. 51-58. Third International Symposium on Coelenterate Biology, University o f Victoria, British Columbia. Connell, J. H. (1978). Diversity in tropical rain forests and coral reefs. Science, Washington, D.C. 199, 1302-1310. Crisp, D. J. (1974). Interspecies and intraspecies chemoreception by marine invertebrates. In “Chemoreception in Marine Organisms” (P. T. Grant and A. M. Mackie, eds), pp. 105-141. Academic Press, London and New York. Dana, T. F. (1970). Coral reef assemblages. Alpha Helix Research Program 1969-1970, pp. 37-38. University of California, San Diego. Dana, T. F. (1975).Development of contemporary eastern Pacific coral reefs. Marine Biology, Berlin 33,355-374. Dana, T. and Wolfson, A. (1970). Eastern Pacific crown-of-thorns starfish populations in the lower Gulf of California. Transactions of the Sun Diego Society of Natural History 16,83-90. Darlington, P. J., J r (1957). “Zoogeography: The Geographical Distribution of Animals”, 675 pp. John Wiley, New York. Darlington, P. J.,Jr (1965). “Biogeography of the Southern End o f the World”, 236 pp. Harvard University Press, Cambridge, Massachusetts. Dart, J. K. G. (1972).Echinoids, algal lawn and coral recolonization. Nature, London 239,5G51. D’Asaro, C.N. (1969).The egg capsules ofJenneriapustulata (Lightfoot, 1786)with notes on spawning in the laboratory. Veliger 11, 182-184. Dawson, C. E. (1973).Occurrence ofan exoticeleotrid fish in Panamti with discussion of probable origin and mode o f introduction. Copeia 1, 141-144.

CORAL COMMUNITIES

125

Dawson, E. Y. (1966). “Marine Botany; an Introduction”, 371 pp. Holt, Rinehart and Winston, Inc., New York. Durham J. W. (1966). Coelenterates, especially stony corals, from the Galapagos and Cocos Islands. In “The GalLpagos” (R. I. Bowman, ed.), pp. 123-135. University of California Press, Berkeley. Durham, J. W. (1980). New fossil Pocillopora (coral) from Guadalupe Island. In “The California Islands: Proceedings of a Multidisciplinary Symposium” (D. M. Power, ed.), pp. 63-70. Santa Barbara Museum of Natural History, California. Durham, J . W. and Allison, E. C. (1960).The geologic history of Baja California and its marine faunas. Systematic Zoology 9, 47-91. Dustan, P. (1977). Vitality of reef coral populations off Key Largo, Florida recruitment and mortality. Environmental Geology 2 , 51-58. Earle, S. A. (1972).A review of the marine plants of Panama. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. I,. Jones, ed.), Bulletin of the Biological Society of Washington, 2, 69-87. Edmunds, M. (1974). “Defence in Animals”, 357 pp. Longman, Essex. Ehrlich, P. R. (1975).The population biology of coral reef fishes. A n n u l Review of Ecology and Systematics 6, 211-247. Elton, C. S. (1958). “The Ecology of Invasions by Animals and Plants”, 181 pp. Methuen and Co. Ltd, London. Emerson, W. K. (1978). Mollusks with Indo-Pacific faunal affinities in the eastern Pacific Ocean. The Nautilus 92, 91-96. Emiliani, C., Gaertner, S. and Lidz, B. (1972).Neogene sedimentation on the Blake Plateau and the emergence of the Central American Isthmus. Palaeogeography, Palaeoclimatology, Palaeoecology 11, 1-10, Endean, R. and Stablum, W. (1973).A study of some aspects of the crown-of-thorns starfish (Acanthaster planci) infestations of reefs of Australia’s Great Barrier Reef. Atoll Research Bulletin 168, 1-26. Feder, H. M. (1966).Cleaning symbiosis in the marine environment. In “Symbiosis” (S. M. Henry, ed.), Vol. I, pp. 327-380. Academic Press, London and New York. Fenical, W. (1975).Halogenation in the Rhodophyta, a review. Journal of Phycoloqy 11, 245-259. Frydl, P. (1977). The geological effect of grazing by parrotfish (Scaridae) on a Barbados coral reef. M.8c. Thesis, McGill University, Montkal, Qu6bec. Garrett, P. and Ducklow, H. (1975).Coral diseases in Bermuda. Nature, London 253, 349-350. Garth, J. S. (1974). On the occurrence in the eastern tropical Pacific of Indo-West Pacific decapod crustaceans commensal with reef-building corals. In “Proceedings: Second International Symposium on Coral Reefs” (A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J . S. Jell, 0. A. Jones, P. Mather and F. H . Talbot, eds), Vol. I, pp. 397404. Great Barrier Reef Committee, Brisbane. Geister, J. (1977).Occurrence of Pocillopora in late Pleistocene Caribbean coral reefs. Second symposium international sur les coraux et r6cifs coralliens fossiles, Paris, 1975, Bureau de Recherche8 Giologiques et Minidres, MLmoires 89, 378-388. Gladfelter, E. H . and Monahan, R. K. (1977). Primary production and calcium carbonate deposition rates in Acropora palmata from different positions in the reef. I n “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. 11, pp. 38S394. University of Miami, Florida.

126

P. W. GLYNN

Glynn, P. W. (1962). Hermodice curunculata and Mithraculus sculptus, two hermatypic coral predators. Abstract, Proceedings, Association of Island Marine Laboratories of the Caribbean 4, 1 G 1 7 . Glynn, P. W. (1972). Observations on the ecology of the Caribbean and Pacific coasts of Panama. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed.). Bulletin ofthe Biological Society of Washington2, 13-30, Glynn, P. W. (1973). Aspects of the ecology of coral reefs in the western Atlantic region. In “Biology and Geology of Coral Reefs” (0.A. Jones and R. Endean, eds), Vol. 11,Biology 1, pp. 271-324. Academic Press, London and New York. Glynn, P. W. (1974a). The impact of Acanthaster on corals and coral reefs in the eastern Pacific. Environmental Conservation 1, 295-304. Glynn, P. W. (197413).Rolling stones among the Scleractinia: mobile coralliths in the Gulf of Panama. In “Proceedings: Second International Symposium on Coral Reefs” (A. M. Cameron etal.,eds, cf. Garth, 1974),Vol. 11,pp. 183-198. The Great Barrier Reef Committee, Brisbane, Australia. Glynn, P. W. (1976).Some physical and biological determinants of coral community structure in the eastern Pacific. Ecological Monographs 46, 431-456. Glynn, P. W. (1977). Coral growth in upwelling and nonupwelling areas off the Pacific coast of Panama. Journal of Marine Research 35, 567-585. Glynn, P. W. and Macintyre, I. G. (1977).Growth rate and age of coral reefs on the Pacific coast of Panama. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. 11, pp. 251-259. University of Miami, Florida. Glynn, P. W . and Stewart, R. H. (1973). Distribution of coral reefs in the Pearl Islands (Gulf of Panama) in relation to thermal conditions. Limnology and Oceanography 18, 367-379. Glynn, P. W., Stewart, R. H. and McCosker, J . E. (1972). Pacific coral reefs of PanamB: structure, distribution and predators. Geologische Rundschau, Stuttgart 61, 483-519. Glynn, P. W. Wellington, G. M. and Birkeland, C. (1979). Coral reef growth in the Galapagos: limitation by sea urchins. Science, New York 203, 4 7 4 9 . Goreau, T. F. and Goreau, N. I. (1973). The ecology of Jamaican coral reefs. 11. Geomorphology, zonation, and sedimentary phases. Bulletin of Marine Science 23, 399-464. Goreau, T. F., Lang, J. C., Graham, E. A. and Goreau, P. D. (1972).Structure and ecology of the Saipan reefs in relation to predation by Acanthaster planci (Linnaeus). Bulletin of Marine Science 22, 113-152. Gygi, R . A. (1975).Sparisoma viride (Bonnaterre),the stoplight parrotfish, a major sediment producer on coral reefs of Bermuda. Ecologae Geologicae Helvetiae 68,327-359. Harleman, D. R. F. (1972).Exchange of water through the proposed sea-level canal a t PanamL. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed.). Bulletin of the Biological Society of Washington 2 , 4 1 4 7 . Heck, K. L., Jr and McCoy, E. D. (1978). Long-distance dispersal and the reefbuilding corals of the eastern Pacific. Marine Biology, Berlin 48, 349-356. Hendler, G. (1977).The differential effects of seasonal stress and predation on the stability of reef-flat echinoid populations. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 217-223. University of Miami, Florida. Highsmith, R. C. (1980). Geographic patterns of coral bioerosion: a productivity hypothesis. Journal of Experimental Marine Biology and Ecology 46, 177-1 96.

CORAL COMMUNITIES

127

Hobson, E. S. (1971). Cleaning symbiosis among California inshore fishes. Fishery Bulletin 69, 491-523. Hobson, E. S. (1974). Feeding relationships of teleostean fishes on coral reefs in Kona, Hawaii. Fishery Bulletin 7 2 , 915-1031. Holcombe, T. L. and Moore, W. S. (1977). Paleocurrents in the eastern Caribbean: geologic evidence and implications. Marine Geology 23, 35-56. Hopkins, S. H. (1957). Parasitism. In “Treatise on Marine Ecology and Paleoecology” (J. W. Hedgpeth, ed.), Vol. I, pp. 413428. Waverly Press, Baltimore, Maryland. Hubbs, C. L. (1952). Antitropical distribution of fishes and other organisms. I n “Symposium on problems of bipolarity and of pan-temperate faunas”, pp. 324-329. Proceedings 7th Pacific Science Congress, Vol. 3. Hubbs, C. L. (1960). Quaternary paleoclimatology of the Pacific coast of North America. California Cooperative Oceanic Fisheries Investigations Reports, Vol. 7, 1 January 1958 to 30 June 1959, pp. 51S-526. Jackson, J . B. C. (1977). Competition on marine hard substrata: the adaptive significance of solitary and colonial strategies. American. Naturalist 111, 743-767. James, N. P., Mountjoy, E. W. and Omura, A. (1971). An early Wisconsin reef terrace a t Barbados, West Indies, and its climatic implications. Geological Society of America Bulletin 82, 2011-2018. Jones, M. L. and Dawson, C. E. (1973).Salinity-temperature profiles in the Panama Canal locks. Marine Biology, Berlin 21, 8G90. Kaufman, L. (1977).The three spot damselfish: effects on benthic biota ofCaribbean coral reefs. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 55S564. University of Miami, Florida. Kinzie, R. A., I11 (1970). The ecology of the gorgonians (Cnidaria, Octocorallia) of Discovery Bay, Jamaica. 269 pp. Ph.D. Thesis, Yale University, New Haven, Connecticut. Lang, J . C. (1970). Inter-specific aggression within the scleractinian reef corals. 177 pp. Ph.D. Thesis, Yale University, New Haven, Connecticut. Lang, J. C. (1973).Interspecific aggression by scleractinian corals. 2. Why the race is not only to the swift. Bulletin of Marine Science 23, 260-279. Lang, J . C. (1974). Biological zonation at the base of a reef. American Scientist 62, 272-281. Laxton, J . H. (1974). Aspects of the ecology of the coral-eating starfish Acarrthaster planci. Biological Journal of the Linnean Society 6 , 19-45. Lewis, J . B., Axelsen, F., Goodbody, I., Page, C. andchislett, G. (1968).Comparative growth rates of some reef corals in the Caribbean. Marine Sciences Manuscript, Report No. 10, McGill University, Montreal 26 pp. MacArthur, R. H. (1972). “Geographical Ecology: Patterns in the Distribution of Species”, 269 pp. Harper and Row Publishers, New York. MacArthur, R. H. and Wilson, E. 0. (1967). “The Theory of Island Biogeography”, 203 pp. Princeton University Press, New Jersey. Macintyre, I . G. (1968). Some submerged cora1,reefs in the Caribbean. F i f t h Caribbean Geological Conference, St Thomas, Virgin Islands 1968, pp. 1-17. Macintyre, I. G. (1972). Submerged reefs of eastern Caribbean. The American Association of Petroleum Geologists Bulletin 56, 720-738. Macintyre, I. G., Burke, R. B. and Stuckenrath, R. (1977). Thickest recorded Holocene reef section, Isla PBrez core hole, Alacran Reef, Mexico. Geology 5 , 749-754.

128

P. W. GLYNN

Macintyre, I . G. and Glynn, P. W. (1976). Evolution of modern Caribbean fringing reef, Geleta Point, Panama. The American Association of Petroleum Geologists Bulletin 60, 1054-1072. Maragos, J. E. (1972).A study of the ecology of Hawaiian reef corals. 290 pp. Ph.D. Thesis, University of Hawaii, Honolulu, Hawaii. Marsden, J . R. (1962). A coral-eating polychaete. Nature, London 193, 598. Marsden, J. R. (1963). The digestive tract of Hermodice carunculata (Pallas). Polychaeta: Amphinomidae. Canadian Journal of Zoology 41, 165-184. Mayr, E. (1965).The nature of colonizations in birds. In “The Genetics ofColonizing Species” (H. G. Baker and G. L. Stebbins, eds), pp. 2 9 4 7 . Academic Press, London and New York. McCosker, J . E. and Dawson, C. E. (1975). Biotic passage through the Panama Canal, with particular reference to fishes. Marine Biology, Berlin 30, 343-351. McCoy, E. D. and Heck, K. L., Jr. (1976). Biogeography of corals, seagrasses, and mangroves: an alternative to the center of origin concept. Systematic Zoology 25, 20 1-2 10. Menzies, R. J . (1968).Transport of marine life between oceans through the Panama Canal. Nature, London 220, 802-803. Mesolella, K . J., Sealy, H . A. and Matthews, R. K . (1970).Facies geometries within Pleistocene reefs of Barbados, West Indies. The American Association of Petroleum Geologists Bulletin 54, 189S1917. Mileikovsky, S. A. (1971). Types of larval development in marine bottom invertebrates, their distribution and ecological significance: a re-evaluation. Marine Biology, Berlin 10, 193-213. Miller, A. C. (1970). Observations on the associations and feeding of six species of prosobranch gastropods on anthozoans in Discovery Bay Jamaica. Organization for Tropical Studies, Marine Biology Course (Jamaica), Miami, pp. 1-5. Miller, R. S. (1967). Pattern and process in competition. Advances in Ecological Research 4, 1-74. Morse, D. E., Morse, A. N. C. and Duncan, H. (1977). Algal “tumors” in the Caribbean sea-fan, Gorgonia ventalina. In “Proceedings: Third International Coral Reef Symposium” (D. L. Taylor, ed.), Vol. I, pp. 623-629. University of Miami, Florida. Mullins, H. T. and Neumann, A. C. (1978). Geology of the Miami terrace and its paleo-oceanographic implications. Marine Geology 30,205-232. Newell, N. D. (1971).An outline history of tropical organic reefs. American Museum Nmitates No. 2465, 1-37. Newman, W. A . (1972).The National Academy of Science Committee on the ecology of the interoceanic canal. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed.), Bulletin ofthe Biological Society of Washington 2, 247-259. Olsson, A. A. (1972).Origin of the existing Panamic molluscan biotas in terms of their geologic history and their separation by the isthmian land barrier. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed.), Bulletin of the Biological Society of Washington 2 , 117-123. Ormond, R. F. G. and Campbell, A. C. (1974). Formation and breakdown of Acanthaster planci aggregations in the Red Sea. In “Proceedings: Second International Symposium on Coral Reefs” (A. M. Cameron et aZ.,eds, cf. Garth, 1974), Vol. I, pp. 5 9 M 1 9 . The Great Barrier Reef Committee, Brisbane, Australia.

CORAL COMMUNITIES

129

Ormond, R. F. G., Hanscomb, N. J. and Beach, D. H. (1976). Food selection and learning in the crown-of-thorns starfish, Acanthaster planci (L.). Marine Behaviour and Physiology 4, 93-105. Ott, B. S. (1975), Quantitative analysis of community pattern and structure on a coral reef bank in Barbados, West lndies. Ph.D. Thesis, McGill University, Montrkal, Qu6bec. 185 pp. Ott, B. and Lewis, J. B. (1972). The importance of the gastropod Coralliophila abbreviata (Lamarck)and the polychaete Hermodice carunculata (Pallas) as coral reef predators. Canadian Journal of Zoology 50, 1651-1656. Patterson, B. and Pascual, R. (1972).The fossil mammal fauna of South America. In “Evolution, Mammals, and Southern Continents” (A. Keust, F. C. Erk and B. Glass, eds), pp. 247-309. State University of New York Press, Albany, N.Y. Patton, W. K. (1976). Animal associates of living reef corals. I n “Biology and Geology of Coral Reefs” (0.A. Jones and R. Endean, eds),Vol. 111,Biology 2, pp. 1-36, Academic Press, London and New York. Pearson, R. G . and Endean, R. (1969). A preliminary study of the coral predator Acanthaster planci (L.) (Asteroidea) on the Great Barrier Reef. Fisheries Notes (Department of Harbours and Marine, Queensland) 3, 2 7 4 8 . Pianka, E. R. (1974). “Evolutionary Ecology”, 356 pp. Harper and Row Publications, New York. Poi., F. D. (1971). One hundred years of Suez Canal-a century of Lessepsian migration: retrospect and viewpoints. Systematic Zoology 20, 13&159. Por, F. D. (1975). Pleistocene pulsation and preadaptation of biotas in Mediterranean seas: consequences for Lessepsian migrat,ion. Systematic Zoology 24, 72-78. Porter, J. W. (1972).Ecology and species diversity of coral reefs on opposite sides of the Isthmus of PanamB. I n “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed), Bulletin ofthe Biological Society of Washington 2, 89-116. Porter, J. W. (1974).Community structure of coral reefs on opposite sides of the Isthmus of Panama. Science 186, 543-545. Potts, G. W. (1973).The ethology of Labroides dimidiatus (Cuv. & Val.) (Labridae, Pisces) on Aldabra. Animal Behaviour 21, 250-291. Randall, J. E. (1965). Grazing effect on sea grasses by herbivorous reef fishes in the West Indies. Ecology 46, 255-260. Randall, J. E . (1967). Food habits of reef fishes of the West Indies. Studies in Tropical Oceanography No. 5, pp. 665-847. University of Miami Institute of Marine Science. Randall, J. E. (1974).The effect of fishes on coral reefs. I n “Proceedings: Second International Coral Reef Symposium” (A. M. Cameron et al., eds, cf. Garth, 1974), Vol. I, pp. 159-166. Great Barrier Reef Committee, Brisbane, Australia. Richardson, C. A., Dustin, P. and Lang, J. C. (1979).Maintenance ofliving space by sweeper tentacles of Montastrea cavernosa, a Caribbean reef coral. Marine Biology, Berlin 55, 181-186. Robertson, R. (1970).Review of the predators and parasites of stony corals, with special reference to symbiotic prosobranch gastropods. Pacific Science 24. 43-54. Rosenblatt, R. H . ,McCosker, J. E . and Rubinoff, I. (1972).Indo-West Pacific fishes from the Gulf of Chiriqui, Panama. Contributions in Science, Natural History Museum, Los Angeles County 234, 1-18.

130

P. M’. (;LYNN

Rubinoff, I. (1970). The sea-level canal controversy. Biological Conservation 3, 33-36. Rubinoff, I. and Kropach, C. (1970). Differential reactions of Atlantic and Pacific predators to sea snakes. Nature, London 228, 1288-1290. Rutzler, K . (1971). Bredin-Archbold-Smithsonian Biological Survey of Dominica: burrowing sponges, genus Siphonodictyon Bergquist, from the Caribbean. Smithsonian Contributions to Zoology 77, 1-37. Rutzler, K. (1972). Principles of sponge distribution in Indo-Pacific coral reefs: results of the Austrian Indo-West Pacific Expedition 1959/60. I n “Symposium on Corals and Coral Reefs” (C. Mukundan and C. S. G. Pillai, eds), pp. 315-332. The Marine Biological Association of India, Cochin. Sachet, M.-H. (1962). Geography and land ecology of Clipperton Island. Atoll Research Bulletin 86, 1-115. Saito, T. (1976). Geologic significance of coiling direction in the planktonic foraminifera Pulleniatina. Geology 4, 305-309. Sammarco, P. W., Levinton, J. S. and Ogden, J . C. (1974).Grazing and control of coral reef community structure by Diadema antillarum Philippi (Echinodermata: Echinoidea): a preliminary study. Journal of Marine Research 32, 47-53. Scheltema, R. S. ( 1977). Dispersal of marine invertebrate organisms: paleobiogeographic and biostratigraphic implications. In “Concepts and Methods of Biostratigraphy” (E. G. Kauffman and J . E. Hazel, eds), pp. 73-108. Dowden, Hutchinson and Ross, Inc., Stroudsburg, Pennsylvania. Sebens, K. P. (1976).The ecology of Caribbean sea anemones in Panama: utilization of space on a coral reef. I n “Coelenterate Ecology and Behavior” ( G . 0. Mackie, ed.), pp. 67-77. Plenum Publishing Corporation, New York. Sheffey, J. P. (1972). Physical characteristics of the proposed sea-level isthmian canal. In “The Panamic Biota: Some Observations Prior to a Sea-level Canal” (M. L. Jones, ed.). Bulletin of the Biological Society of Washington 2 , 31-39. Sheppard, C. R. C. (1979).Interspecific aggression between reef corals with reference to their distribution. Marine Ecology-Progress Series 1, 237-247. Shinn, E. A. (1976). Coral reef recovery in Florida and the Persian Gulf. Environmental Geology 1, 241-254. Silva, P. C. (1966). Status of our knowledge of the Galapagos benthic marine algal flora prior to the GaIBpagos International Scientific Project. In “The Galapagos” (R. J . Bowman, ed.), pp. 149-156. University of California Press, Berkeley. Simpson, G. G. (1947).Holarctic mammalian faunas and continental relationships during the Cenozoic. Bulletin of the Geological Society of America 58, 613-688. Slobodkin, L. B. (1968).Toward a predictive theory of evolution. In “Population Biology and Evolution” (R. C. Lewontin, ed.), pp. 187-205. Proceedings of the International Symposium, 1967. Syracuse University Press, New York. Slobodkin, L. B. and Fishelson, L. (1974).The effect of the cleaner-fish Labroides dimidiatus on the point diversity offishes on the reeffront a t Eilat. The American Naturalist 108, 36S376. Springer, V. G. (1958). Systematics and zoogeography of the clinid fishes of the subtribe Labrisomini Hubbs. Publications of the Institute of Marine Science, The University of Texas, Port Aransas 5 , 417492. Squires, D. F. (1959).Results of the Puritan-American Museum of Natural History Expedition to Western Mexico. Corals and coral reefs in the Gulf of California. Bulletin of the American Museum of Natural History 118, 367432.

CORAL COMMLINITIES

131

Stearns, S. C. (1976).Life-history tactics: a review of the ideas. The Quarterly Hevieui of Biology 51, 3 4 7 . Stephenson, W. and Gearles, R. B. (1960).Experimental studies on the ecology of intertidal environments a t Heron Island. Australian Journal of Mari,ne and Freshwater Research 11. 241-267. Stoddart, D. R. (1969a). Ecology and morphology of recent coral reefs. Biological Reviews 44, 433498. Stoddart, D. R. (1969b).Post-hurricane changes on the British Honduras reefs and cays: re-survey of 1965. Atoll Research Bulletin 131, 1-25. Storr, J. F. (1964).Ecology of the Gulf of Mexico commercial sponges and its relation to the fishery. Special ScientiJicReport-Fisheries United States Department of the Interior, Fish and Wildlife Service No. 466, p. 73. Taylor, W. R. (1945).Pacific marine algae of the Allan Hancock Expeditions to the Galapagos Islands. Allan Hancock PaciJic Expeditions 12, 1-528. Thorson, G. (1950). Reproduction and larval ecology of marine bottom invertebrates. Biological Reviews 25, 1 4 5 . Thonon, G. (1961). Length of pelagic larval life in marine bottom invertebrates as related to larval transport by ocean currents. In “Oceanography” (M. Sears, ed.), pp. 455-474. American Association for the Advancement of Science, Washington, D.C. Van den Hoek, C., Cortel-Breeman,A. and Wanders, J . (1975).Algal zonation in the fringing coral reef of Curaqao, Netherlands Antilles, in relation to zonation of corals and gorgonians. Aquatic Botany 1, 269-308. Vaughan, T. W. (1919). Fossil corals from Central America, Cuba, and Porto Rico, with an account of the American Tertiary, Pleistocene, and Recent coral reefs. United States National Museum Bulletin 103, 18S524. Vermeij, G. J. (1974).Marine faunal dominance and molluscan shell form. Evolution 28, 65-64. Vermeij, G. J. (1977). Patterns in crab claw size: the geography of crushing. Systematic Zoology 26, 138-151. Vermeij, G. J. (1978).“Biogeography and Adaptation: Patterns of Marine Life”, 332 pp. Harvard University Press, Cambridge. Vine, P. J. (1973).Crown-of-thorns (Acanthuster planci) plagues: the natural causes theory. Atoll Research Bulletin 166, 1-10. Wanders, J. B. W. (1977).The role of benthic algae in the shallow reef of Curacao (Netherlands Antilles) 111: the significance- of grazing. Aquatic Botany 3, 357-390. Ward, J . (1965). The digestive tract and its relation to feeding habits in the stenoglossan prosobranch Coralliophila abbreviata (Lamarck). Canadian Journal of Zoology 43, 447464. Webb, S. D. (1976).Mammalian faunal dynamics of the great American interchange. Paleobiology 2 , 22C~234. Weber, J. N. and Woodhead, P. M. J . (1970).Ecological studies of the coral predator Acanthaster planci in the South Pacific. Marine Biology, Berlin 6 , 12- 17. Wellington, G. M. (1975).The Galapagos coastal marine environments: a resource report to the Department of National Parks and Wildlife, Quito. Ministerio de Agricultura y Ganaderia, 357 pp., Quito, Ecuador. Wellington G. M. (1980). Reversal of digestive interactions between Pacific reef corals: mediation by sweeper tentacles. Oecologia 47, 340-343.

132

P. W . CLYNN

Wells, J. W. (1956).Scleractinia. I n “Treatise on Invertebrate Paleontology” (R. C. Moore, ed.), Part F. Coelenterata, pp. 328-444. Geological Society of America, University of Kansas Press, Lawrence. Wells, J. W. (1957).Coral reefs. I n “Treatise on Marine Ecology and Paleoecology” (J. W. Hedgpeth, ed.), Vol. I, pp. 1087-1104. Waverly Press, Baltimore, Maryland. Wells, J. W. (1978). Annotated list of the scleractinian corals of the Galapagos (manuscript, 112 pp.). Wickler, W. (1968). “Mimicry in Plants and Animals”, 255 pp. World University Library, McGraw-Hill Book Company, New York. Wilson, D. P. (1960). Some problems in larval ecology related to the localized distribution of bottom animals. I n “Perspectives in Marine Biology” (A. A. Buzzati-Traverso, ed.), pp. 87-103. University of California Press, Berkeley. Wilson, D. P. (1968).The settlement behaviour of the larvae of Subellaria alveolata (L.). Journal of the Marine Biological Association of the United Kingdom 48, 387435. Wilson, E. 0 . and Willis, E. 0 . (1975). Applied biogeography. In “Ecology and Evolution of Communities” (M. L. Cody and J. M. Diamond, eds), pp. 522-534. Belknap Press, Harvard University Press, Cambridge, Massachusetts. Woodriw, W. P. (1954).Caribbean land and sea through the ages. Bulletin of the Geological Society of America 65, 7 1S732. Woodring, W. P. (1957). Geology and paleontology of Canal Zone and adjoining parts of Panama. Geological Survey Professional Paper No. 306-A, 145 pp. Wulff, J. L. and Buss, L. W. (1979). Do sponges help hold coral reefs together? Nature, London 281, 474475. Yonge, C. M. (1960). “Oysters”, 209 pp. Collins, London.

Adv . Mar . B i d . , Vol . 19, 1982. pp. 133-256

ENVl RONM ENTAL SIM ULATlON EXPERl MENTS ON MARINE A N D ESTUARINE ANIMALS J . DAVENPORT N . E . R . C . Unit of Marine Invertebrate Biology, Marine Science Laboratories. University College of North Wales. U .K . I . Introduction . . . . . . . . . . I1. Variability of the Inshore Environment . . A . Temperature and salinity fluctuations at an intertidal estuarine site .. B. Rock pool physico-chemical conditions I11. Development of Simulation Equipment . . IV . Regimes . . . . . . . . . . . V . Temperature Experiments . . . . . . A . Survival . . . . . . . . . B . Development . . . . . . . . C . Reproduction . . . . . . . . D . Adaptation . . . . . . . . E . Interaction with other factors .. . . . . . . . . VI . Salinity Studies A . Survival . . . . . . . . . B . Behavioural responses . . . . . C . Reproduction . . . . . . . . D . Growth . . . . . . . . . . E . Feeding . . . . . . . . . . F . Osmotic/ionic responses . . . . G. Oxygen consumption . . . . . VII . Oxygen Tension Studies . . . . . . VIII . Pollutant Studies . . . . . . . . IX . Conclusions . . . . . . . . . . X . Acknowledgements . . . . . . . . X I . References . . . . . . . . . .

.. ..

.. ..

.. ..

.. ..

.. ..

133 137

.. ..

..

..

..

..

..

..

..

..

.. ..

..

..

..

.. .. ..

.. .. ..

.. ..

.. .. .. .. ..

..

..

..

..

..

.. ..

.. ..

..

.. ..

.. .. ..

..

..

.. .. .. ..

.. .. ..

..

..

..

..

..

..

..

.. .. . . .. .. .. ..

139 143 145 160 166 166 175 177 179 in2 184 185 190 207 2008 209 209 230 233 236 240 242 242

.. .. ..

.. .. ..

..

..

..

.. . . . . ..

. . ..

..

..

..

..

..

..

..

..

..

..

.. ..

.. ..

..

..

.. ..

. . ..

..

.. ..

..

..

..

.. .. .. .. .. . .

I . INTRODUCTION While the open sea is relatively stable in its physio-chemical makeup. indeed impressively so at depth. coastal and estuarine aquatic environments exhibit substantial fluctuations in physical

134

J. DAVENPORT

and chemical characteristics such as light intensity, turbidity, temperature, salinity and oxygen tension. These changes result from terrestrial, seasonal climatic, die1 and tidal influences. I n coastal and estuarine littoral zones these various fluctuations, which interact in their effects upon organisms, are further complicated by periodic tidal emersion, with the consequent possibility of exposure of the littoral fauna to aerial and freshwater influences. Inshore waters and estuarine areas bear the brunt of man’s impact on the seas; they receive most types of deliberately released pollutants (thermal, radiological, inorganic and organic) for eventual dilution and dispersion in the great bulk of the marine hydrosphere and are also affected by accidently discharged pollutants (mainly oil). Because pollutant delivery to the environment is often intermittent, and because the subsequent fate of pollutants is affected by tides, currents, variable freshwater run-off and various chemical or biological processes, it is most unlikely that marine animals will ever be exposed to sustained steady pollutant concentrations. For a variety of reasons animals of inshore waters have attracted rather more attention from experimental workers than have animals from the open sea. Obviously their availability and economy of collection have contributed to this, but their special attributes, evolved in response to the changeable nature of their environment have been particularly attractive to experimentalists. There are likely to be more challenging difficulties in understanding the adaptations of eurythermal or euryhaline animals compared with understanding their offshore stenothermal and stenohaline relatives. Similarly, i t is more logical to test the effects of heavy metal pollutants on inshore fish, crustaceans and molluscs, than on their deep water relatives. Despite a predominant interest in experimental material from changeable environments, the main experimental approach adopted in the study of the effects of various environmental factors on coastal and estuarine animals has been the “steady state” or “direct transfer” experiment (see Fig. 1 ) . Conventionally this approach is used to study the effects of one environmental factor in isolation. Animals are taken from their natural habitat and held in the laboratory under constant conditions which approximate t o field conditions at the time of collection, insofar as knowledge of those conditions and available experimental facilities allow. With the exception of the environmental factor under consideration, all conditions are held constant throughout the experimental period.

ENVIRONMENTAL SIMULATION EXPERIMENTS

135

Experiment a I Accl imat ion Level n?

Native ‘I eve1

t

t

t

Sampling procedures (for mortality, physiological changes etc.)

t

Collect ion

t

of animals

Transfer Time I )

FIG.1 . Format of the “steady state” or “direct transfer” experiment

The experimental animals are divided into groups. One, the control, remains at the acclimation level while the other groups are each transferred, usually directly, t o situations where they encounter various levels of the factor being studied. This type of experiment is simple, does not require complex apparatus, and is easy to repeat at almost any marine laboratory in the world. With portable equipment, and the acceptance of some compromises in the degree of control over factors other than the one under investigation useful studies may be carried out in remote areas. The ‘Lsteady-state”experiment is, however, open to criticism on several grounds, particularly when the results obtained are extrapolated uncritically to field conditions (for example when assessing likely limits to distribution of the animals). The experiment essentially consists of suddenly changing the environmental level of a particular factor and then sustaining the new level, often for long periods. Commonly, any changes induced by the procedure are attributed to the new factor level, and the trauma involved in transfer is ignored. Yet in nature such a sequence ,of events never occurs. An example from salinity studies serves to illustrate this point. The mussel, Mytilus edulis (L,),occurs in a variety of brackish water areas. I n the Baltic Sea, populations of mussels are found at sites where the salinity is much lower than in the open sea (down to

136

J . DAVENPORT

5%,; Theede, 1965), but the salinity levels are quite stable, varying little even on a seasonal basis, so the animals are never exposed to

sudden osmotic shocks. On the other hand, many estuarine populations of Mytilus are exposed to severe tidal fluctuations in salinity (Milne, 1940; Cawthorne, 1979a). They encounter sharp salinity changes, but no particular concentration is sustained for very long, and the mussels, like many other bivalves, survive exposure to low salinities for a few hours by closing the shell valves, and retaining water of high salinity within the mantle cavity (Milne, 1940; Gilles, 1972; Hoyaux et al., 1976; Shumway, 1977; Davenport, 1979a). Consequently, the conventional steady state experiment does not represent the environmental situation of either the Baltic or estuarine mussels. This example also illustrates another pitfall of steady state experiments. Many animals have mechanisms for avoiding damage caused by short term exposure to sub-optimal or deleterious, environmental conditions. Thus, as discussed above, bivalve molluscs reduce contact with the external environment to a greater or lesser extent when the external salinity is low; analogous behaviour is found in barnacles, gastropods, burrowing worms and hermit crabs (Newman, 1967; Foster, 1970; Davenport, 1976; Shumway and Davenport, 1977; Davenport et al., 1980). Wells (1949a, b ) suggested that ArenicoZu marina (L.) avoided exposure to unfavourable media by ceasing to irrigate its burrow, while recent work (Davenport, 1977; Davenport and Manley, 1978; Manley and Davenport, 1979) has shown that several bivalve species are capable of detecting heightened environmental copper levels and can close their shell valves, or at least reduce the rate of irrigation of the mantle cavity to avoid damage during a few hours’ exposure to this pollutant. It has also been known for a long time (Turner et ul., 1948) by workers interested in antifouling techniques that mussels can detect chlorine and close their shell valves to survive intermittent exposure to chlorinated sea water. All of these behaviour patterns are devices to counteract transient unfavourable conditions and evolved only for short term, notably tidal, periods. I n the extended direct transfer type of experiment, animals are often inspected at daily or even longer intervals; by this time overriding respiratory or nutritional demands may have forced animals to abandon their avoidance behaviour and resume normal activity in conditions to which they would not usually be exposed. A refinement of the basic steady state experiment has been the multivariate approach pioneered especially by Costlow et aZ. (1960) and Alderdice (1963, 1972). I n this type of investigation animals are

ENVIRONMESTAL SIMULATIOK EXPERIMENT8

137

exposed to Combinations of factors (e.g. salinity and temperature, or salinity, temperature and oxygen tension, etc.) rather than single factor stresses. Obviously this approach is most useful for finding out how factors interact, and also for determining optimal laboratory or rearing conditions (Box, 1956) very rapidly. However, the experiments have also been used for the construction of multidimensional survival envelopes (e.g. McLeese, 1956) with the implication that these are valid for distribution in the field. The objections set out above for single stress steady state experiments apply equally to multivariate studies, with the additional comment that, if it is unnatural or unphysiological to expose animals to instantaneous single factor shocks, then it is surely even more artificial to impose sudden multifactor changes. Dissatisfaction with some aspects of the steady state experimental approach has spurred an increasing number of workers to develop apparatus to provide experimental regimes which more closely reflect the changeable nature of the environment, at least for one factor (e.g. temperature, salinity, oxygen tension) at a time. Initial steps in this direction were taken with temperature alone (e.g. Grainger, 1956; Khan, 1965). However, following the salinity studies of Tucker (1970a), several researohing groups have developed apparatus t o mimic the salinity fluctuations of estuaries or coastal lagoons. Such salinity studies have been the most common form of environmental simulation experiment performed so far, although the project at the Netherlands Institute for Sea Research, Texel has been running for several years, and some thermal, oxygen tension and pollutant studies have also been performed. Since the literature devoted to such studies is still relatively compact, and because the recent advent of microprocessor techniques holds out the promise of relatively cheap and flexible apparatus to facilitate simulation studies, the time appears ripe for a review of the field.

11. VARIABILITY OF THE INSHORE ENVIRONMENT

It would normally be appropriate to replace the unphysiological aspects of steady state experiments by conditions of temperature, salinity or oxygen tension which animals are likely to encounter in nature. For realistic simulation studies therefore adequate information is required about the source and form of natural fluctuations ir!

138

,J. DAVENPORT

levels of various environmental factors. For the sublittoral environment, whether coastal or estuarine, there is an extensive amount of information about temperature and salinities, simply because physical oceanographers interested in currents and mixing processes rely heavily upon these data in their calculations. Unfortunately much of this information is either somewhat inaccessible, or not well correlated with biological data. Also, except for a few estuarine studies (e.g. Sanders et al., 1965), little detailed information is available about short term (i.e. tidal or diel) changes in temperature or salinity. For the intertidal environment data is much sparser and often incomplete. For example, Southward (1958) showed that barnacles at Plymouth U.K. were exposed to a 16 deg C thermal change between tidal extremes in the summer; this observation was valuable, but gave no idea of the exact form of the temperature changes encountered by the animals over a 24 h period, although it seemed likely that temperature changes associated with emersion/immersion would be quite abrupt. Detailed information about the form and amplitude of changes in salinities and temperatures occurring in the intertidal zone over periods of as much as 24 h at different times of the year was virtually absent until the work of Cawthorne (1979a), itself indequate, being collected from only one particular estuarine site. An area of rather greater knowledge lies in the characteristics of the specialized rock pool environment, including fluctuations in oxygen tension and pH, which have been monitored by several workers (e.g. Stephenson et al., 1934; Pyefinch, 1943; Read, 1969; Ganning, 1971; Daniel and Boyden, 1975) and will be discussed in detail later. Most monitoring of polluted marine or estuarine environments has been carried out over weeks, months or years with emphasis being placed upon relatively infrequent chemical and biological sampling; the possibility or consequences of tidal or diel fluctuations in pollutant concentration have largely been ignored. However, the effects of thermal pollution, in the form of power station discharges, have attracted the attention of many researchers, and the first major review of the field was published as long ago as 1965 (by Naylor). Broadly speaking, power stations have two effects upon the environment. First, and most obviously, their discharges heat up the inshore waters nearby, the temperature increment decreasing with distance from the discharge point. I n terms of fluctuations in environmental temperature this effect simply means that the fluctuations occur rather higher up the temperature scale than they would have done in the absence of the station. Secondly, power

ENVIRONMENTAL SIMULATION EXPEHIMESTS

139

stations apply sudden thermal increases (usually of the order of 10degC in about 5s) to organisms entrained in the cooling water passing through the stations’ condensers. This latter effect can be very damaging, especially in summer when organisms are living closer to their upper thermal limits; Briand (1975) showed that biomass amounting to 1700 tons of organic carbon was destroyed in a year by passage through the condensers of two Californian power stations. The entrainment thermal shock has been simulated in the laboratory notably by Heinle (1969) and Diaz (1973, 1975); more work on both types of thermal effects caused by power stations would be valuable especially upon inshore tropical organisms which may be particularly vulnerable. However, it should be remembered that the thermal shock of entrainment is accompanied by great pressure increases as the cooling water passes through the narrow bores of the condenser. Also the cooling water may be chlorinated, to a greater or lesser extent, to prevent fouling and this may kill animals where thermal shock does not (Heinle, 1969). A comprehensive account of all types of environmental fluctuations which have been monitored by marine biologists would constitute a sizeable review in itself; the following two sections are derived from a few selected examples relevant to the simulation studies already carried out or in progress. A. Temperature and salinity Juctuations at a n intertidal estuarine site When Davenport et al. (1975) employed sinusoidal and abrupt salinity changes in their studies of larval salinity tolerances they were relying upon the work of Sanders et al. (1965),who monitored salinity changes in the Pocasset river, an estuary with a pronounced salt wedge effect, to justify the abrupt salinity regimes. On the other hand, measurements made by students of the University College of North Wales indicated that the salinity of the bottom water of the well mixed Conwy estuary fluctuated in a roughly sinusoidal manner (and often fell to near zero salinity). When these same “idealized” regimes were later applied to estuarine littoral organisms (e.g. barnacles, bivalves) some unease was felt about their validity. For example, were animals of marine origin such as Mytilus edulis ever exposed to fresh water even in estuaries? The work of Tucker (1970a) and Stickle and Ahokas (1974), the only other workers in the field at the time, wits of no assistance, since they too relied on simulations based (somewhat loosely) upon assumptions from sublittoral data.

140

,J. DAVENPORT

Stickle and Denoux (1976) described tidal salinity fluctuations on an Alaskan shore but these only varied from about 8 to 25%,. With the later development of apparatus which could also simulate complex temperature changes (Davenport and Cawthorne, 1978), the problem of using realistic simulated intertidal regimes became more acute. A survey of literature revealed that, although a little information was available about temperature extremes during tidal fluctuations (e.g. Southward, 1958; Lewis, 1960,1963),nowhere did their data show the form of the fluctuations occurring between the extremes. Consequently, during the year from November 1977 to October 1978, monthly visits were paid to a site on the shore of the Conwy estuary in North Wales to monitor salinity and temperature changes over a 24 h period (Cawthorne, 1979a). The recording site was on the lower shore and chosen because it was quite close to the landward limit of common members of the coastal littoral fauna and flora and accessible at tides roughly midway between springs and neaps. Knight and West (1975) had shown that the estuary was well mixed, but the site selected was beside a narrow channel where turbulence was thought likely to produce especially effective mixing. Table I lists the epibenthic organisms found at the recording site (mobile or infaunal organisms were not considered since they might be able to swim or crawl away or might retreat into the substrate to avoid exposure to the full rigours of salinity and temperature changes recorded at a single point above the substrate). Temperature and salinity curves for two monthly recording periods are shown in Fig. 2; the periods of aerial exposure and timings of high water are also indicated. The results showed a great deal of variation in the amplitude and shape of the salinity profiles, despite being collected TAHLE I. LISTOF EPIHENTHIC SPECIES RECORIIEU WITHIN 5 METRES OF RECORDING PROBE ox THE LOWER SHORE OF THE CONWY ESTUARY (FROM CAWTHORSE, 1979a)

Algae

Cirripedes

Molluscs

Fucus vesiculosus Fucus serratus Ascophyllum nodosum Ulva lactuca" Elminius modestus Balanus balanoides Balanus crenatus Littorina saxatilis Mytilus edulis

*Seasonel occurrence (summer/autumn)only.

:

< E

I-" 0

24

4

0

12 Hours

FIG.2[a), (b). Examples oftemperature and salinity fluctuations recorded at a single site in the Conwy Estuary. Honzontal bars indicate periods of emersion; arrows correspond to high water. [a)Values recorded in January 1978. Note large salinity range and small temperature changes during immersion. [b) Values recorded in May 1978. Note smaller salinity range (following a period of low rainfall) and large time temperature changes associated with emersion. From Cawthorne (1979a).

142

J . DAVENPORT

from the same point on occasions of similar tidal height. Changing seasonal and climatic influences, particularly where they affected the freshwater input to the estuary, were responsible for most of the variation. The maximum salinity range noted on a single tide was between 0 and 31%,, recorded twice during the 12 month study period; clearly a number of littoral organisms of marine origin may survive exposure to fresh water for brief periods. Maximum rates of salinity change during a tidal cycle varied from 7 to 18y&,/h. Temperature fluctuations were equally interesting. When the shore was covered with water, the temperature changes were associated with salinity changes. The amplitude of the aquatic temperature fluctuations appeared t o be determined by the difference between river-water and seawater temperatures, and was not influenced by short term changes in air temperature even when these were quite extensive. Thus, in November 1977, when the seawater temperature was 7-4"C, and the river-water input was about 505°C there was a fluctuation in water temperature between these levels, with rising salinity being associated with rising temperature. I n April 1978, however, the freshwater input to the estuary was warmer than the sea, so that rising salinities were associated with falling temperatures. Generally, water temperatures changed slowly and amplitudes of fluctuations during tidal cycles were small, often of little more than 2-2 deg C. However, aerial emersion was always associated with far more abrupt temperature changes of much greater magnitude. The most extreme example of this occurred in May 1978 when the temperature a t the recording point rose by 12.5 deg C in 15 min on one falling tide. Typically, therefore, a single tidal cycle will expose epibenthic intertidal organisms to both gentle and abrupt temperature changes. During the 12 visits freshwater temperatures ranged from 1-0to 19.5"C,seawater temperatures from 2.8 to 16*2"C,while air f,emperatures varied between 0-6 and 32.1"C. The maximum amplitude of thermal fluctuation recorded during a 24 h period also occurred in May 1978, and amounted to 18degC ( 1 4 ~ 3 2 ° C )this ; agrees quite well with the observations of Southward (1958). Newel1 ( 1969) pointed out that temperature fluctuations in temperate intertidal zones were usually greater than in tropical areas where air i%ndsea temperatures were often reasonably close (Lewis, 1960,1963). Temperature fluctuations on the shore a t high latitudes have not attracted much study, although Davenport et al. (1979) suggested that eggs of the capelin MaZZotus viZ2osus (a salmonid teleost) laid on the sandy shores of Balsfjord, N. Norway might be exposed to die1 temperature changes of 25 deg C or more during the subarctic spring.

ENVIRONMENTAL SIMULATION EXPERIMENTS

143

B. Rock pool physico-chemical conditions Unlike the general estuarine or coastal shore, and contrary to popular opinion, intertidal rock pools are usually characterized by stable salinites (Orr and Moorhouse, 1933; Pyefinch, 1943; Ganning, 1971; Daniel and Boyden, 1975), although there may be vertical salinity gradients within them. Even when exposed to heavy freshwater inflow, the salinity at the bottom of littoral pools, left by the tide, is usually almost unchanged from that of sea water (Ganning, 1971; Davenport et al., 1980).This stability follows from stratification and generally poor mixing within the pools. On the other hand, layering and slow mixing contribute, with the metabolic processes of rock pool organisms, to substantial fluctuations in the temperature, pH and oxygen tension (PO,) of the pool water. Examples of such fluctuations are shown in Fig. 3. Maximum recorded die1fluctuations in temperatures in temperate areas vary from 10-15 deg C (Ganning, 1971; Daniel and Boyden, 1975) and so rock pool organisms may encounter amplitudes of temperature fluctuation approaching those associated with aerial exposure in other littoral organisms. However, as shown in Fig. 3, the temperature changes tend to be gentle rather than sudden. Similar changes were observed by Read (1969); Ganning (1971) indicated that such amplitudes of temperature fluctuation were not normally exceeded in tropical rockpools; in Barrier Reef pools Orr and Moorhouse (1933) recorded 10°C diurnal changes. The work of the author (Davenport, 1979c) upon a Norwegian subarctic rock pool which fluctuated between - 8°C (high salinity water beneath ice -see below) and 3°C in spring, suggests that this is true at high latitudes also. These subarctic pools are also interesting because they break the pattern of stable salinities. Ganning (1971) was the first to note that freezing of the surface water in pools produced higher salinities in the underlying water, but this information was derived from rock pools in the Baltic splash zone where the maximum salinity recorded was only 14-3%,. However, Davenport (1979c), working on the shore at Tromso, N. Norway, where the sea is of normal salinity, found that salinities as high as 65%, were regularly developed on each tide beneath the ice formed in shore pools exposed to air temperatures below - 10°C. The increase in salinity appears to take several hours, whereas the return to normal sea water salinities (32-34%,) probably occurs in a few minutes, when the ice is melted and the pools are flushed out by the relatively warm ( + 3 t o +5"C) water of the incoming tide.

1

Pool 1

1

20

E

P 26 24 22 20 18 16 14

12 '08 10 12 14 16 18 20 22 24 02 0 4 06 08 08 10 12 14 16 18 20 22 24 02 04 06 08 08 10 12 14 16 18 20 22 24 02 0 4 06 08 Pools submerged Periods of daylight Pools exposed =Darkness

Time of day ( h )

0

FIG.3. Diurnal variation in oxygen concentration (solid circles) and temperature (open circles) in six rock pools. From Daniel and Boyden (1975)

ENVIRONMENTAL SIMULATION EXPERIMENTS

145

The most striking short term changes which occur in the rock pool environment consist of massive fluctuations in oxygen tension. Both Ganning (1971)and Daniel and Boyden (1975)reported fluctuations in oxygen tension, between 3% and more than 300y0 air saturation in rock pools which contained large quantities of macroalgae or, in pools cut off from the sea for periods of days or weeks, dense populations of phytoflagellates. These fluctuations are basically with the highest values tending t o occur when photosynthetic activity has peaked at around 1300 h or 1400 h. Minimum air saturation levels occur before dawn (approx. 0400 h ) after respiration has proceeded in the absence of photosynthesis for several hours. However, tidal influences complicate this picture because immersed pools when subject to wave and current action will always contain water which is close to full air saturation, whatever the time of day or night. The oxygen content and pH of rock pool waters are closely related; both are affected by algal photosynthesis and respiration of the total ecosystem. Typically night time is characterized by low pH values (around 6-5-7-5) associated with oxygen depletion and CO, release, while daytime figures may rise to 9 or even 10 under conditions of oxygen hypersaturation (Ganning, 1970; Ganning and Wulff, 1969; Daniel and Boyden, 1975). Continuous monitoring of rockpool pH during tidal or die1 cycles appears not to have been carried out, although Ganning (1971) stressed the futility of occasional pH sampling. However, it seems probable that the form of pH fluctuations, like that of the changes in oxygen tension is made up of both gentle and abrupt components. As virtually no work has been carried out upon the direct effect of pH upon marine organisms, this would appear to be a fruitful field for laboratory simulation studies.

111. DEVELOPMENT OF SIMULATION EQUIPMENT Simulation studies started with temperature investigations, the earliest of which appear to have stemmed from the interest of Grainger (1956, 1959) in varying rather than constant temperatures. Initially, experiments were simply direct transfer experiments in which observations were made immediately after the transfer (e.g. Grainger, 1956); a similar approach has been adopted by workers interested in power station entrainment thermal shocks (e.g. Heinle, 1969; Diaz, 1973, 1975). Khan (1965),in her work upon the development of the freshwater species Acanthocyclops viridis (Jurine), employed an irregular

146

$1. 1)AVENPORT

temperature regime (see Fig. 4), produced by keeping aquaria out of doors and occasionally switching immersion heaters within them on or off by hand. Monitoring of temperatures was continuous, but the analytical technique required to evaluate the effects of fluctuating temperature upon development, which followed from the work of Grainger (1959), was somewhat tedious. Also, by definition, experiments of this type cannot be repeated, and the approach does not appear to have been used subsequently. I n any case, as is clear from Section 11, the short term temperature changes encountered in nature by marine and estuarine animals, although very variable, are not truly irregular, but are rhythmically related to tidal and die1 patterns of events. 3 7

0

2

4

6

8

Days Flu. 4. Irregular temperature regime employed by Khan (1965) in her studies on copepods.

The bulk of temperature cycle studies have involved animals living in aquaria with no through flow, and exposed to uncomplicated temperature regimes. Thus Heath (1963) used square wave temperature profiles (wavelengths +I8 h, amplitude 10 deg C, while Hubbs (1964) used rather poorly defined changes. Linear increases and decreases over 24 h periods were employed by Thorp and Hoss

ENVIRONMENTAL SIMULATION EXPERIMENTS

147

(1975) and Widdows (1976). Costlow and Bookhout (1971) and Feldmeth et al. (1974) used die1 thermal regimes which consisted of linear changes in temperature, interspersed with periods of constant high or low temperature. Most of these investigations were .carried out with the aid of relatively simple equipment involving standard items such as time clocks, immersion heaters and motor-driven contact thermometers. Such apparatus has long been used for producing temperature increases for conditioning bivalves to produce ripe gametes (e.g. Loosanoff, 1945a; Loosanoff and Davis, 1950; Gruffydd and Beaumont, 1970). Costlow and Bookhout (1971) used rather more refined equipment; their experiments were performed in environment cabinets with Honeywell temperature programmers. Such cabinets, which usually have light regime control as well, have been used for many years in botanical work, and are ideal for thermal studies upon small organisms (e.g. larvae) in small vessels, but may not easily be used with large animals or organisms needing a large exchange of water. Moreover, they are of no use in the simulation of fluctuations in other variables such as salinity or oxygen tension. Salinity simulation studies started with the work of Tucker (1970a) who imposed simulated tidal salinity cycles upon the gastropod, Scutus brevicutus (Montfort) t o investigate its osmotic responses using the equipment shown in Fig. 5 . Both she, and Stickle and Ahokas ( 1974) employed apparatus which essentially consisted of an aquarium which could be flushed out at a constant flow rate by water of a different salinity from a header tank. The salinity changes thus produced were of rather arbitrary form (see Fig. 14) and, in the case of Stickle and Ahokas’ study, were not accurately repeatable (Tucker gave no information on this point). I n the light of these various defects, Davenport et al. (1975), in their descriptions of a salinity fluctuation apparatus which has since been used in several studies, set out the following design criteria which were thought essential for routine and comparative studies: 1. Salinity must be repeatable, so that different species, or various larval stages of one species, may be exposed to the same salinity regime. 2. Automatic repetition of profiles must be possible to allow the exposure of organisms to simulated tidal salinity fluctuations over long periods of time during feeding, growth or activity experiments. 3. Substitution of one profile for another should be simple and must not involve adjustment which is difficult to reverse.

148

J. DAVENPORT

- Constant pressure bottle

/

Needle valve

--

ions mi-' Constant with time 1 [XI

[ y l i o n s ml-' ( a function of time)

Volume=Vml [c] ions m1-l ( a function o f time)

Aquarium-,--

<

2

FIG.5. Experimental set-up used to obtain a predictable gradual dilution or concentration of a solution in an aquarium. The solution was continuously aerated and mixed by air being bubbled through a diffusion block. From Tucker (1970a).

4. The design of the apparatus must allow salinity fluctuations of any form between the limits of fresh water and full sea water. 5 . Flexibility in the time taken to complete a fluctuating salinity programme is necessary, so that both 24h tidal cycles and spring/neap cycles may be reproduced. 6. The apparatus should be capable of delivery up to, say, 30 litres of water per hour, so that the response of several animals may be studied simultaneously. 7 . The salinity of the outflowing water must be automatically monitored at all times. 8. All factors except salinity should be kept constant, e.g. the temperature of the outflow should be controlled and the freshwater supply should be deionized (because the ionic content of fresh water can change with time and source). The apparatus described by Davenport et al. (1975) met all of these criteria, and is depicted diagrammatically in Fig. 6. Filtered sea water and deionized water supplies were fed into two polypropylene header tanks, the inlet of each tank being regulated by a domestic

ENVIRONMEKTAL SIMULATION EXPERIMENTS

149

FIG.6. Diagram of apparatus used to produce fluctuating salinity regimes. S.V., solenoid valve; CC, conductivity cell. From Davenport et al. (1975).

ball cock valve. The header tanks were connected, by wide bore P.V.C. tubes each guarded by a solenoid valve, to a 200 ml perspex mixing chamber where the water was magnetically stirred. Mixed water flowed from this chamber through long heat exchange coils of P.V.C. tubing immersed in a thermostat and thence to apparatus containing experimental animals. The salinity of the mixed water was monitored by a platinum conductivity cell connectdd via a Carwyn Instruments* salinity monitor to a chart recorder. A control seawater supply was taken from the seawater header tank and also passed through the thermostat before delivery to apparatus containing control organisms. The form of the salinity regime was determined by the programmed opening and closure of the two solenoid valves. The * Carwyn Instruments, Pentraeth Road, Menai Bridge, Gwynedd, U.K.

150

J . DAVENPORT

p r a p m r n e r was a L.K.B. 11300 Ultrograd,* an instrument normally used to control the mixing of relatively small volumes of liquids at low flow rates to produce chromatography gradients. The programmer was the heart of the apparatus, since the rest of the equipment was similar in principle to the apparatus used by Alderdice et al. (1958)to supply water of steady low oxygen tension to Pacific salmon eggs. Recognition of the fact that the programmer could be modified to operate larger valves than those for which it was designed ended a long period of frustration during which other mechanical or electrical programmers were considered, and discarded by the author because of their complexity or lack of flexibility. The Ultrograd operated by photoelectric scanning of a rectangular screen, divided into black and white areas by a cut out piece of black paper, is shown in Fig. 7 . The black paper programme was held over the screen by a glass plate. Each time the scanning device crossed the border between the black and white areas the valve controlling the flow of one liquid opened while the other closed. Sea water flowed when the scanner was traversing the black areas. Scanning of the concentration axis took place every 7*3s, while scanning of the time axis was adjustable between 15min and 16 days. When the scanner reached the end of its travel along the time axis i t automatically returned to the start of the profile and began again. The paper profiles were precisely located on the scanning screen, and near perfect accuracy and repeatability of salinity regimes was

0

12 Hours

24

Flu. 7. Diagram to illustrate cut-out scanning procedure of programmer. White area indicates paper removed. Arrowed path represents the track of the photoelectric scanner, but with the time displacement between scans greatly exaggerated. From Davenport et al. (1975). *L. K . B.-Produkter, ABS-161 25 Brommit 1 , Sweden

151

ENVIRONMENTAL SIMULATION EXPERIMENTS T.L.

18

12

6

0

Time (h)

FIG. 8. Conductivity chart trace to illustrate calibration accuracy and repeatability of concentration programming (flow rate 100ml/min). (1) Start of calibration programme, consisting of a stepwise increase in seawater concentration, each step representing 10% S.W. The seawater concentrations superimposed were obtained from seawater samples taken a t times corresponding to the midpoint of each step. (2) Start of programme to illustrate accuracy of programming. The regime produced consists of a drop in concentration from 100'$(oS.W.followed by an abrupt rise to 30% S.W.which was held for several hours. After an abrupt drop back to fresh water a 6 h sinusoidal rise back to 100yoS.W. was programmed. Solid circles and dashed lines represent the programmed concentrations; their positions on the chart trace were determined by the calibration data obtained from programme ( l ) . T . L . indicates the time lag caused by the space in the apparatus between the solenoid valves and the conductivity cell sensor. (3)Start of programme to illustrate repeatability by repeating the sinusoidal section of programme (2). From Davenport et al. (1975).

demonstrated, as shown by the test results displayed in Fig. 8. Since the description of the apparatus was published in 1975 the design of the equipment has not been altered, except that the deionized water supply, which proved to be very expensive to operate, was first replaced by a distilled water source and later in 1977 by a tap water supply, the water being filtered and passed through a charcoal filter before being delivered to a 2000 gal. (9080litre) storage vessel from which the salinity apparatus was supplied by centrifugal pumps. The L.K.B. Ultrograd is an expensive instrument, and this has led workers to use the same basic layout as Davenport et al. (1975), but with different (and hopefully cheaper!) programmers. Thus Atkins and Ritz (1977)described a salinity apparatus controlled by a programmer of their own design; this programmer also relied upon photoelectric scanning of a paper cut-out, was nearly as flexible as the Ultrograd, yet cost a great deal less. Hokanson et al. (1977)produced sinusoidal die1 temperature fluctuations with the aid of a cam

152

,J. DAVENPORT

controlled programmer which operated two solenoid valves which controlled hot and cold water supplies. Spaargaren (unpublished), at the Netherlands Institute of Sea Research, also designed and built an ingenious salinity fluctuation apparatus during the period 1974-75 in which the solenoid valves were controlled by a modified baker’s oven programmer! Instead of paper programmes, this apparatus employed drums formed from thin copper sheet and cut to an appropriate profile. Spaargaren’s apparatus also had the refinement of feedback control; a conductivity cell recorded the conductivity of the water delivered by the equipment, and if this deviated from the programmed value, then one or other of the solenoid valves delivering salt or fresh water was opened to compensate. Such feedback control is also a feature of the latest microprocessor controlled salinity fluctuation apparatus at the Department of Oceanography, Southamptom University (Lockwood, personal communication), and the computer-interfaced, multifactor simulation equipment under construction a t the Institute of Marine Environmental Research, Plymouth, U.K. (Bayne, personal communication). Feedback control has its dangers though. Sensors must be totally stable and reliable or the experimenter will not be aware of errors caused by drift of the sensor response! All the equipment described so far was designed to change only one environmental factor (either salinity or temperature), although Davenport et al. (1975) had suggested that simple modifications of their salinity apparatus would allow the experimental possibilities listed below: 1. Temperature fluctuations (of more complex form than those employed by Heath (1963) or Costlow and Bookhout (1971)); produced by programmed switching between two seawater supplies maintained at different temperatures. 2. Oxygen tension fluctuations; obtained by switching between deoxygenated and air (or oxygen) saturated seawater supplies. 3. Fluctuations in food availability for filter-feeding species; produced by mixing filtered sea water and sea water containing algae. 4. Pollutant fluctuations; obtained by mixing polluted sea water with pure sea water.

I n fact, although Ritz (1980)used the apparatus of Davenport et al. (1975) in slightly modified form to deliver water of fluctuating oxygen tension and salinity to intertidal amphipods, it was clear that completely anoxic conditions could not be accomplished with the

ENVIRONMENTAL SIMULATION EXPERIMENTS

153

apparatus as i t stood, since the minimum air saturation values attained by Ritz was 4%; temperature fluctuations could not be accomplished either, as long as water flowed from the thermally uncontrolled header tanks, through solenoid valves to a constant temperature thermostat. Because of these inadequacies, and spurred by a desire to be able to produce a greater variety of temperature regimes than attained previously, Davenport and Cawthorne (1978) designed and built apparatus which could deliver water of fluctuating salinity, temperature or oxygen tension. Its design is shown semidiagrammatically in Fig. 9 while some test results demonstrating its capabilities are displayed in Fig. 10. As in the earlier salinity apparatus (Davenport et aE., 1975), programming was accomplished

P

FIG.9. Apparatus of Davenport and Cawthorne (1978).Key: CC = cooling coil: CT = contact thermometer, EC = experimental chamber, FW = freshwater header tank, H = heater, MC = mixing chamber, P = pump, Prog = Ultrograd programmer, S = solenoid valve, SW = seawater header tank, T = tap.

Cycle 1

f 8

t

0

Hours

Cycle2

t

4

0

8

Cycle 8

s

0

0

8 Hours

16

FIG. 10. Apparatus test data from Davenport and Cawthorne (1978). (a) Repeatability of' temperature regimes. (b)Factor interaction. The results show a sinusoidal programme of8 h wavelength, high salinity being associated with low oxygen tension.

ENVIRONMENTAL SIMPLATIOX EXPRRIMESTS

155

by a L.K.B. 11300 Ultrograd which switched between two solenoid valves located close to the bases of two rigid PVC cylinders, each about 1.5 m tall, and 2.3 m in diameter, sealed at the bottom but open at the top. Sea water could be supplied to both chambers from a header tank regulated by a float switch and solenoid valve; alternatively one of the pair could be filled with fresh water instead from a freshwater header tank. The water within each cylinder was circulated by a centrifugal pump, while its temperature was regulated to a value between extremes of 3 and 50°C by a contact thermometer controlling both a 1.5kW glass heater and the supply valve of a resin-coated copper coil, through which flowed cold ( - 3°C) ethylene glycol from the laboratory supply. The cylinders and all piping were insulated with polystyrene and rubber foam. The tall cylindrical shape was adopted to provide sufficient pressure head to drive water flows of 300-1500ml/min, to minimize the distance between the temperature controlled water and the experimental vessel (which was also well insulated), and to reduce the water surface area for heatlgaseous exchange. The seawater or freshwater contents of each cylinder were temperature controlled and could be saturated with atmospheric air from a compressor, supersaturated by oxygen delivered from a gas bottle, or be deoxygenated by bubbling with oxygen-free nitrogen. Thus it was possible to deliver water which fluctuated in temperature, salinity, or oxygen tension either separately or in combination; i t was also possible to ensure that salinity fell while temperature rose, or that a falling oxygen tension was associated with rising temperature and salinity. However, factors could not be varied independently; the basic form of the factor fluctuations had to reflect the paper programme either directly or inversely. It was not possible to deliver water of slowly changing salinity and swiftly changing temperature. For simplicity’s sake the above description has dealt with a programmer controlling one pair of solenoid valves, and one pair of cylinders. Obviously it is quite simple to control several pairs of valves and cylinders with one programmer; this is especially valuable with temperature work where a single shape of profile (e.g. sinusoid) may operate between different temperature extremes. I n fact the apparatus described by Davenport and Cawthorne consisted of three pairs of cylinders. Obviously an apparatus delivering water of continually varying quality must be backed by continuous monitoring systems, which, in this case, consisted of platinum conductivity cells, thermistors and

+

+

156

J . DAVENPORT

oxygen electrodes. The apparatus has proved to be very useful in the thermal and salinity studies for which i t was primarily intended (e.g. Cawthorne, 1980). On the other hand, its use in oxygen tension fluctuation studies is still somewhat marginal as the water within such large cylinders required considerable quantities of nitrogen to maintain i t at zero oxygen tension. However, if a “stripping column” cylinder, of the type described by Fry (1951) and Alderdice et al. (1958) (i.e. tall, narrow and filled with glass chips t o give a high water-gas interface surface area), were substituted, great economies in the use of nitrogen (or other gases) should follow, and long term or routine oxygen tension studies would be possible. During the period of development of the temperature/salinityfoxygen tension apparatus by Davenport and Cawthorne, it was discovered that the mussel, Mytilus edulis L., could survive intermittent exposure to concentrations of copper, which would be lethal if delivered continually (Davenport, 1977).These observations were extended by Davenport and Manley (1978)who determined the threshold concentrations of copper inducing shell valve closure in Mytilus. During these investigations it was realized that a pollutant delivery system which had the following characteristics would be desirable: 1. It should be capable of supplying a flow of any steady concentration to experimental organisms. 2. It should be possible to supply water of increasing (or decreasing) pollutant concentration (at various linear rates of concentration change) to allow the assessment of threshold concentrations for organisms’ behavioural responses. 3. Mimicking of pollutant regimes occurring under field conditions should be feasible, where such information is available.

The need for providing the first is not generally appreciated. With some types of conservative, unreactive pollutant (e.g. detergents, acid pollutants, phenols) i t is possible to perform “steady state” tolerance or mortality studies, in the manner of salinity or temperature tolerance investigations, by placing organisms in vessels filled with polluted sea water, lids being necessary to prevent the loss of volatile components. However, with pollutants which are accumulated from the environment or eliminated by organisms (e.g. some heavy metals, free chlorine, radionuclides, organochlorine compounds), such experiments can be faulty unless the vessels used are very large in relation t o the organisms’ size. To illustrate this problem, consider a mussel placed in a 500 ml vessel filled with sea

ENVIRONMENTAL SIMPLATION EXPERIMEETS

157

water contaminated with 0.5p.p.m. added copper. No matter how long the mussel remains in the vessel i t cannot take up more than 0-25mgm copper. However, a similar animal placed in flowing sea water containing 0.5 p.p.m. added copper and filtering 50 ml/min would be exposed to 36mgm of copper passing through its mantle cavity each day. Not surprisingly, in the former situation the animal would survive indefinitely, whereas in the latter, the mussel would succumb in 1-2 days (Davenport, 1977). Requirement 1 could be satisfied simply by dosing pollutants at a constant rate into a constant flow of sea water, and this approach was adopted by Manley and Davenport (1979). However, to meet requirements 2 and 3 as well, a more complex apparatus was required; this has been described by Manley (1980) and is shown in Fig. 11. Again a L.K.B. 11300 programmer was used, but in this case i t was coupled with a L.K.B. 11300 “switch over” two position solenoid valve as in the normal, low flow rate chromatography application of this equipment. Teflon tubes from two constant head vessels led to the two position solenoid valve. One of the constant head vessels was supplied by a pump with deionized water from a stock tank, the other with concentrated pollutant from a similar stock vessel. The overflow pipes of the constant head devices returned excess fluid to the appropriate stock tank to prevent wastage. Inert materials were used throughout the construction of the apparatus. Mixed deionized water and pollutant were delivered from the solenoid valve at a flow rate of 12 ml/min into a constant flow of sea water (approx. 500ml/min) and thence to the experimental animals. Reliable linear pollutant gradients were produced by this apparatus (see Fig. 12),and it has already been used in heavy metal studies (Manley, in preparation); experiments with changing p H levels and high salinities are also envisaged. Having described the simulation equipment developed so far, future trends in such apparatus are worth consideration. It seems certain that the present mechanically based programmers will tend to be replaced with microprocessors which may be linked to computer or desk top minicomputers. If the microprocessors chosen have sufficient memory capacity, then they hold out the possibility of enormous flexibility and complexity in apparatus performance. Four alternative directions in simulation studies appear likely with such equipment. First, i t would be possible to continue with simple “idealized” regimes of the single factor type used so far, but with the minicomputer/microprocessor combination allowing several of such experiments to proceed independently, with different factors and timing in different parts of a single laboratory. This would

158

J . DAVENPORT

I1

-

7

I

I

I

I

I I T

I

I I I I

I

I I I

I I

I

FIG.11. Pollutant delivery apparatus; from Manley (1980). Key. A-animal

chamber, Cconstant head tanks (overflows returning to stock tanks), CW-unpolluted water a t constant flow, D S 4 i s t i l l e d deionized water stock tank, M-mixing chamber, Pperistaltic pump, PS-pollutant stock tank, T-teflon tubing, UCTultrograd, UVultrograd valve.

0 Time (minutes)

FIG.12. Test data from apparatus of Manley (1980).P-programmed

pollutant concentration, A-measured concentration in animal chamber of apparatus shown in Fig. 11, Mconcentration in mixing chamber.

ENVIRONMENTAL SIMULATION EXPERIMEKTS

159

substantially reduce the cost of individual experimental set ups, each of which require a separate programmer at present. Second, programmes to duplicate precisely the complexity of longer term fluctuations in one factor might be written; this is an approach adopted by Lockwood (personal communication) in mimicking the salinity changes which occur during a complete spring-neap cycle in a particular estuary. Thirdly, it is now theoretically possible to model the fluctuations in several environmental physico-chemical factors independently, so that the “natural” environment at a particular site may be closely simulated in the laboratory. This approach, though expensive, is somewhat analogous to the construction of hydrodynamic models used in the study of the likely effects of proposed dams, barrages etc., and may be of especial use in the simulating of the effects of pollutant delivery to a real estuary. Apparatus of this type is under development at the Institute of Marine Environmental Research, Plymouth (Bayne, personal communication), but has apparently only been used in fluctuating salinity experiments so far (Livingstone et al., 1979). Obviously, though, if five or six factors (e.g. salinity, temperature, oxygen tension, pH, light intensity and algal concentration) are all being continually changed in more or less independent fashion, then it will not be possible to attribute responses of organisms to any particular factor, and the simulation can only have a background function against which other experiments are carried out. Finally there may be possible applications of the simulation techniques in aquaculture. Most rearing studies for commercially important marine organisms have concentrated upon the use of optimum steady conditions, following the principles and using the analytical techniques of Box and associates (Box and Wilson, 1951; Box, 1954, 1956). However, such conditions may not be appropriate in all cases. For example, Ling (1969),in his study of the large tropical prawn, Macrobrachium rosenbergii (de Man), reported that the adult prawns, which normally live in fresh water, migrate to estuaries during the breeding season and spawn at quite high ‘salinities. The subsequent development of the young stages occur in brackish waters of various salinities. Given adequate information about the salinities associated with each larval stage of M . rosenbergii it would be technically feasible to duplicate these conditions in a hatchery/rearing establishment.

160

J . DAVENPORT

IV. REGIMES The main types of fluctuating temperature, salinity, oxygen tension and pollutant regimes employed so far in simulation studies are summarized in Figs 13, 14 and 15. From these i t is clear that there has been little standardization of regime form, although sinusoidal and square wave profiles (of tidal, 12 h or die1 wavelength) have been chosen independently by several workers. It is also obvious that all of the regimes displayed are of “idealized” form; none is a precise replica of real fluct$uations. Superficially the lack of standardization might seem undesirable, but close scrutiny shows that conventional “steady-state” experiments are equally chaotic in their design details; one researcher may use 2 deg C intervals in temperature tolerance studies, while another

I

0

I 24

1 48

--

0

24 0

24

Hours

FIG. 13. Examples of temperature regimes used in simulation studies. ( 1 ) Costlow and Bookhout (1971); (2) Thorp and Hoss (1975); (3) Hokanson et al. (1977); (4) Cawthorne (1979a).

161

EXVIKONMENTAL SIMCLATION EXPERIMENTS

might employ 5 deg C intervals basically, but use 1 deg C gaps in critical parts of the temperature range. In similar vein, pollutant investigations have been extremely variable in the duration of exposure and range of pollutant concentrations employed. Precise mimicking of the conditions at a particular site, on a particular tide or day, would seem to have limited usefulness; i t would have little value in comparative studies and in any case, as would be expected, conditions at a single site can change markedly from day to day, and certainly alter seasonally. An exact simulation is therefore meaningless, but regimes generally representative of certain estuaries might well be selected. Some of the regimes chosen by workers have obviously been dependent on the apparatus used. This is particularly true of the

x

-

0

L

0)

a

24

l'O-1nl0

24

0

24

0

1 0

24

nnn

0

12 0

12 0

Hours

12

nn

0

12 0

12

Hours

FIG.14. Examples of salinity regimes used in simulation studies. ( 1 ) Tucker (1970s);( 2 )Stickle and Ahokas (1974); (3) Davenport et al. (1975). FIG. 15. Examples of pollutant and oxygen tension regimes used in simulation studies. ( 1 ) Davenport (1977); (2) Ritz (1980);(3) Ritz (1980).

162

.I DAVENPORT

salinity studies of Tucker (1970a), Zachary and Haven (1973) and Stickle and co-workers (e.g. Stickle and Ahokas, 1974; Stickle and Howey, 1975;Findley and Stickle, 1978),whose equipment produced an asymmetrical pattern of falling and rising asymptotic changes in salinity by alternately flushing out an aquarium with dilute and concentrated sea water. The concentrations produced were predictable from the equations published by Wells and Ledingham (1940), who devised a technique for changing the salinity surrounding isolated preparations from the polychaetes Arenicola marina L., Nereis diversicolor (Muller) and Perinereis cultrifera (Grube). However, precise duplication of salinity regimes, especially at different laboratories, would be difficult since fluid flow rates and vessel sizes were factors in Wells and Ledingham’s equations. Since the data presented by Tucker (1970a) also showed that the regimes employed were not particularly closely representative of the environment from which her experimental animals were obtained (see Fig. 16), they seem to simply represent a somewhat arbitrary means of changing salinity in a relatively gradual manner. With the flexibility of more modern equipment there would appear to be little reason to persist with such an approach.

during two tidal cyclesat the estuary

8 4 1

0

,

I

1

I

I

I

2

I

3

sea water l l 4

1

Time, hours

FIG.16. Dilution of sea water in the environment and that produced in the laboratory. From Tucker (1970a).

The thermal regimes used by Hubbs (1964) were restricted by apparatus limitations in a similar fashion. However, the temperature programmes developed by Costlow and Bookhout (1971) and since used in several studies by Costlow and his co-workers (e.g. Christiansen and Costlow, 1975; Rosenberg and Costlow, 1976; Christiansen et al., 1977a, b; Lucas and Costlow, 1979) appear to be

EKVIRONMENTAL SIMULATION EXPERIMENTS

163

much more generally useful and repeatable. They also represent the major exception to the general use of square wave fluctuations in thermal studies. These programmes consist of a repeated sequence of 6 h at low temperature, 6 h of linear rising temperature, 6 h at high temperature and 6 h of falling temperature. The simple zig-zag die1 temperature changes used by Widdows (1976) appear not to have been used elsewhere, but were admirably suited to their design role in establishing whether the mussel Mytilus edulis could become acclimated to fluctuating temperatures. This work followed the earlier observations of Widdows and Bayne ( 197 1 ) that M . edulis could acclimate to changed but steady temperatures. Square wave and sinusoidal fluctuations appear to have particular advantages, especially where both are used for comparative work. They are symmetrical and so do not obscure asymmetries in animals’ reponses. Square wave profiles represent the most abrupt means of changing between two levels of a particular factor, while sinusoidal curves are among the more gentle means of accomplishing such transitions. Natural rates of change must fall between these extremes. The essential feature of the sinusoidal pattern is that the return to the mean level accelerates in proportion to the deviation from the mean, just as tidal height varies sinusoidally with time. Moreover the field salinity data presented by Stickle and Denoux (1976), the rock pool oxygen tension measurement of Daniel and Boyden (1975),and the salinity and temperature results presented by Cawthorne (1979a) all approximate to sinusoidal curves. A further advantage of size and square wave profiles may be appreciated from the salinity regime programmes illustrated in Fig. 17. Sinusoidal ( P l ) and square wave (P2) profiles of similar wavelength and amplitude are shown. Because areas A1 and A2 are equal, the quantities of salt and water delivered at a given flow rate over a complete cycle are the same for both types of profile. This means that animals exposed t o the P1 profiles will have access to the same total amount of salts and water during a simulated tidal cycle as animals exposed to the corresponding P2 profiles, but will not be exposed to the sudden osmotic shocks and long periods at low sea water concentrations characteristic of the P2 profiles. Similar considerations obviously apply to temperature, oxygen tension or pollutant regimes. If square wave and sinusoidal profiles or some other patterns of change are adopted in similar studies, there still remains the question of appropriate wavelengths and amplitudes. The situation for salinity investigations is relatively uncomplicated; except in bodies

164

J. DAVENPORT

7

100-

r.

so-

vi

8

0.

t,

I

12

Hdurs

FIG.17. Diagram to illustrate the two types of fluctuating salinity programme used. P1: this programme produces sinusoidally fluctuating seawater concentrations of near tidal frequency; P2: this programme delivers water of abruptly changing seawater concentration; A1 and A2 are equal areas; this means that animals exposed to either type ofsalinity regime will have access t o similar total amounts of salt and water a t a given flow rate. From Davenport et al. (1975).

of water cut off from the sea for periods of days or weeks (e.g. lagoons, salt marshes) which tend to change their salinity slowly, salinity fluctuations are normally of tidal wavelength. The only decision which then has to be made for laboratory simulation studies is whether a precise tidal wavelength should be mimicked or whether the more experimentally convenient 12 h wavelength should be adopted. So far 12 h wavelengths have been almost universally used (e.g. Stickle and Ahokas, 1974; Davenport et al., 1975; Livingstone et al., 1979),although Stickle and Howey (1975)used a tidal 12 h 25 min wavelength regime. This approach is also being adopted by Lockwood (personal communication). A few regimes of rather odd wavelength have also been used. Tucker (1970a) employed a short (4 h ) regime which appears to have been a simulation of part of an estuarine tidal cycle, and therefore reasonably realistic. Findley and Stickle (1978), in their study of the haemolymph composition of the blue crab Callinectes sapidus Rathbun, used a 24.8 h wavelength salinity cycle variously described in their paper as tidal and diurnal, yet without field data to justify its double tidal duration. The amplitudes of salinity change employed in simulation studies will

ENVIRONMENTAL SIMULATION EXPNRIMESTS

165

obviously depend upon the euryhalinity of the species being studied although amplitudes greater than those likely to be encountered in nature may be of value for comparative purposes in some investigations where the responses of euryhaline and relatively stenohaline species are being compared (e.g. Shumway, 1977a). Appropriate wavelengths for thermal fluctuation simulation studies pose problems. In terrestrial and freshwater investigations, diel temperature fluctuations have been used (e.g. Edney, 1964; Feldmeth et at., 1974; Hokanson et aE., 1977) and these make sense since warm days tend to alternate with cold nights. I n marine, estuarine and intertidal environments the situation is more complex. In shallow coastal waters temperatures may fluctuate in diel fashion, but sublittorally in estuaries temperature changes are associated with salinity changes (Cawthorne, 1979a) and are therefore of tidal wavelength. I n the intertidal zone temperature fluctuations result from a mixture of tidal and diel influences. Aerial emersion provides thermal shocks of tidal periodicity, but the direction and magnitude of the shock are determined by diel influences. Costlow and Bookhout (1971) employed diel temperature changes in studies upon larvae of the mud crab Rhithropanopeus harrisii (Gould); these would appear appropriate for the inshore environment near Beaufort, N. Carolina, where the experiments were carried out, but not for the Miramichi estuary in Canada where the species has been observed by Bousfield (1955). Sastry (1978) also exposed marine crustacean larvae to diurnal temperature changes which were justified by the field data of Hillman (1964). Oxygen tension and pollutant studies have been few in number. Ritz (1980) used both sinusoidal and square wave oxygen tension regimes of 12 h, near tidal wavelength. These were associated with 12h wavelength salinity fluctuations with low salinity and low oxygen tension coinciding; this situation simulates an estuary where the freshwater input is organically polluted. From the work of Daniel and Boyden (1975) it would seem that rockpool oxygen tensions tend to fluctuate diurnally, but this pattern is distorted when the tide reaches the pool-an analogous situation to that described above for intertidal temperatures. Pollutant fluctuation experiments appear to have been limited to those of Davenport (1977) and Ritz (1980) who both employed square wave copper regimes of 12 h wavelength. These were rather arbitrary profiles intended to simulate a situation where animals were alternatively exposed to clean and polluted seawater on a tidal basis. No field data concerning short term pollutant fluctuations appear to be available.

166

.I. DAVENPORT

Before leaving the topic of regimes a little space needs to be devoted to the usefulness of linear profiles. Linear regimes consist of constant rate changes in factor level or concentration; an example of a linear pollutant concentration gradient may be seen in Fig. 12. Of course such changes, whether in temperature, salinity or other factors, are not likely in nature, but they are useful in accurately establishing threshold concentrations for behaviour or tolerance and studying the effects of different rates of change of factor level upon such concentrations. Linear regimes are therefore useful in augmenting the data gained from the use of more realistic simulations.

v. TEMPERATURE EXPERIMENTS The literature devoted to the study of direct and indirect thermal effects upon marine and estuarine animals is extensive and has been reviewed on many occasions. Relatively recent reviews include the excellent comprehensive articles of Kinne (1970) and Newell and Branch (1980) on invertebrates and Brett (1970) on fish. Of course much of this literature is devoted to single and multifactor steady-state experiments. Newell (1969) reviewed the effects of fluctuations in temperature on the metabolic processes of intertidal invertebrates, but the time scales of the experiments described were all closer to the seasonal pattern of temperature change rather than the short term tidal and die1 fluctuations which are of interest here. In any case all of the results described by Newell were derived from conventional steady-state experiments. The majority of fluctuating temperature studies on marine and estuarine animals have emanated from the Duke University marine laboratory at Beaufort, North Carolina, where they originated with the study of Costlow and Bookhout (1971). Other researchers have employed cyclic temperatures in their studies, but the animals concerned, mainly fish, have been freshwater (e.g. Feldmeth et al., 1974; Hokanson et al., 1977). However, as Kinne (1970)pointed out, the thermal reactions of marine and freshwater forms are often similar so such studies will be referred to where appropriate. For background information about the responses of freshwater fish to temperature, an earlier review by Fry (1967) is still most useful.

A. Survival The upper and lower lethal temperatures of aquatic organisms have been measured in two ways. The less common method used by

ENVIRONMENTAL SIMULATION EXPERIMENTS

167

Tamura (1944), Tsukada and Ohsawa (1958), Southward (1958), Tsukuda (1960) and Crisp and Ritz (1967), involves gradually altering the animals’ ambient temperature and periodically sampling for mortality. The more widespread approach has been the instantaneous transfer steady-state technique which may be regarded as being analogous to the classic LD50 pharmacological assay method (Kinne, 1970). Lower and upper lethal limits are expressed as the temperature which kills 50% of a test population within a particular period-usually 24 h (LT5&24 h). Plotted temperaturemortality data are normally of sigmoid form (e.g. McLeese, 1956; Mihursky and Kennedy, 1967) and susceptible to Probit analysis which enable researchers to calculate LT50s with considerable precision. Thermal tolerance simulation studies performed so far fall into two categories. I n the first are those which mimic the sudden thermal stresses imposed by entrainment of organisms in the cooling systems of electricity generating plants. Of course the designs of such experiments grade imperceptibly into the conventional steady-state experimental format, so a somewhat subjective decision has had to be made to decide which studies merit discussion here. It should also be stressed that damage of organisms by power station entrainment is rarely simply a matter of exposure to transient increases in temperature. Usually the combined effects of thermal stress and chlorination are responsible (Waugh, 1964; Heinle, 1969; Morgan and Stross, 1969; McLean, 1973; Muchmore and Epel, 1973; Hoss et al., 1975).There are also pressure fluctuations of considerable magnitude and heavy metal contamination of the cooling flow is not uncommon. Heinle (1969) exposed specimens of the copepods Acartia tonsa Dana and Eurytemora afJinis (Nordqvist) to sudden thermal shocks. He took considerable care to ensure that temperature increases took place rapidly ( < 20 deg C change within 5-10 min) but sustained the subsequent elevated temperatures for 24 h-an unlikely situation in the field where thermal discharge plumes lose heat to the atmosphere and mix quickly with colder water. Not surprisingly, both species survived a given temperature increase quite well if they were living in cold water, but not if their native water was warm. Thus, specimens of E . afinis living at 5°C survived an increase of 20 deg C quite normally and even an increase of 25 deg C did not kill 50% of the copepods in 24 h. However, animals acclimated to 25°C could not survive a 5 deg C increase for 24 h although rather revealingly there was no mortality during the first 4 h of exposure to elevated temperature.

168

J DAVEXPORT

Diaz (1973, 1975), in his studies on bivalve larvae, used a more accurate simulation of power station entrainment. In 1973 he worked with the larvae of the American Oyster Crassostrea, virginica (Gmelin) which were reared and maintained a t +25"C. To simulate entrainment the larvae were passed, at a concentration of 5-10 individual/ml, through a tube placed in a constant temperature bath. This ensured exposure for 5s to a particular elevated temperature. Three to four litres of sea water containing larvae which had received a thermal shock were then placed in a jar and cooled t o 25°C over a period of 30 min. Similar experiments were performed upon larvae of the bivalve Mulinaria lateralis (Say) but with a basic temperature of 20°C. The experimental design appears to be quite representative of the thermal aspects of the power plant entrainment situation. Heating is almost instantaneous, as in condensers, while cooling is quick, as seems likely in rapidly dispersed thermal plumes. A similar experimental procedure has been adopted more recently by Sherberger et al., (1977) who have simulated the type of thermal shocks encountered by drifting aquatic insects which cross a thermal plume. The major results derived from Diaz' studies are displayed in Fig. 18 and Table 11. The data for Crassostrea virginica demonstrate that the deleterious effects of a brief exposure to elevated temperature may not be expressed for some time (4 days in this case). This could not have been predicted from steady-state experiments, and also casts a certain amount of doubt upon the usefulness of LT50-24 h determinations. The results for Mulinaria lateralis displayed in Table I1 indicate that susceptibility to thermal shock may alter with developmental state; 2-day-old larvae appeared to be relatively

+

+

TABLE 11. MORTAI~ITIES o~ LARVAE OF Mulinia lateralis 48 H AFTER BKIEF Elrohr.rt~1'0 THREE ISC'KLSASES IN TEMPEKATVRE ABOVE 20°C (SELEVTEI) FKOM THE DATA OF DIAZ. 1975)

Age of larvae (days) 2 4 6

8 10 12 Mean

0 deg C' increase (control) yo mortality

10 deg C increase yo mortality

15 deg C incrrase yo mortality

20 dsg C increase Yo mortality

1.5 9.8 1.3 2.6 1.8 1.6

7.5 28.5 9.1 27.8 209 23.8

23.1 15.6 34.1 33.6 36.7 27.0

41.9 38.8 77.1 63.9 42.7 44.8

3.1

19.6

28.4

51.5

ENVIRONMENTAL SIMULATION EXPERIMEKTS

0

2

6

10

DAYS AFTER

14

18

169

22

EXPOSURE

FI(:. 18. Cumulative mortalities for 24 days of American oyster (Crussostrea virginicu) larvae

exposed for 5 s to three increases in temperature above 25°C when 3 days old. From Diaz (1973).

unaffected by temperature increases, while 6-day-old larvae tolerated a 10deg C increase particularly well but were affected badly by a 20 deg C increase. This species did not die until 2 days after a temperature shock. Delayed mortality is perhaps t o be expected. Although proteins coagulate leading to cell death (Nassonov and Alexandrov, 1943) at high temperatures, most somatic cells can tolerate substantially higher temperatures than the intact individual (Ushakov, 1968).For vertebrates, heat death of the individual generally seems to be caused by interference with the working of the central nervous system (Fry, 1967). Control processes appear to be less centralized in invertebrates and peripheral activity of bivalve larval tissues may persist for some time after the CNS has been irreversibly damaged. A more recent entrainment simulation study has been carried out upon larvae of a teleost fish by Middaugh et al. (1978). Much of this study was concerned with the effects of brief exposure to residual chlorine. However some investigations on thermal shock were conducted upon embryos and larvae of the euryhaline, eurythermal estuarine mummichog, Fundutus heteroclitus L. The animals were reared at 24°C and exposed to 5 or 10 deg C temperature elevations lasting 7.5, 15, 30 or 60min before a return to 24°C. Mortality was assessed 24 h later. All embryonic and larval stages showed increased mortality with both amplitudes of temperature increase. I n embryos

170

J . DAVEXPOKT

longer exposure to elevated temperature did not increase mortality, though it did in newly hatched and 7-day-old larvae. However the latter point is somewhat academic; Middaugh et al. admit that only the embryos are likely to be entrained in nature. The second category of temperature tolerance simulation study encompasses investigations where cyclical temperature regimes have been employed. It has long been known that organisms often survive changing temperatures better than they do constant temperatures (Allee et al., 1949). However the earliest laboratory tolerance experiments with repeatable regimes upon aquatic organisms were those of Heath (1963) and Hubbs (1964).Heath was interested in the difference in critical thermal maximum (C.T.M. as defined by Lowe and Vance, 1955) temperature between species of cutthroat trout, Salmo clarki clarki previously acclimated to either constant or cyclic temperatures. Fish acclimated to diel square wave fluctuations between 10 and 20°C survived to a significantly higher C.T.M. (29-77°C)than trout acclimated to 15°C (C.T.M.29.06"C).Fish kept at a constant 10°C had a C.T.M.of 27.63"C,while trout held at 20°C had a significantly higher C.T.M. (29.88"C)very similar to that of temperature cycled fish. Hubbs (1964)showed that eggs and larvae of the freshwater teleost Etheostoma lepidum Baird and Girard survived diel fluctuations between 4 and 12°Cjust as well as at a constant 9"C, yet constant exposure to 7°C was lethal. Again this illustrates the enhanced survival often caused by thermal fluctuations, although it should be remembered that instantaneous, and usually unphysiological, thermal shocks may kill fish (Fry, 1957). The abrupt fluctuations used by Heath (1963) were small in amplitude and well within the thermal tolerances of cutthroat trout, while the temperature changes employed by Hubbs (1964) were quite slow (c. 2 deg C/h). More recently Feldmeth et al. (1974), working on the pupfish Cyprinodon nevadensis amargosae Eigenmann and Eigenmann, which inhabits shallow thermally unstable desert streams, have demonstrated an increase in temperature tolerance at both ends of the temperature scale induced by prior acclimation t o diel temperature fluctuation regime (15-+35"C;7 h a t low temperature, 5 h of warming, 7 h at high temperature, 5 h of cooling et seq). These results conflicted with an earlier hypothesis of Brett (1944) which suggested that both upper and lower lethal temperatures were determined by the maximum temperature experienced during a diel cycle. How does prior acclimation to cyclic rather than steady regimes extend thermal tolerances in fish? Studies upon acclimation to steady

EKVIRONMENTAL HIMULATIOX EX PERIMEKTS

171

temperatures in fish (e.g. Hochachka and Somero, 1968; Hochachka and Lewis, 1970; Moon and Hochachka, 1971) indicate that acclimation occurs by elaboration of isoenzymes uniquely suited to metabolic function at a particular environmental temperature. Feldmeth et al. (1974) suggest that the pupfish acclimated to cyclic temperatures simultaneously induce separate isoenzymes for both warm and cold conditions, thus gaining survival attributes not seen in fish acclimated to steady average temperatures. Unfortunately this type of study, where survival of extreme temperatures is tested after a period of exposure to moderate thermal fluctuations, appears to have been carried out only upon freshwater fish or migratory fish during the freshwater phase of their life cycle. There would appear to be no reason to expect marine fish not to have similar responses. Certainly a similar response has been demonstrated in a very different group-the terrestrial isopods. Edney (1964) worked on Porcellio laevis Latreille and Armadillidium vulgare Latreille. Isopods acclimated for 2 weeks to a diel regime of 12 h at 10°Cand 12 h at 30°C had a maximum lethal temperature similar to animals acclimated to a steady temperature of 30°C while the lower lethal temperature of the cyclically treated isopods approximated to that of animals acclimated to a steady 20°C. I n other words, the animals held in cyclical regimes had a total temperature range greater than that of isopods held at any single temperature. Cyclic temperature studies on marine and estuarine organisms appear to have started with the investigation of Costlow and Bookhout (1971) on larval development in the mud-crab Rhithropanopeus harrisii. These workers compared the effects of a variety of diel temperature fluctuations and constant temperatures upon survival and duration of larval stages of R. harrisii; the survival results obtained are displayed in Table 111. Clearly Costlow and Bookhout's approach differs somewhat from that employed in the freshwater and terrestrial studies described above. Instead of investigating how prior acclimation to modest temperature fluctuations affected tolerance of extreme temperatures, these workers studied the larval tolerance of the cycling temperatures themselves, some of which approached the lethal limits of the animals. Larvae were reared from hatching to the first crab stage at a constant salinity of25%,. Control larvae were kept at steady temperatures of 15,20,25, 30 and 35°C. Experimental larvae were either exposed to 5 deg C fluctuations (15 + 20°C, 20 -+ 25"C, 25 + 30"C, 30 + 35°C and 35+4OoC) or 10 deg C changes (15+25"C, 2O+3O0C and 25+35"C). The picture that emerges from the results is rather complex. Survival

172

J. DAVENPORT

TABLE 111. SURVIVAL OF LARVAL STAGES O F R . ha?'riSii CULTURE]) AT 250- (FROM COSTLOW AND BOOKHOUT. 1971 Hatch to megalopa

Megalopa to crab

Hatch to crab

No. of meae

Temperature ( " C )

No.

%

No.

Yo

No.

%

150 42 1 442 434 100 400 400 400 400 100 650 lo00 1150

15 20 25 30 35 1&20 2&25 25-30 30-35 3540 15-25 2&30 25-35

87 380 389 288 5 253 337 329 362 0 503 826 733

58.0 90.3 88.0 664 50 63-2 842 82.2 90.5 0 77.4 82.6 63.7

32 35 1 359 252 2 230 306 292 334 0 437 738 525

368 92.3 92.2 87.5

32 351 359 252 2 230 306 292 334 0 437 738 525

21.3 83.4 81.2 58.1 2.0 57.5 765 73.0 83.5 0 67.2 73.8 457

40 90.9 908 88.7 92.2 0 869 89.3 7 1.6

to the crab stage in the 15 -+ 25°C cycling regime (67.2%) was better than the survival at a constant 15°C (21.3y0) while the regime fluctuating between 25 and 35°C was tolerated better (45.7% survival) than a constant 35°C (2.0% survival). Again extremes of fluctuating temperature appear to be tolerated better than extreme constant temperatures. However, survival in the 20 -+ 25°C regime was slightly poorer than at a steady 20 or 25°C indicating that the same situation did not hold at less stressful temperatures. Finally the survival in the 30 -+ 35°C regime (83.5%)was far better than that at either a constant 30 or 35°C (58.1% and 2% respectively). This last result seems particularly anomalous since the survival of the 30 -,35°C regime represents the maximum survival to metamorphosis for any of the experimental series of larvae. Costlow and Bookhout could only speculate that this regime might eliminate pathogenic organisms not killed by constant temperatures, or alternatively, that enzyme systems or physiological functions might in some way be particularly favoured by it. In a later study, Christiansen and Costlow (1975) were unable to duplicate this particular result with Rhithropanopeus harrisii so perhaps i t should be regarded as suspect. I n a later study (though published earlier) Regnault and Costlow (1970) obtained results for the larvae of the shrimp Crangon septemspinosu Say. Broadly speaking the results were similar to those obtained for R . harrisii with extremes of

ENVIRONMENTAL SIMULATIOK FXPEKIMESTS

173

temperature being tolerated better if the animals were given intermittent respite at less stressful temperatures. I n subsequent studies, largely on R. harrisii, Costlow and his coworkers (Christiansen and Costlow, 1975; Rosenberg and Costlow, 1976; Christiansen et al., 1977a, b; Lucas and Costlow, 1979) have concentrated upon the effects of cyclic temperature on larval development and on the interaction of cyclic temperature with the effects of toxic compounds (heavy metals, insecticides).These studies will be discussed elsewhere in the review. As part of a broader study, Widdows (1976) showed that adult mussels, Mytilus edulis, could sustain filtration, and hence survive, in a cyclic temperature regime fluctuating between 21 and 29°C. At a constant 29"C, however, filtration was halted and long term survival presumably impossible. The most recent study devoted purely t o the effects of fluctuating temperature upon the survival of marine organisms is that of Cawthorne (1980). He studied nauplius larvae of the barnacles Balanus balanoides L. and Elminius modestus Darwin and compared their tolerances to: (a) steady temperatures, (b) temperature fluctuations of square wave form and 12h wavelength, (c) temperature changes of sinusoidal form and 12h near tidal wavelength. The data derived from this investigation are displayed in Fig. 19. They consist of sigmoid curves relating the maximum temperature encountered in a given temperature regime to the mortality induced by that regime. The LT50-24 h values calculated from these curves are set out in Table IV. These results show once again that survival a t extreme temperatures is better in animals exposed to cyclic rather than steady temperatures and resemble the data collected for a desert population of a freshwater teleost Gambustia a j j n i s a j j n i s (Baird and Girard) (see Otto, 1974). Additionally i t is clear that gentle (sinusoidal) salinity changes are tolerated better than abrupt fluctuations. Particularly interesting is the disparity between the two species in their response to fluctuating temperatures; although Balanus balanoides and Elminius modestus exhibit very similar upper lethal temperatures when exposed to constant conditions, E . modestus is much the more tolerant of high temperatures when the latter are encountered briefly or intermittently. Fry (1947) defined the limit of the zone of activity or thermal tolerance as the point beyond which an organism is unable to

TABLE I v . MAXIMITM TEMPERATURES ("c)CAITSINC: 50% MORTALITY IN C I I t R I P R D E ~ A Z I P I ~ I I T S LARVAE EXPOSED TO VARIOUS TEMPERATIJRE REGIMES. 95% CONFIDENCT INTERVALS IN BRACKETS. (FROM THE DATA OF CAWTHORNE, 1979a, 1980) Type of temperature regime (a)Steady state (b) Square wave

Species

( c ) Sinusoid

~~

32.2 ( 0 6 ) 32.4 ( 0 9 )

Balanus balanoides Elminius modestw

36.6 (04) 388 ( 0 4 )

345 (0.5) 3 5 5 (0.5)

100(a)

0-r

I

31

I

32

33

34

35

i

I

36

37

38

39

40

41

42

Maximum temperature ("C

Maximum temperature ("C)

FIG.19. Temperatureof newly released nauplius larvae of (a)E. modestus and (b) B. balunddes. Circles indicate animals exposed to sinusoidal temperature changes of 12 h wavelength between 10°Cand various highertemperatures;squares indicate animals exposed to squarewave temperature changes; triangles represent animals exposed to steady elevated temperatures. Modified from Cawthorne (1979a).

ENVIRONMENTAL SIMULATION EXPERIMENTS

175

maintain its activity and survive indefinitely. This definition is clearly relevant only to constant thermal conditions which rarely occur in coastal and estuarine marine habitats. B. Development A large number of steady state investigations into the effects of temperature on development and growth in aquatic fish and in vertebrates have been carried out (see Gray, 1929; Sandoz and Rogers, 1944; Precht et al., 1955; Costlow et al., 1960;Costlow, 1967; Ong and Costlow, 1970; Alderdice and Velsen, 1971; Brett, 1970; Kinne, 1971;Lough 1975;Peterson et al., 1977 for some examples and reviews). It has long been known that development in most ectothermic animals is accelerated at higher temperatures (Belehradek, 1935), but the effects of varying temperatures are more complicated. Khan (1965) lists three possibilities, mainly derived from the freshwater studies of Grainger (1959): ( 1 ) Development proceeds at a rate to be expected from the results of constant temperature experiments. Here the alterations in temperature do not themselves affect the rate, and embryonic development is, for instance, speeded up by the expected amount if the temperature is raised. (2) Development does not proceed at a rate predictable from steady temperature studies. Here the alterations in temperature have an effect on developmental rate at the time they are taking place. This has been called an immediate effect (Grainger, 1959). (3) Like (2) development does not proceed at the expected rate because the temperature experienced in early development retards or accelerates subsequent development. This was described as an after effect by Grainger.

Khan’s classification appears to have been ignored in cyclic temperature studies on marine and estuarine animals, as have the detailed and tedious methods she employed in her temperature studies on the freshwater crustacean Acanthocyclops viridis. Costlow and Bookhout (1971)found that developmental times (as distinct from survival) of larval stages of Rhithropanopeus harrisii were barely affected by fluctuating temperatures; development proceeded at a rate intermediate between the rates occurring at constant temperatures corresponding to the extremes of the tem-

176

J DAVEXPORT

Hatch to crab

Megalops to crab

50 40

Hatch to megalops

PI:. 20. Comparison of time required for development of larval stages of R.harriaii in the laboratory at constant temperatures and 5 and 10°C daily cycles of temperature. From Costlow and Bookhout (1971).

perature cycle (see Fig. 20). There was some indication that 10 deg C fluctuations slightly slowed development but this was certainly not the case with 5 degC changes.It would appear that larvaeof R.harrisil: are well adapted to temperature fluctuations which are a feature of their coastal/estuarine habitat. A similar situation was found for the larvae of the shrimp, Crangon septemspinosa by Regnault and Costlow (1970). I n contrast, Sastry (1978)found that the duration of development of larvae of the sublittoral crab, Cancer irroratus Say was affected by fluctuating temperatures; zoeae developed faster and megalopae developed slower when exposed to 1O+2O0C or 12.5 + 17.5"C than they did at a constant 15°C. However, both of these amplitudes of temperature fluctuation may be somewhat

ENVIRONMENTAL SIMULATION EXPERIMENTS

177

unphysiological, since Hillman (1964) recorded diurnal variations of only k 2 deg C in the environment of these larvae. Recently, Lucas and Costlow (1979) have carried out a cyclical temperature study on some molluscan larvae-veligers of the intertidal and estuarine prosobranch gastropod Crepidula fornicata (L.). As for Rhithropanqeus harrisii growth under cyclic temperature regimes was intermediate between that observed in veligers living at corresponding constant temperatures (see Fig. 21). Given that only a small number of investigations have been performed so far, this would appear to be the usual pattern for animals normally exposed to temperature fluctuations in the wild, and consequently well acclimated. C. Reproduction Although several studies have shown the importance of raised temperatures to the induction of spawning in bivalves (e.g. Loosanoff, 1945a; Loosanoff and Davis, 1950; Gruffydd and Beaumont, 1970) there appears to have been only one investigation so far in which the effects of fluctuating temperature upon reproductive processes have been monitored. Cawthorne and Davenport (1980) investigated the effects of square wave and sinusoidal tidal temperature cycles u,pon the release of larvae by adults of two common barnacle species, Balanus balanoides and

30 t

- 200 E

I

c

E?!

/;:!:!

B

25O-3OoC

/

OCOEP 2Oo-25OC

.C

x 0 E

100

P

OCOEP15°-200C

0 0

6

12

Age (days)

FIG 21 Crepnddaforntratn Growth of vehgers at varioub constant temperatures and 5deg (' temperature cycles of equal penodicity (COEP) Values are from measurements of 75 larvae From Lucas and Costlow (1979)

178

250200I

j4

. I . DAVENPORT

500-

(b)

r--,f&+ '

100-

k!-i,i

..........

:I

!/I

Llkl ' I !

:

'Y.1

.Sl$V'

#4-1

400.

(bl r - - - - - - - - - ~ .~

I

!

~

0,

'

,

'

,

'

,

'

1

0

'

..........

ENVIRONMENTAL SIMULATION EXPERIMENTS

179

produced by abrupt temperature changes would have advantages ecologically. Abrupt temperature changes of significant amplitude only happen in the intertidal habitat when immersion or emersion occurs (Cawthorne, 1971a). Because E . modestus closes its opercular valves and effectively isolates its mantle cavity in response to emersion, any enhanced hatching triggered by temperature change at this time cannot be expressed as release until the animal is reimmersed. Temperature changes associated with immersion generally tend to be decreases in the summer and increases in the winter although die1 or climatic influences may sometimes modify this pattern. Because E. modestus breeds throughout the year a hatching response to sharp changes of temperature in either direction will always ensure that most larvae are released on a rising tide. This is particularly important in the estuarine habitat where rising tides tend t o be associated with rising salinity. The experiments which led to the results recorded in Fig. 22 were carriedout in winter; it is noticeable that decreases in temperature were more effective in triggering heavy larval release than rising temperatures. Possibly this pattern reflects a degree of seasonal acclimation.

D. Adaptation Precht et al. (1955) showed that ectothermic organisms held at a new temperature for some days show changes in the activity and concentration of some enzymes and other substances; these changes are often associated with changes in oxygen consumption. The initial stages of this process of adaptation were studies by Grainger (1956, 1958) who was interested in the immediate metabolic responses of various crustaceans to abrupt temperature change. Kinne (1964a) reviewed the then extant literature devoted to non-genetic adaptation using the various concepts of temperature acclimation proposed by Precht (1958) and since adopted by many workers (e.g. Alderdice, 1972). Not until the work of Widdows (1976) was the question of adaptation to cyclic temperatures tackled for any marine organisms. Widdows and Bayne (1971) and Bayne (1976) had previously shown that the intertidal mussel Mytilus edulis could adapt completely (Precht Type 2) in its filtration rate to gradual temperature changes between 2 and 20°C. Obviously this allowed feeding activity to continue independent of season. Widdows (1976) showed that specimens of M . edulis acclimated their rates of filtration and oxygen consumption t o cyclic temperature fluctuations (11 + 19°C) by

180

J . DAVENPORT

reducing the amplitude of response over a period of about 2 weeks (see Figs 23 and 24). Compensation was only partial for oxygen consumption but complete for filtration rates, which became independent of the temperature changes. The oxygen uptake and filtration rates of animals adapted to cyclic temperatures did not differ significantly from the responses of mussels adapted to equivalent constant temperatures. This agrees well with the observations on growth and development of eurythermal organisms reported in the previous section. Widdows also showed (see Fig. 23) that acclimation to cyclic temperatures occurred over a similar period to constant temperature adaptation. Two suggestions to explain this were put forward. The first was that only a brief exposure to a new temperature was required to stimulate the initiation of the adaptive response, although presumably the stimulus had to be repeated at intervals to sustain adaptation. Alternatively, the mussel might integrate its response over the whole temperature cycle to become independent of thermal extremes. Finally, Widdows

0'7[

I

FIG.23. Thermal acclimation of oxygen consumption by Mytilus edulis. (a) Rate of oxygen consumption in response t o cyclic temperatures. TF (11-19°C); 1I T , 15"C, 19°C.(b) Rate of oxygen consumption in response t o constant temperatures, 11"C, 15"C, 19°C. Mean1S.E. From Widdows (1976).

ENVIRONMENTAL SIMULATIOK EXPERIMENTS

I

0

I

I

I

I

I

I

I

20

10

I

I

I

I

181

I

30

Days

FIG.24. Thermal acclimation of filtration rate of Mytilus edulis. (a) Filtration rate in response to cyclic temperatures, TF (11-19°C); 1 1 T , 15"C, 19°C. (b) filtration rate in response to constant temperatures. 11"C, 15"C, 19°C. Mean+S.E. From Widdows (1976).

demonstrated that animals taken from relatively constant temperature environments and tested immediately showed metabolic temperature-dependent responses, whereas mussels collected near a power station which imposed marked temperature fluctuations on them exhibited temperature-independence, thus confirming that the conclusions from the laboratory cyclic temperature studies were applicable in the field. In contrast to Widdows' results, Sastry (1978), who worked on larvae of the crab Cancer irroratus, found that oxygen consumption in animals exposed to cyclic temperatures (10 -,20°C) was shifted towards values characteristic of larvae living at 20°C rather than towards those applicable to the mean value of 15°C;Sastry suggested that enzymes which were inactive at 15°C might become activated during temperature cycles which periodically exceeded 15"C, and that this activation might allow increased metabolic activity as suggested by Somero (1969). Sastry also showed that there were significant differences between the activities of various enzymes

182

J. DAVENPORT

(lactate dehydrogenase, malate dehydrogenase and glucose-6phosphate dehydrogenase) in cycled and constant temperature crab larvae. However, the reasons for these differences were and remain obscure, particularly as there appeared to be no correlation with oxygen consumption.

E. Interaction with other factors Since response surface techniques were introduced by Box and Wilson (1951), a large number of marine biologists have used such representations of multiple factor studies. I n many of these investigations, and also in simpler two factor studies, temperature has been a variable (e.g. McLeese, 1956;Costlow et al., 1962;Crisp and Costlow, 1963; Alderdice, 1963; Forrester and Alderdice, 1966; Alderdice andvelsen, 1971;Ahokas and Sorg, 1977).The most recent major reviews of the field appear to be those of Kinne (1971) and Alderdice (1972), the latter being concerned mainly with the principles and mathematical techniques involved in response surface methodology . Studies involving the interaction of cyclic temperatures with other factors (except for biological factors such as age or stage of development) have been few in number. So far no special method of analysis or presenting the results has arisen and no-one has yet had the temerity to draw conclusions from experiments in which more than one factor was cycled! Regnault and Costlow (1970) investigated the effects of cyclic temperature upon larvae of Crangon septemspinosa at two salinities, 20 and 30%,. The enhanced survival shown in response to temperature cycles by comparison with that exhibited at equivalent constant temperatures was more marked a t 30%,than at 20%,.Thorp and Hoss (1975)employed a much wider range of salinities ( 5 -,35%,) in their studies upon adults of two species of grass shrimp, Palaemonetes pugio Holthius and Palaemonetes vulgaris (Say).Using a zig-zag die1 pattern of temperature change like that used by Widdows (1976), they obtained the data summarized in Table V. Clearly, cycling temperatures significantly depressed the survival of both species in low salinity (5%,) water. As a result of this study Thorp and Hoss stressed that more work was necessary to detect possible positive or negative effects of cyclic temperature regimes, particularly as they might have a bearing on environmental toxicity standards which have been based upon LC50 values obtained at constant, often optimal, temperatures. The investigation of

ENVIRONMENTAL SIMULATION EXPERIMENTS

183

TABLE V. MEAN PERCENTAGE SURVIVAL OF Palaemonetes pugio AND P . vulgaris HEM) AI' THREE SALINITIES AND THREE TEMPERATURE REGIMES FOR 21 DAYS (FROM THE RESULTS OF THORP AND Hoss, 1975) Salinity Temperature regime 1 . Cyclic 7-13°C P. pugio P . vulgaris 2. Constant 10°C P. pugio P. vulgaris 3. Constant 7°C P . pugio P . vulaaris

5% 20% 35%

467 9 3 3 6 2 2 17.8 75.6 66.7

800 93.3 77% 64.5 84.5 75.6 75.6 91.1 95.6 467 91.1 8 0 0

Christiansen and Costlow (1975)lent weight to this advice since they found that survival of the larvae of the mud crab Rhithrqanopeus harrisii at low salinity (5%,), in contrast to that of the Palaemonetes adults, was better in a fluctuating temperature regime than had previously been demonstrated in an equivalent constant temperature study (Costlow et al., 1966). I n much of their subsequent work, Costlow's group have pursued the matter of the interaction of pollutants with cyclic thermal regimes. Rosenberg and Costlow (1976)studied the synergistic effects of cadmium and salinity combined with constant or cycling temperatures on larvae of R. harrisii and the eurythermal, euryhaline blue crab Callinectes sapidus. The multi-factorial design was complicated (63 different temperature/salinity/cadmium concentration combinations were used for R. harrisii alone) and this, combined with the variety of larval stages used, generated a great mass of data which are rather difficult to compress. However, Fig. 25 gives a general picture for R. harrisii. At both the lower and upper ends of the temperature scale fluctuating temperatures apparently stimulate survival in all cadmium concentrations; this is particularly noticeable in the 30 + 35°C temperature regime. On the other hand, to complete an already complicated picture, more recent studies upon the same species by Christiansen et al. (1977a, b) using combinations of temperature cycles and various concentrations of the juvenile hormone mimicking insecticides methroprene and hydroprene, failed to demonstrate any synergism with temperature.

184

J UAVESPORT

Hatch

100

-

Crab

80

-

60

$

40

:

Ln

20

0 Salinit

11

11

Temperature

x)

20-25

25

25-30

I/

l l

II

%o

I 1

30

30-35

35

OC

FIG.25. Rhithropanopeus harrisii. The percentage of animals surviving from hatching to the first crab stage when exposed to various combinations ofcadmium, salinity and constant or cycling temperatures. From Rosenberg and Costlow (1976).

VI. SALINITY STUDIES The greater part of the world’s aquatic environment consists of sea water with a salinity close to 35%,. The invertebrates in it are not under appreciable osmotic stress as their tissues and body fluids are approximately isosmotic with the external medium. It is generally held that marine invertebrates have a long marine ancestry without significant change in the composition of their seawater surroundings, except possibly over a geological time scale. For a recent discussion of the controversial question of the stability of seawater composition over geological time, Spaargaren (1978) should be consulted. Even marine fish, which have a lower ionic concentration in their blood than in the surrounding medium, are exposed to a constant rather than changeable stress. The fraction of the hydrosphere which is characterized by low, high or varying salinities is very small. Consequently, unusual or changeable salinities influence only a small proportion of the aquatic biomass. However, for reasons already discussed in the introduction, littoral, estuarine and other brackish water habitats have attracted the attention of many researchers. I n such areas the controlling influence of salinity on distribution is often obvious, and the numerous attempts to correlate particular faunal types with certain salinity levels started nearly a century ago (e.g. Mobius and Heincke, 1883) and have persisted with considerable confusion and some acrimony to recent times (e.g. Kinne, 1964b; Khlebovich, 1969).

ENVIRONMENTAL SIMULATION EXPERIMENTS

185

Distributional, tolerance and metabolic data in relation to salinity have been reviewed by Kinne (1971)for invertebrates, and for fish by Holliday (1971). A great deal of this literature has been concerned with responses of animals t o osmotic stress. The physiological mechanisms were first reviewed by Krogh (1939);more recent extensive reviews have been given by Potts and Parry (1964) and Gilles (1979). Studies containing elements of simulations of the real environment have been performed by Wells and Ledingham (1940) who changed the concentration of the medium around their tissue preparations from euryhaline polychaetes, and by Haskin ( 1964)who exposed oyster larvae to changing salinities to see if increased swimming activity was associated with rising salinity. However, the rates of salinity change employed, and the form of the changes were quite arbitrary. The first real attempt to mimic natural fluctuations in the laboratory appears to have been that of Tucker (1970a) who investigated body weight and blood composition changes on the fissurellid gastropod Scutus breviculus. Since then a number of investigations at several laboratories have been performed, which may be subdivided in the following manner. A. Survival Three studies comparing the tolerances of marine or estuarine organisms to fluctuating rather than steady salinities have been carried out by Zachary and Haven (1973),by Davenport et al. (1975) and by Cawthorne (1978). Zachary and Haven (1973) were interested in the effects of fluctuating low salinity levels on survival and activity of the Oyster Drill Urosalpinx cinerea Say. Amplitudes of salinity fluctuations were small (roughly 2-3%,) and mean salinities ranged from 7.9 to 16*8%,.Drills were also held at various steady salinities ranging from 8 to 12%,.No investigations into the effects of higher salinities were performed as earlier workers (Frederighi, 1931; Galtsoff et al., 1937; Manzi, 1970) had shown that only salinities below 20%, were lethal. Zachary and Haven's experimental design and analytical methods were somewhat imprecise and cumbersome, and it is difficult t o compare their fluctuating and constant salinity results. However, broadly speaking the drills were more tolerant of fluctuating conditions than of steady low salinities, particularly for the first 10 days of exposure. The study of Davenport et al. (1975) was carried out on larvae of

186

J. DAVENPORT

the scallop Pecten maximus (L.).Although larvae of the sublittoral stenohaline scallop were unlikely to encounter salinity fluctuations in their natural habitat, the results proved to be interesting. Larvae were exposed for 24h either to steady salinities or various 12h wavelength sinusoidal and square wave salinity regimes fluctuating between full sea water (34%,) and various lower salinities. At the end of the 24 h period the larvae were returned to full sea water for 18 h before being assessed for mortality against full seawater controls. The characteristics of these three types of salinity regime are detailed in Table VI, and the results shown in Pig. 26. The two fluctuating salinity regimes were much less damaging than constant exposure to lowered salinity levels and these results give a strong indication that conventional steady-state salinity tolerance studies can yield incorrect assessments of the likely salinity limits to distribution for a given species. Even more interesting is the difference between the effects of sinusoidal and square wave salinity regimes. Over a 24 h period, at a given flow rate, the total amounts of salt and water supplied to experimental animals were the same whichever cycling regime was used. On the one hand, larvae in the square wave regimes were exposed t o abrupt osmotic shocks and prolonged periods at low salinity, neither of which were encountered by larvae exposed to the sinusoidal cycle. On the other, animals in the sinusoidal regime were exposed t o full seawater conditions only momentarily unlike larvae T.4~31,~ VI. A COMPARISON OF THE FACTORS I NF I , ~N CAxImw IN ( : IN THREE TYPES OF SALINITY REGIME EMPLOYED IN A LARVAL SALINITY TOLERANCE STWYUPOKPecten maximus (FROM DAVENPORT et al., 1975) Regime type (all with 50% sea water minimum concentration)

No. of osmotic shocks per 24 h Period of exposure to minimum salinity Percentage of salts delivered per 24 h by comparison with 100% S.W. control 100% S.W. = 34%.

(a) Sinusoidal

(b) Square wave

(c) Steady state

0

4

2

Negligible

12h

24 h

75%

75%

50%

ENVIRONMENTAL SIMULATION EXPERIMENTS

187

100

0 0

50 Minimum sea waterencountered

100

(yo)

FIG.26. The effects of sinusoidal,abrupt and steady state salinity profiles upon larvae of Pecten m x i m w r . Triangles indicate steady state profile results; LC50 (24 h) = 49.7f0.4yO S.W.; Y = 1 3 . 9 8 7 4 1 8 1 ~where c = yo S.W. Squares indicate abrupt (P2) profile results: minimum LC50 (24 h) = 3 5 7 f 0 4 % S.W.;Y = 14.4544265~.Circles indicate sinusoidal (Pl) profile results: minimum Lc50 (24h) = 244&02% S.W.; Y = 7.918-0.120~. From Davenport et al. (1975).

from the square wave profile which were in full seawater for 12 h in every 24. So what was the major factor in allowing enhanced survival in the animals exposed to the gentle sinusoidal salinity fluctuations; was i t the absence of osmotic shocks or the freedom from prolonged exposure to low salinities? To clarify this point Pecten larvae were exposed to the two special salinity regimes shown in Fig. 27. These represent gradual (P3) and abrupt (P4) introductions to a steady salinity (equivalent to 40% sea water). After 10 h of exposure to 40% sea water assessment for mortality commenced and was repeated every 2 h until 18h of exposure had been completed. The results obtained are displayed in Table VII. It appears that a sudden osmotic shock, characteristic of the P4 type of regime, causes damage which leads to death after several hours. I n contrast, the mortality of the larvae exposed to gradual osmotic changes (P3 regimes) rises very little after the minimum salinity is reached. It seems likely that if the change is gradual, animals which are particularly susceptible to low seawater concentrations die off rapidly leaving a residue of more tolerant individuals. It follows that for Pecten larvae, sudden osmotic

188 TABLEVII. ANASSESSMENTOFTHE RELATIVECONTRIBUTIONS'I'O MORTALITY o~Esrosr.~~~o Low SEAWATER CON('ENTI~ATIONS, A N D SUBJECTION TO ABRUPT OSMOTIC SHOCKS 15LARVAE OF Pecten maximus ( looyo S.W. = 33.5%,) (a)Results of experiments: values in relative percentage mortalities (correctedfor control results).

Hours of exposure to 40%

S.W.

Regime type

Experiment order

10

12

14

16

18

P3

1 4 2 3

18.99 19.07 7.49 15.65

23.32 21.25 1503 19.96

1673 30.39 30.38 35.54

23.33 20.68 5080 58.58

22.41 21.37 67.59 66.91

P4

(b) Results of analysis of variance: performed after angular transformation of above data.

Variation in mortality due to: (i) Differences between regimes P3 and P4 (ii) Differences between sampling times (iii) Interaction between regimes and sampling times (iv) Replication

0

Degrees of freedom

Calculated d.f. ratio

d.f. ratio for P = 0.001

1

393

21-0

4

25.2

11.3

4

22.0

11.3

10

(1)

-

I c

I 10

6h

I

I

12

14

I 16

i 18

0 Exposure (h)

FIG.27. Diagram to illustrate the salinity programmes used t o determine the relative importance of exposure t o low seawater concentration and osmotic shocks. Arrows indicate sampling times. From Davenport et al. (1975).

189

ENVIRONMESTAL SIMITLATIOX EXPRKIMEN'I'S

TAHLE VIII. MKIXASMINIMUMLETHALyo SEALZAT~W CONCEN'I'HATIOS~ (100~o S.W.=335%,) FOR NEWLY RELEASED CIHRIPKDE NAL-PLII EsI.osE1) FOR 24 H TO THI
Species Elminius modestus Balanus balanoides Balanus hameri

(b) Square wave

( c ) Steady

(a)Sinusoidal

8.9 0 3 17.6k0.3 198f03

17.3 + 0 7 23.3 & 0 7 30.8 & 1.0

163+07 23.4 f0.7 31.4 f0.2

state

shock, causing damaging swelling, is the primary cause of increased mortality shown by larvae exposed to the square wave salinity regime by comparison with animals subjected to the sinusoidal type. Cawthorne (1978) performed a very similar study on the newly released nauplii of the barnacles Elminius modestus, Balanus balanoides and Balanus hameri Ascanius. The first two species are relatively euryhaline and the nauplii are known t o occur in estuarine waters at salinities as low as 8%, (Cawthorne, 1979a);B. hameri is a sublittoral offshore form which was included for comparative purposes. The results obtained are summarized in Table VIII. Again there was clear evidence that gentle salinity fluctuations are much less damaging than salinity regimes including abrupt changes, but in this instance steady state and square wave regimes were equally lethal. A separate investigation showed that, although osmotic shocks were rather more important than prolonged exposure to low salinity in causing mortality in all three species, this bias in cause of death was much less marked than for larval Pecten maximus. Cawthorne's investigation also revealed that the difference between the salinity tolerances of larvae of E . modestus and B . balanoides was greater, with E . modestus being more euryhaline, than steady-state investigations (e.g. Barnes, 1953; Bhatnagar and Crisp, 1965) had previously indicated. Although these three investigations have been the only ones specifically designed to evaluate salinity tolerances under cyclical conditions, a certain amount of information has arisen from other cyclic salinity studies. Many common intertidal animals have proved to be capable of tolerance fluctuations between full sea water and pure fresh water during simulated tidal cycles. These species are listed in Table IX.

190

J . DAVENPORT

TABLE IX. ORGANISMS KNOWN TO TOLERATE FULLRANGE OF SALIXITIES (@34%,) REGULAR SIMULATED TIDAL SALINITY CYCLES Species

Source of information ~~

Balanus balanoides Balanus crenatus Balanus improvisus Elminius modestus Bivalves Mytilus edulis Cerastoderma edule Crassostrea gigas M y a arenaria Scrobicularia plana Anadara senilis Amphipods Marinogammarus marinus G a m m r u s duebeni Crabs Carcinus maenas Fish Blennius pholis Algae Fucus serratus Enteromorpha spp Ulva lactuca Barnacles

IN

~

~

Davenport (1976) Davenport (1976) Davenport (1976) Davenport (1976) Shumway (1977) Bettison (unpublished data) Bettison (unpublished data) Bettison (unpublished data) Bettison (unpublished data) Djangmah et al. (1979) Ritz (1980) Ritz (1980) Davenport (unpublished data) Davenport and Vahl (1980) Dickson (1978) Dickson (unpublished data) Dickson (unpublished data)

B. Behavioural responses Reduction of tissue contact in response to adverse external salinities is common in sessile or slow moving benthic intertidal or estuarine organisms (Kinne, 1971). Cronklin and Krogh (1938) and Milne (1940) showed that the mussel Mytilus edulis, reduced contact by closing the shell valves. Since then similar mechanisms have been demonstrated for a variety of invertebrates including other bivalves (e.g. Freeman and Rigler, 1957), gastropods (e.g. Segal and Dehnel, 1962; Avens and Sleigh, 1965) and barnacles (e.g. Barnes and Barnes, 1958; Foster, 1970). Several mobile nektonic invertebrates and fish have been shown to detect deleterious salinities and to be able to select more favourable or even optimal conditions (e.g. Bull, 1938; Gross, 1957; Baggerman, 1960; Jansson, 1962; McLusky, 1970; Davenport, 1972a;Bettison and Davenport, 1976).However, there is insufficient information on the form of short term salinity changes encountered by such animals in nature because their behaviour may allow them to avoid environmental fluctuations partially or completely. In consequence, laboratory simulation studies have largely been restricted to studies of planktonic animals or of sessile or slow moving benthic organisms.

ENVIRONMENTAL SIMULATION EXPERIMENTS

191

Studies on planktonic animals have been mainly concerned with oyster larvae (Crassostrea spp.) living in estuaries. There has long been controversy about the means by which the larvae of estuarine animals maintain their upstream position in estuaries prior to settlement. Three stances have been taken by various researchers. First, some authors believe that oyster larvae are simply transported by currents, and exhibit no differential vertical position with tidal stage (Loosanoff, 1949; Andrew, 1954; Korringa, 1952). Second, Pritchard (1953) stated that in a two layered estuarine situation, oyster larvae would only have to maintain a benthic position to stay in high salinity water and move upstream. Finally, there are those workers who have suggested that oyster larvae can detect increased salinity levels and become more active in response to them, thus leaving the bottom and being swept upstream on the flood tide (Nelson, 1912; Haskin, 1964; Wood and Hargis, 1971). The work of Hidu and Haskin (1978) appears to be the latest in this sequence. These investigators observed larvae swimming in a “salinity cell” (Haskin, 1964) supplied with water rising in salinity a t about 0*5%,/h (somewhat slow in most real estuarine situations). The results obtained were rather confusing, particularly as considerable differences in swimming speed between small and large larvae were demonstrated. Reference to the study is included because Hidu and Haskin suggest that “the trials should be extended to large scale experimental water columns which could simulate estuarine water column conditions and in which in addition to temperature and salinity, pressure and quantitative aspects of light could be controlled”-an ambitious and expensive aim which would require more information about short term changes in estuarine conditions than is available at present! Cawthorne and Davenport (1980) working on the larvae of the barnacles Elminius modestus and Balanus balanoides approached the problem of the vertical position of estuarine planktonic organisms from a different viewpoint. Unlike oyster larvae, newly released barnacle nauplii need to be carried downstream towards the sea to survive. Also they cannot isolate themselves fram low salinities by closing up and descending to the bottom, as may bivalve larvae. It seemed likely that nauplii might react to falling salinities by ceasing to swim and subsequently sinking into higher salinity water down in the water column. Once in high salinity water their survival would be assured as they would tend to remain within it. For the settlement cyprid stage in estuaries the situation is somewhat different. Cyprids can isolate themselves from deleterious salinities by closure of the

192

J . DAVESPOHT

carapace valves and would benefit by sinking in low salinity water so that they were not carried too far upstream before settling. Cawthorne and Davenport, using linear falling salinity regimes of realistic rates of salinity fall derived from the field study of Cawthorne (1979a), obtained data which support these hypotheses (see Table X). Larvae of the generally more euryhaline E. modestus kept swimming in rather lower salinities than those of B . balanoides. This difference in response is less marked between the cyprids of the two species. Interestingly, the salinities which prevent swimming and result in the nauplius sinking are quite close to the lethal limits of both species as predicted from the sinusoidal salinity cycle studies of Cawthorne (1978); cessation of movement occurs a t salinities much lower than those causing mortality in steady-state studies. Investigations into the behaviour of sessile organisms exposed to simulations of estuarine salinity regimes started with a study of the responses of barnacles by Davenport (1976). A novel method of continuously recording activity in barnacles, derived from the impedance pneumograph technique of Trueman (1967),was used (see Fig. 28) to monitor opercular valve and cirral movements in barnacles exposed to sinusoidal, near tidal salinity regimes fluctuating between full sea water and 20% sea water (33.4%, and 6.7%,). The proportion of time that the animals were active in each hour was noted, together with the limiting seawater concentrations required to induce the cessation of activity or permit its recovery. Such inactivity in barnacles at low salinities has been described TABLEx. RESPONSES OF BARSAVLE LARVAE TO FALLIN; SALISITIEh ( MOIHPIED FROM CAWTHOKNE A N D DA V E N P OI 1980). ~ T, MEANSALIKITIES (WITH 95% CONFIDENCE IST~CRVALS) COKRESPOSDINU TO CESSATION OF SWIMMING IN 50% OP LARVAE, AT RATES OF SALINITY FALL ~h SHOWN

A. Elminius modestus Stage Nauplii (stage I ) Rate of change

(Wh) Critical salinity

16.75

4.81

Cyprids 16.75

.81

8.8 f0.9 6.6 f0.3 9.0 k 1.3 8.3 f0.9

B. Balanus balanoides Stage Rate of change (YJh) Critical salinity

Nauplii (stage I ) 16.75 12.3f1.0

4.81

Cyprids 16.75

9 3 k 1 . 0 109+1.0

4.8 1

9.1k1.1

ENVIRONMENTAL SIMlTLATION EXPEKIMESW

Shielded cable

B

Minutes +it--r-, r

193

-$&a,-.-*. I

1

+t

Ealanus rrnprovrsus-Conwy

h-h

Ealanur improvisur-Baltic

Balanus crenatus

-W-YcY-IC-jJw(

Elminius rnodestur

Ealanur balanus

1~~~~~~~~~~~~~~~~~~~ / j -w

Eaianus harnerr

Balanus balanoides

F I G . 28. ( A ) Arrangement for activity recording. (B) Examples of activity records. Two sets of records for B. improuisus are displayed to demonstrate that the recording technique is equally applicable to small Baltic specimens (2-5 mm basal diameter) and large Conwy animals (15-20 mm basal diameter). From Davenport (1976).

somewhat confusingly as “salt sleep” by Barnes and Barnes (1958). Examples of activity patterns are shown in Fig. 29; critical seawater concentrations for several littoral/estuarine balanomorphs appear in Table XI. From this table it may be seen that the responses of Balanus crenatus Bruguikre, Balanus balanoides and Elminius modestus to salinity fluctuations are very similar, despite the lower shore distribution of B. crenatus. This last species appears to be barred from higher placement on the shore because of its inablity to close its opercular valves in response to aerial emersion; certainly it is not because of sensitivity to low salinity. It is also noteworthy that the responses are remarkably symmetrical, with activity stopping in falling salinities at about 20% and being resumed at much the same seawater concentration when the external salinity rises again. Symmetry of such responses to salinity appears to characterize sessile animals which are well adapted to pronounced salinity fluctuations. Thus, intertidal barnacles (Davenport, 1976), bivalves (Shumway, 1977a; Shumway and Youngson, 1979; Djangmah et al.,

100

100

L

W

0

c

c ?

0

50

50

E

:: v)

8

c

8 0

00 100

L

W

0

50

8 00 0

12 Hours

24

FIG.29. Activity in Conwy estuary barnacles. (a)Balanus crenalus, (b) Elminius modestus, (c) Balanus improvisus, (d) Balanus improvisus in more extreme salinity regime. Dotted line =presents salinity regime; other symbols represent mean values with 95% confidence intervals. From Davenport (1976).

ENVIRONMENTAL SIMULATION EXPERIMEBTS

195

TABLE X I . CRITICALSALINITY LEVELS FOR INTERTIDAL BALANOMORPH BARNACLES EXPOSED TO A SINIJSOIDAL SALINITY REGIME FLUCTUATING BETWEEN 3340,b A N D 6.7% (SIMPLIFIED AND MODIFIED FROM DATAOF DAVENPOHT, 1976)

Mean salinities ( X ) (b) Inducing activity cessation onset

(a)Inducing activity

Species A From Menai Strait Balanus crenatus Balanus balanoides Elminius modestus B From Conwy estuary Balanus crenatus Elminius modestus Balanus improvisus

23.3 23.6 22.0

21.7 23.3 19.5

191 204

19.4 24.4 No Activity Cessation

1979) and the lugworm Arenicola marina (see Shumway and Davenport, 1977) all show symmetrical activity responses (see Table XII). On the other hand, sublittoral or poorly adapted lower shore forms, such as offshore barnacles (Davenport, 1976), scallops (Bettison, unpublished data), sea anemones and sea squirts (Shumway, 1978a, b), and hermit crabs (Davenport et al., 1980) all exhibit asymmetrical responses. I n these activity ceases at salinities similar to, or even lower than, those inducing inactivity in better adapted forms; activity is not not resumed until much higher salinities have been attained. The source of the asymmetry appears to vary from species to species. I n specimens of the hermit crab Pagurus bernhardus (L.), which do not resume activity after exposure to low salinities until they have been in full sea water for some time (Davenport et al., 1980),the probable reason is that their sense organs are isolated from the environment when the crabs retreat into their shells in response to lowered salinity (Shumway, 1 9 7 8 ~and ) register a rise in external salinity only after enough salt has diffused into the retained water within the shell (Davenport et al., 1980). On the other hand, forms such as the deep water balanomorph Balanus hameri appear to be rendered comatose by exposure to low salinities 2nd are then unable to resume activity until a period spent in high salinity water has revived them (Davenport, 1976). From Table X I and Fig. 29 it may also be seen t h a t the salinity reactions of the barnacle Balanus improvisus Darwin are different from those of other intertidal barnacles. In a regime fluctuating between full sea water and 20% sea water specimens do not cease

196

J. DAVENPORT

activity. Even when exposed to fluctuations between full sea water and pure fresh water they only close their opercular plates and stop cirral activity at salinities below S%, (see Table XII). This result agrees quite well with the laboratory experimental results of Foster (1970). However, Foster also suggested that estuarine (Tamar) B. improvisus closed its opercular plates at about la%,in nature. This suggestion arose from blood osmolarity readings taken from freshly collected animals. He presented no direct evidence to support this, and the recent study of Cawthorne (1979b) has since confirmed that B. improvisus does not become inactive at salinities above 10%. Fyhn (1976) presented data which suggests that the species, unlike other balanomorphs, is capable of osmoregulation. This would help to explain the unusual degree of euryhalinity which allows B. improvisus to penetrate far into the Baltic Sea where other barnacles cannot survive. TABLE XII. ACOMPARISON OFTHE RESPONSESTO SALINITY OFBalanuScrenatus FROM DIFFERENT SOURCES EXPOSED FOR 24 H TO SINUSOIDAL SALINITY REQIMES (12 H WAVELENGTH) FLUCTUATING BETWEEN 33.4 AND 6.7ym(CALCTJLATEDFROM DATA OF DAVENPORT, 1976)

Source 1 . Conwy

estuary 2. Conwy estuary but held for 1 month in full S.W. (33-4%0) 3. Menai Strait

1st activity cessation salinity

1st activity onset salinity

2nd activity 2nd activity cessation onset salinity salinity

19.6k 1.3

191 f 3 . 5

18%+ 1.4

19.6k3.6

212f2.1

21.2k 1.8

20.6+ 1.8

2 3 2 k 1.0

244f2-6

21.05 1-7

2 2 2 A 1.5

22.4k 1.3

Values are mean salinities with 95% confidence intervals.

Davenport (1976) also showed that the salinity responses of barnacles were susceptible to alteration by non-genetic adaptation (i.e. acclimatization-Kinne, 1964a).This phenomenon is illustrated in Table XI11 for B. crenatus. From these data it may be seen that the Conwy estuary barnacles cease and start activity at lower salinities than B . crenatus from the Menai Strait, where barnacles are normally exposed to full seawater conditions. However, Conwy B. crenatus held for 1 month in full sea water could not be distinguished in their responses from Menai Strait specimens, so had clearly adapted.

197

ENVIRONMENTAL SIMULATION EXPERIMESTS

TAHLE XIII. CKITI~AI, SALINITIES FOR BI.:HAVIOURAL RESPONSEX OF WELLh A P 1 ' R i ) INTERTIDAL ORGANISM^ (TO NEAREST ym)IN SIML-LATEI)TIDAL CY('LES Species Barnacles Balanus crenatus Balanus balanoides Elminius modestus Balanus improvisus Bivalves Mytilus edulis Crassostrea gigas Modiolus demissus

Activity cessation salinity

Activity onset salinity

23 24 22 8

22 23 19 8

Davenport (1976)

14 20 21

12 13 20

Rhuinway (1977a)

LSource of data

Shumway and Youngson

Anadara senilis Polychaetes Arenicola marina"

15

15

(1979) Djangmah rt al. (1979)

19

19

Shumway and Davenport

(1978) "These values are for water overlying a substrate characterized by high interstitial salinities in which the lugworms were burrowed.

Much of the interest in the study of Davenport (1976) lay in comparisons with previous steady-state work on intertidal barnacles by Foster (1970). Foster made most of his observations after barnacles had been held in media of constant concentration for 24 h. For Elminius modestus he reported that activity was present down to 50% sea water (i.e. 17%,) in Menai Strait specimens, and 42% sea water (14%,) in Conwy animals. Apart from the fact that his technique gave no indication of salinities corresponding to activity resumption in rising salinities, these concentrations are significantly lower than the values reported by Davenport (1976) (see Table XI). It seems likely that Foster's animals were active in lower concentrations partly because they may have had longer to adaptthough the results of Cawthorne (197913) largely deny this possibility-but mainly because respiratory needs tended to override the closure response to lowered salinity when such seawater concentrations were held for as long as 24 h. Foster also found that Elminius died after 30 h continuous immersion in 25% sea water (8*5%,),but in both laboratory simulation experiments (Davenport, 1976) and field studies (Cawthorne, 1979a), it has been established

198

J . DAVENPORT

that the species can and does survive regular exposure to much lower salinities. Finally, Davenport (1976) collected some data which suggested that the salinities including closure in intertidal barnacles might be affected by rate of change in external salinity prior to closure. Cawthorne (1979b) used linear falling salinity regimes of various rates of change to clarify this point. He established that in B. crenatus, B. improvisus and E . modestus the seawater concentration corresponding to shell closure was independent of the rate of fall in salinity except at unphysiologically high rates of salinity change. In contrast B. balanoides did show a sensitivity to rate of salinity change. Animals from the Menai Strait exposed to a 48 h duration drop from 33.5 + O X , ceased activity at 24*9%,;with increasing rate of salinity change this value rose to a peak of 27.6%, for a salinity drop over 7.5 min. Corresponding figures for Conwy specimens were 23.6 and 26-3%, respectively. Interestingly, Cawthorne ( 1979a) showed that retained mantle fluid salinities in B. balanoides adults exposed by the falling tide in the Conwy estuary were about 26%,-again indicating that simulation studies give a realistic picture of responses in habitats characterized by short term fluctuations in physicochemical conditions. Comparable data concerning the salinities which induce shell valve closure in bivalve molluscs have been collected for several species (see Table XII), but the most detailed studies have been carried out on the common mussel, Mytilus edulis, a prominent member of the fauna of estuaries. Milne (1940),working on mussels from the Aberdeenshire Dee, showed that the salinity of the water retained within the mantle cavity of mussels exposed to pronounced tidal fluctuations in salinity never fell below 21%. He also noted that the retained mantle fluid salinity was rather higher in areas characterized by severe salinity changes than in places where the changes were less extreme. Similarly, Shumway (1977a) found that the mantle fluid osmolarity of mussels which had closed in response to an abrupt salinity profile fluctuating between 33*5%,and O X , was significantly higher than the corresponding value found for mussels exposed to a regime which changed sinusoidally between 33.~5%~ and fresh water. Together, these observations suggested that the adduction response of Mytilus to declining salinities might be at least partially dependent upon rate of change of salinity, as it is in the barnacle B. balanoides. Davenport (1979a) found that this was indeed the case, but, as is evident from Fig. 30, there was no simple relationship

199

ENVIRONMENTAL SIMULATION EXPERIMENTS

A

1

0 1 0.1

I

I

1

I

I

I

1

1.o

10

100

lo00

Rate of fall in medium salinity (%o/h)

30. Mytilus edulis. (a)The effect of rate of change in external salinity, prior to shell valve closure, on the mantle fluid salinity maintained after closure. (b) The effect of rate of changes in salinity on the external salinity inducing complete shell valve adduction. Symbols represent mean values with 95% confidence intervals (for six animals). From Davenport (1979a).

FIG.

between external salinity at the time of shell valve closure and the salinity of the mantle fluid found inside after closure. Although the salinity corresponding to valve closure fell with increasing rate of external salinity change, retained mantle fluid salinities actually rose! This led to the discovery that the isolation of the mantle cavity of a mussel from the external environment is not produced simply by shell valve adduction, but results from a three-part sequence of reactions. First, it became clear from the data presented in Table XIV that the salinity of the retained mantle fluid is primarily determined by closure of the exhalant siphon; once this siphon is closed, effective irrigation of the mantle cavity ceases. A t this time lateral and frontal ciliary activity upon ctenidia stops or slows

200

d . DAVENPORT

Taur,~ XIV. SAIJSITY LEVKLS Assocumu RTI'H THE COMPONKNXS OY T H E Isoixrrox RESPONSE: OF Mytilus edulia Exros~i) TO FAI.I.ING SAI.INITIW (FIWM DAVENPOKT, 1979a)

A. Summary

(x0)

Salinity External salinity regime (time (mean values with 95% conjdence intervals) (min)taken A t exhalant A t inhalant At complete f r o m 33.5 to siphon siphon shell valve O%,, closure closure closure 3.75 7.5 15 30 60

25.3 & 2.2 25.6 0 7 25.9 t0.9 28.4 k 1.7 24.8 k 1.3

+

21.8 f3.4 2 0 4 & 3.3 19.3 f2.5 24.9 f3.3 23%f 1.9

10.7 2.9 17.5 f3.7 17.4f1.5 18.8 3% 20.2 f3.7

Salinity ?Ao) o j mantle fluid after complete closure (mean values with 95% conjdeme intervals) 3 0 6 k 1.2 29.2 f1.2 28.9 & 1.2 31.0f09 29.7 f 1.6

B. Example (335%, -+ Oo/, in 3 7 5 min) A t exhalant siphon Animal no. closure 1

2 3 4 5 6

26.0 23.5 22.0 27.5 265 260

External salinity A t inhalant siphon closure 25.5 19.5 19.0 23.0 18.5 25.5

(X) At complete shell valve closure

Salinity (%,) of mantle Puid after complete closure

8.5 140 7.0 12.5 12.5 9.5

31.5 295 290 31.5 30.5 31.5

(Sleigh, 1962;Ajana, 1975;Davenport and Fletcher, 1978).When the external salinity has dropped further the inhalant siphon shuts; later still the shell valves close. Once the valves are firmly adducted, the mantle fluid salinity remains unchanged for many hours (see Fig. 31). Of course there had to be some advantage gained by Mytilus from this behaviour pattern which is quite prolonged under conditions of slowly changing salinity and therefore contrasts strongly with the immediate total closure of Mytilus to tactile stimuli. Since M . edulis can close its valves in seconds when touched, why does it close in piecemeal fashion over a period of many minutes when exposed to slowly falling salinities? The answer to this question is evident from the results presented in Fig. 32 and Table XV. Mantle fluid oxygen tensions were monitored while mussels were exposed either to abrupt drops in salinity (regime A) or to slowly

ENVIRONMENTAL SIMULATION EXPERIMENrS

20

1, 0

I

8 Hours after closure

20 1

1 16

FIG.31.The salinity of the retained mantle cavity Auid of mussels (Mytilusedutis) exposed to a sudden fall in external salinity (335%, + O X , in 7.5 min) to a sustained freshwater level. Symbols represent mean values with 95% confidence intervals (for six animals). From Davenport (1979a).

falling salinity (regime B). This was done by slow circulation of mantle cavity fluid through a low volume external circuit which included an oxygen electrode. I n the abrupt salinity regime closure of siphons and valves occurred almost simultaneously, and the mantle fluid oxygen tension fell rapidly as the animals tissues took up oxygen. I n the slowly falling salinity regimes the tripartite sequence of closure was prolonged; the mantle fluid oxygen tension fell more slowly and irregularly. It must be remembered that although sessile estuarine osmoconformers benefit osmotically by isolation from the environment when external salinities are low, they also incur penalties. Food may not be collected, nor gases or metabolites exchanged during the periods when isolation is effectively complete. I n Mytilus once the exhalant siphon is closed, the animal may continue to gape for periods determined by the rate of external salinity change. During this period of gaping the mantle fluid salinity does not fall much (perhaps 1 -4%,) because of poor exchange with the exterior, but its oxygen tension takes much longer to fall than in animals which close suddenly. Several factors are probably responsible for the slower rate of oxygen depletion in partially isolated, gaping mussels. First, the mantle cavity of a gaping mussel is physically larger than that of a tightly closed one and hence contains more oxygen; this factor will operate until final shell valve closure. Second, diffusion of oxygen may take place through the inhalant siphon if open or across the exposed mantle edge if closed.

202

J. DAVENPORT EIV

E

IV

150

E

I V

I

I 1

50

E

l

V

E

EIV

IV

-Y

El

EIV

V

-+it

A

--

0

l

o

Time (h)

IV

j:'i B

A

E

1

B

7 0

1 0 Time (h)

1

FIG.32. Mantle fluid oxygen tensions in six mussels exposed to sudden (a) and slow (b) falls in external salinity (bottom diagram). E corresponds to exhalant siphon closure, I to inhalant siphon closure and V to complete shell valve adduction. Drawings between traces show the size and shape of each of the six mussels studied (scalebar = 5 cm) and the positioning of the hypodermic needles used in the oxygen tension measurement circuit. The values superimposed a t the end of each oxygen tension trace correspond to the salinities of retained water samples taken a t the end of each experiment. From Davenport (1979a).

Finally there may be small bulk movements of water between the interior and exterior because of small movements of mantle, valves, or foot. Supporting evidence for this last suggested factor lies in the irregular form of the oxygen traces derived from animals in the slow (B) salinity regime and displayed in Fig. 32. Davenport suggested that, at the Conwy estuary site studied by Cawthorne (1979a),this behaviour pattern would gain mussels living there an extra hour of available oxygen each day-a worthwhile bonus for animals which

ENVIRONMENTAL SIMULATION EXPERIMENTS

203

TABLE xv. THEEFFWTOF RATEOF FAIL I N EXTERNAL SALINITY U P O N THE DHIINE OF MANTLEFI,UID OXYGEN TENSION (PROM DAVENPORT, 1979a) Time interval (man)between closure of the exhalant siphon and mantle fiuid oxygen tension falling to 15 mmHg , , ,’’

Animal no. 7.5 min (33.54%,, S)

1 2 3 4 5 6 Mean

S.D.

29.5 19 18 255 32 24.5 24.75 5.56

2 h (33&00/,S) 45 49 33 47 69 47 48.33 11.64

are deprived of an external oxygen source for up to 10 h per day because of aerial exposure. There are other implications from this study. Several physiologists have relied upon data from “propped open” bivalves, in which the shell valves are held apart by glass tubes or plastic wedges (e.g. Pierce, 1971; Shumway, 1977a; Costa and Pritchard, 1978) implying that “propped open” animals are in full tissue contact with the environment. Costa and Pritchard (1978) even postulated that the mussel was a short term osmoregulator on the basis of such studies, despite the extensive earlier work which had shown Mytilus and other marine bivalves to be osmoconformers (e.g. Lange, 1963,1970; Lange and Mostad, 1967; Pierce, 1970, 1971; Pierce and Greenberg, 1972,1973).My results showed that a mussel could largely isolate the mantle space simply by closing the exhalant siphon without shell valve adduction. Hence the tissues of “propped open” bivalves are not in intimate contact with the external medium. This indication was confirmed in full in a further study (Davenport, 1979b) which demonstrated t h a t M . edulis was an osmoconformer even in the short term. Thirty years ago Wells (1949b) postulated that parts of the complex burrowing behaviour of the lugworm Arenicola marina prevented the irrigation of the worm’s burrow with noxious water. Arenicola is moderately euryhaline, but cannot survive in water of less than loo&,constant salinity, and even that concentration only applies to the Baltic Sea populations; it is the classical example of an

204

J. DAVENPORT

osmoconforming polychaete (Schlieper, 1929). I n estuaries the lugworm may be found on mud flats which are washed with virtually pure fresh water at some stages of the tide although the interstitial salinity almost certainly remains high and varies little (Kinne, 1971 ). Shumway and Davenport ( 1977) used simulation equipment to confirm Wells’ hypothesis for salinity responses. Lugworms were placed in seawater-laden sand in a vessel of the type shown in Fig. 33. Sea water flowed over the top of the sand from the variable salinity apparatus and the worm’s activity was monitored by a sensitive pressure transducer connected to a fine capillary tube thrust deep into the head shaft of the Arenicola burrow. When the animal had settled, sinusoidal or square wave salinity fluctuations between 32 and 9*6%,were programmed for the water overlying the burrow. In both types of regime the worms responded to low salinity by becoming quiescent at the bottom of the burrow until the salinity rose again. The promptness of the salinity response to abrupt decreases in salinity is evident from the traces shown in Fig. 34. T

FIc:.33. Sand-filled box with A . marina in burrow and glass capillary, T, positioned in anterior end of burrow. Arrows indicate direction of water flow. From Shumway and Davenport (1977).

ENVIRONMENTAL SIMULATION EXPERIMENTS

205

Results from worms exposed to sinusoidal regimes established that a salinity of about 20%, both induced inactivity in falling salinity regimes and triggered a resumption of irrigation in rising salinities. Periodically lugworms would “test” the water above the substrate by briefly pumping water (see Fig. 35).However, they quickly became inactive again if the salinity was still low. Presumably low salinity water drawn into the burrow during “testing” rapidly equilibrates with the higher concentration of the interstitial water of the substrate. As will be discussed in more detail in the next section, the control exerted over body fluid osmoconcentration as a result of this behaviour is quite remarkable and comparable with that attained by the closing responses of intertidal barnacles and bivalves. A few studies have been performed upon animals which are less likely to encounter regular salinity fluctuations. In three separate studies Shumway (1978a, b, c) studied specimens of the coelenterate Metridium senile (L.), the ascidian Ciona intestinalis L. and the hermit crab Pagurus bernhardus exposed to simulated tidal salinity cycles. For M . senile and C . intestinalis fluctuations between lOOyo and 30% sea water were used although some specimens of the former species did not survive this; P . bernhardus experiments were performed with regimes which did not drop below 50% sea water; more extreme regimes were lethal. It has to be pointed out that only one of these three species, C. intestinalis, is known to extend into estuarine areas (MacGinitie and MacGinitie, 1968).The other two are littoral rather than estuarine, and the anemone M . senile is confined to the lowest intertidal levels and below. It is difficult to envisage its ever being exposed to salinities as low as the lo%, minima employed by Shumway, and certainly not for 6 h at a time. Large P . bernhardus are found only in the sublittoral, while the littoral population consists of small crabs usually living in rockpools. Davenport (197213,c, d) showed that the species were moderately euryhaline, that the small intertidal specimens, unlike their large offshore relatives, were capable of volume regulation and that the soft abdomen of the hermit crabs played a part in resisting the effects of low salinity. However, a recent field and laboratory investigation by Davenport et al. (1980) suggests that the species rarely encounters salinities below 25%, and then only briefly; Davenport’s earlier studies and the investigation of Shumway (197th) almost certainly exposed the hermit crabs to far lower salinities than they ever meet in nature. The reason for this is simple; Shumway ( 1 9 7 8 ~stated ) that tide pools are subject to dilution by rain and terrestrial run off. While this is true of very shallow pools, perhaps up to 5 c m in depth, and

206

J. DAVENPORT 100 % sea water A

\\ \

\

B U

1 min

30% sea water

FIG.34. Changes in A . marina activity when exposed t o a 30% seawater minimum abrupt salinity change, and the animal’s position in the U-tube after activity had e a s e d . From Shumway and Davenport (1977).

30 % sea water

FIG.35. (a) Trace recorded during sampling excursion of A . m a ~ i n aexposed to decreased salinity and (b) the position of the worm in the glass U-tube a t the time of sampling. From Shumway and Davenport (1977).

also of the surface layers of deeper ones, the bottom water of most littoral pools of the type inhabited by hermit crabs remains high and stable (Pyefinch, 1943; Ganning, 1971; Daniel and Boyden, 1975; Davenport et al., 1980). Given that there are some reservations about the severity of the regimes used by Shumway, her studies do extend our knowledge of the convergent evolution of the behavioural responses of benthic osmoconformers to salinity. Metridium reduced tissue contact with

207

ENVIRONMENTAL SIMULATION EXPERIMENTS

the external medium in low salinities by expelling water from the coelenteron, retracting the tentacles and finally contracting the whole body to become as small as possible (see Fig. 36). Ciona ceased its regular squirting activity and tightly closed its siphons to prevent entry of water for filtration whenever the external salinity fell to about 20%,; this left only the relatively small surface area of the tough, thick and leathery and probably impermeable outer skin of the sea squirt in contact with the external medium. Hermit crabs retreated into their shells when the seawater concentration fell by about 25%; Davenport et al. (1980) used linear salinity regimes to pinpoint the critical concentration for Pagurus at 20*5-22.5%,,a value unaffected by rate of salinity change prior to retreating. All three species showed pronounced asymmetry of response indicating their relatively poor level of adaptation to estuarine conditions.

1

0

6

Hour.

I I2

PIC:.36. The effect of fluctuating salinity on activity in M . senile: all animals are drawn to the same scale. From Shumway (1978a).

C. Reproduction The only work so far concerned with the effects of simulated tidal salinity regimes on reproductive processes is that by Cawthorne and Davenport (1980),which consisted of a study of larval release in the intertidal barnacles Balanus balanoides and Elminius modestus. Unlike cyclic thermal regimes which revealed differences between the two species (see Section VC), the salinity fluctuations elicited similar responses from both species. As might be expected &hebulk of larval release occurred at high salinities (e.g. Fig. 37) in the range of the adults’ activity. At low salinity, as during aerial emersion, the adult barnacles close their opercular plates. Consequently larvae which hatch within the adult mantle cavity during the period of isolation cannot be liberated until external conditions become favourable again and cirral activity is resumed. The results obtained from animals exposed to both square wave and sinusoidal regimes

208

J . I)AVESPOHT

PIC:.37. Larval release from adult specimens of Balanus balanoides exposed to a sinusoidal salinity regime fluctuating between full sea water (hour 0,12 and 24) and fresh water (hour 6 and 18).From Cawthorne and Davenport (1980).Symbols represent mean values with 959” confidence intervals. Dashed line indicates salinity profile.

indicated that a small number of larvae are released at salinities below those associated with activity in the adults. It has been reported (e.g. Davenport, 1976; Cawthorne, 1978) that the mantle cavity is not totally isolated under these conditions. Pneumostonie formation (i.e. parting of the soft lips of the opercular flaps to form a small hole) and “testing” (Crisp and Southward, 1961) offer opportunities for larval release in these species a t low salinities.

D. Growth It seems likely that studies of growth during exposure to salinity cycles in sessile invertebrates such as bivalves and barnacles would be fruitful since the extremes of distribution of such organisms in estuaries are probably determined not simply by salinity tolerance. Growth rates may be equally important and will be determined by the proportion of time that the animals are able to collect food and exchange gases or metabolites. At the time of writing no published information is available although such studies are in progress on Mytilus (Gruffydd, personal communication) with the aid of the laser diffraction growth measurement technique of Stromgren ( 1975) which allows daily growth measurements to be performed on young mussels.

ENVIRONMENTAL SIMULATION EXPERIMEKTS

209

E. Feeding Again little information is available at present although the papers of Davenport and Vahl (1979) and of Vahl and Davenport (1979) explore the impact of salinity fluctuations upon the feeding strategy of the intertidal teleost Blennius pholis L.; this work will be discussed in more detail in a later section. Also Bettison (in preparation) has shown that filtration rates in bivalves such as Mytilus and Crassostrea gigas Thunberg are enhanced after periods of shell valve closure at low salinity in simulated estuarine salinity cycles. Matthiessen (1960) had previously shown that feeding rate decreased with decreasing salinity in the bivalve M y a arenaria L. Obviously changing external and internal ionic composition in animals, particularly osmoconformers, exposed to fluctuating salinities are likely to affect physiological processes such as nerve conduction, ciliary activity and circulatory efficiency, which will themselves in turn influence feeding.

F, Osmoticlionic responses 1, Extracellular jluid composition

Kinne (1964b) suggested that investigations into the effects of salinity as an ecological factor should include studies of the consequences of fluctuating patterns of salinity. Six years later Tucker (1970a) carried out the first laboratory study of tidal salinity fluctuation on the prosobranch gastropod Scutus breviculus. This species is normally marine but penetrates the Heathcote- Avon estuary in New Zealand to a limited extent. Tucker’s preliminary field salinity measurements showed that the environmental salinity of S. breviculus fluctuates only between 34 and 29%, and values below 34% persisted only for about 1h during each tidal cycle. Her laboratory simulations reflected these conditions and consisted of a sequence of 1h at 34%,, 30 min falling to 29%,, 30 min at 29%,, 30 min rising to 34%,, followed by 2 h at 34%, (see Fig. 38). Tucker was interested in the osmotic and ionic responses of the gastropod. First she showed that S. breviculus was an osmoconformer, becoming isosmotic over the salinity range of 25.5-34%, in about 8 h. Using her estuarine simulation she showed that the haemolymph concentration changes were damped by comparison with those of the external medium (see Fig. 38). Thus, whereas the salinity of the external

210

J. DAVENPORT

e

2

1

-

84 -

:

:---;Sea I

l

water concentration l

1

1

1

I

I

I

I

I

medium fell by 15% during the simulated salinity cycle, the prosobranch's haemolymph concentration fell only by about 8%. Such damping was first remarked upon in polychaete worms by Wells and Ledingham (1940). Also there was a time lag or retardation (hysteresis) of haemolymph osmotic changes compared with the medium concentration changes. Damping and retardation are presumably caused because osmotic uptake of water and outward diffusion of salt through the integument and other membranes takes time. From dimensional considerations damping and hysteresis will tend to be more marked in larger animals. Since Tucker's study an excellent mathematical treatment of the damping and time lag effects of tidal salinity cycles has been written by Spaargaren (1974). Obviously damping and time lags are responsible in large measure for the enhanced survival of many animals (except perhaps small larval forms) exposed to fluctuating rather than steady salinities; if the body fluids do not have time to equilibrate fully with the extremes of external salinity fluctuations, then the animal concerned will survive conditions which would be lethal if sustained. Tucker also measured the haemolymph concentrations of various cations; her results are displayed in Fig. 39. From previous work it would seem likely that hydrated sodium and potassium ions would pass through membranes more readily than larger hydrated calcium and magnesium ions (e.g. Dakin and Edmonds, 1931; Webb, 1940; Conway, 1956, 1960). Koizumi (1935) showed that rates of cationic penetration through the skin of a holothurian Caudina chilensis (Gould) could be arranged in the following expected sequence: K+>Na+>Ca2+>Mg2+.

21 1

ENVIRONMENTAL SIMIJLATION EXPERIMENTS

Sodium Potassium

x-x

----

I

0

I

I

I

I

I

2

I

I

I

3 Time, hours

I

4

I

I

5

I

PIC:.39. Changes in the concentrations of cations in the blood of animals subjected to changing seawater concentration simulating environmental conditions. (Each curve mean of values for 22 animals). From Tucker (1970a).

From the results displayed in Fig. 39 and from some additional direct transfer experiments Tucker established that the sequence for Scutus was as follows: Ca2+>K > Na+ > Mg2+. +

In other words, the rates of movement of potassium, sodium and magnesium were as expected if their rate of permeation through the integument was determined by the size of the hydrated ion. However the loss rate of calcium in lowered salinities was much greater than expected. Tucker could only offer various speculations to explain this discrepancy. Enhanced membrane binding of haemolymph calcium at low salinities seems to be the most plausible of her suggestions. Similar unexpected and unexplained anomalies about the behaviour of calcium have been reported in subsequent studies on other animals. I n a later related study Tucker (1970b) showed that the nerve conduction velocity in Scutus was reduced during periods of low salinity exposure; changes in ionic concentration and/or ratios, particularly of the divalent cations were probably responsible for this phenomenon. Since Tucker's work several similar studies have been performed on various invertebrates and fish, usually employing rather more generally applicable salinity regimes. For convenience these studies may be subdivided into three categories: (a) those concerned with osmoconformers, (b) those investigating osmoconformers which possess behavioural responses to salinity and (c) those concerned with osmoregulators.

212

J . DAVENPORT

(a) Osmoconformers. Stickle and Ahokas (1974) monitored the perivisceral fluid composition of three species of Pacific echinoderms, Pisaster ochraceus (Brandt), Cucumaria miniata (Brandt) and Strongytocentrotus drobachiensis (Muller) exposed to simulated estuarine conditions. Binyon (1966) had stated that echinoderms were an exclusively marine stenohaline group. A few asteroid starfish do occur in stable low salinity areas such as the Baltic and Black Seas, but these they have penetrated over long periods of geological time and so their essential intolerance of salinity change remains. However, Loosanoff (1945b) showed that the starfish Asterias forbesi (Desor) tolerated very dilute seawater for brief periods. Armed with this information and some field data which strongly suggested that the three echinoderms under investigation did encounter pronounced salinity fluctuations on the Alaska coast, Stickle and Ahokas were able to justify their use of simulated tidal salinity fluctuations between 30%,and lo%,. As may be seen from Fig. 40 the perivisceral fluid concentration changes were much damped by comparison with those of the external medium; environmental osmolarities fluctuated between 900 and 300 mOsmoles kg- while the minimum body fluid concentration in all three species approximated to 650 mOsmoles/kg. Shifts in concentrations of chloride, sodium, potassium and magnesium tended t o follow the form of osmotic changes, but calcium levels in the perivisceral fluid exhibited a pattern of change very different from the other ions. I n P . ochraceus and S. drobachiensis the calcium levels declined with lowered salinity, but more rapidly than would be expected from the size of the hydrated calcium ion (see Table XVI). However, calcium levels did not recover when the salinity rose again. Similar findings were reported for several bivalves by Shumway (1977a). I n contrast, the calcium concentrations of C . miniata hardly changed at all until salinities started to rise after the period of low salinity exposure when calcium levels actually rose above the initial values. No explanation was offered for these phenomena. Stickle and Ahokas apparently did not appreciate that the sequences of ionic loss rates in falling salinities (estimated from their data and shown in Table XVI) were, in all three species, different from those expected from a simple consideration of hydrated ion size. In a later study on molluscs, Stickle and Ahokas (1975)remarked on deviations from the expected sequences in the animals studied. Such interspecific differences merit further study-clearly losses of cations are not determined solely by the concentration gradient between the extracellular fluid concentration and the exterior together with the porosity/permeability of

213

ENVIRONMENTAL SIMCLATION EX PEKIMENTS

( 0

900 -

1

800 -

--

I

700-

0

p

600-

. I

-E

500-

L

g

400-

D

300-

100 -

200

1

1

1

1

1

1

1

1

1

1

1

1

1

95 90

g

= 0

85

0 n

80 E

:: L

75

2

70

0

1

2

3

4

5

6 Hour

7

8

9

1011

12

FIG.40. (a) Osmolarity of the ambient sea water of P. ochracew (open circles), S.drobaehiensis (open squares) and C. miniata (closed circles). (b) Perivisceral fluid (above) and yo body water (below) values for all three species (symbols as in (a)).Vertical bars represent 95% confidence intervals. Modified from Stickle and Ahokas (1974).

the integument. Since the work of Tucker (1970a,b) and of Stickle and Ahokas (1974) three very similar studies have been performed. Stickle and Ahokas ( 1975) studied amphineuran and prosobranch molluscs, Stickle and Howey (1975) investigated an oyster drill Thais haemastoma (L.), while Shumway (1977b) investigated four more

214

J. DAVENPORT

TAMEXVI. SEQUENCE O F CATIONIC LOSS RATESIN THREE SPk:.:('IES OY ECHISO1)ELLMS (ESTIMATED FROM THE FIGURES OF STICKLE ANI) AHOKAS, 1974) Species

Loss rate sequence

Piaaster ochraceus Cucumaria miniata Strongylocentrotus drobachiensis 'Ca2+ concentrations in the haemolymph of C . miniata did not decrease in falling salinities and

actually increased in the rising salinity area of the salinity regime.

species of echinoderms, all asteroid starfish. Damping of changes in body fluid osmotic and ionic concentrations in comparison with external salinity fluctuations was demonstrated in all of these animals, but apart from various interspecific differences little novel information was gleaned. The study by Stickle and Howey (1975) showed predictably that the body fluid concentrations of the oyster drill T . huemastoma exposed to diurnal 24h wavelength salinity fluctuations between 30%,and lo%,fell lower than when exposed to a 12 h wavelength cycle of the same amplitude. They also found that the time lag between external and internal osmotic and ionic changes was affected by the composition of the fresh water used as a dilutant in their experiments, presumably because of the influence of calcium on permeabilities. (b) Osmoconformers with behavioural reactions to salinity. The first, and so far the most comprehensive of osmotic/ionic studies upon osomoconformers which exhibit behavioural osmotic control, is that of Shumway (1977a). She worked upon six bivalve species; two sublittoral offshore forms, the queen scallop Chlamys opercularis L. and the horse mussel Modiolus modiolus L.; two were species with both littoral and estuarine distributions, the mussel Mytilus edulis and the oyster Crassostrea gigas while the last two species, Scrobicularia plana (da Costa) and the clam M y a arenaria were both characteristic of brackish water. Shumway collected a wealth of data, too numerous to present here in full, some of which are summarized in Figs 41, 42, 43 and 44. Throughout, her assumption was that these bivalves were osmoconformers. From a large body of work reviewed by Potts and Parry (1964) and Prosser (1973) this would seem entirely reasonable; the only marine bivalve known unequivocably to be an osmoregulator is the low salinity brackish water clam Rangia cuneata (Gray) (see Bedford and Anderson, 1972). The data for

ENVIRONMENTAL SIMULATION EXPERIMENTS

215

intertidal and estuarine bivalves demonstrate how effective behavioural osmotic control can be in protecting osmoconformers. Whereas the sublittoral C. opercularis, which cannot exclude the external medium since it cannot sustain shell valve closure (Brand and Roberts, 1973), exhibits an almost perfect osmoconformer response (see Fig. 41),M . edulis closes its shell valves completely to restrict fluctuations in the concentration of the haemolymph to half the amplitude of those of the external environment (see Fig. 42). Mya arenaria, a clam particularly well adapted to brackish water habitats

Hours

Hours

Hours

Hours

Hours

Hours

Hours

Hours

FIG.41. Chlamys opercularis. Changes in haemolymph osmolality, and N a + , Mg' and Caz+ concentrations during exposure to (a)50% seawater minimum sinusoidal salinity regime and (b) 50% seawater minimum square-wave salinity regime. Stippled areas represent changes in external medium. Each point is mean of three scallops. Error bars represent 95% confidence limits. From Shumway (1977a).

216 55

100

L

aJ 0

65 v)

.-" 30 12 Hours

Hours

100

24

100

10

L

W

c

L

0

gz 65g 5 6

65

s

do

.$ 30 24 Hours

Hours

30 2 0

12 Hours

24

Hours

Fu:. 42. Mytilus edulis. Changes in haemolymph (circles)and mantle fluid (squares) osmolality and Na', Mg2+ and Ca2+concentrations during exposure t o (a)30% and (b)0% seawater minimum sinusoidal salinity regime. Stippled areas represent changes in external medium. Arrows indicate points of shell-valve closure ( J. ) and opening ( f ) (from Bettison, unpublished). Each point is mean of three mussels. Error bars represent 95% confidence limits. From Shumway (1977a).

217

Hours

Hours

Hours

Hours

Hours

Hours

43. Myu arenaria. Changes in haemolymph osmolality and Na', Mgz+ and Ca2+ concentrations of 1 0 0 ~ oseawater-acclimated burrowed (open circles) and non-burrowed (filled circles) clams exposed to (a)30% seawater minimum sinusoidal salinity regime and (b) 30% seawater minimum square-wave salinity regime. Stippled area represents rhanges in external medium. Each point is mean ofthree clams. Error bars represent 95% confidence limits. From Shumway (1977a).

FIG.

218

.J. DAVENPORT

Hours

Hours

5 E V 0

Hours

Hours

Hours

Hours

Hours

FIG.44. Modiolus modiolus. Changes in haemolymph (circles) and mantle fluid (squares) osmolality and Na', Mgz+ and Ca2+ concentrations during exposure t o (a) 50% seawater minimum sinusoidal salinity regime and (b)50% seawater minimum square-wave salinity regime. Stippled areas represent changes in external medium. Arrows indicate points of shell-valve closure ( ) and opening ( ). Each point is mean of three bivalves. Error b a n represent 95'7' confidence limits. From Shumway (1977a).

4

t

cannot close its shell valves nor withdraw its well developed siphons. However the edges of the mantle, like the siphons are covered by periostracum and are completely united except for a small pedal aperture. I n conjunction with muscular sphincters at the tip of each siphon which control pumping (Chapman and Newell, 1956), these structural adaptations appear to minimize the influence of the external medium as effectively as shell valve closure does in M . edulis. M y a arenaria lives deep in the substratum, but, as Shumway's

ENVIKONMEKTAL SIMULATION EXPERIMENTS

219

data show, the burrowing habit does not contribute to osmotic control (see Fig. 43) and presumably has more importance as a protection against predators. The results for one species studied by Shumway were rather unexpected. As may be seen from Fig. 44,the horse mussel Modiolus modiolus exhibited quite effective behavioural osmotic control despite its large byssal aperture, and the general acceptance that its distribution is sublittoral. Previously Coleman and Trueman (1971) had shown that M . modiolus could not retain water within the mantle cavity during aerial emersion and Pierce (1970) using steady-state salinity experiments had established that the species could not survive below 80% sea water. Shumway suggested that her results might have some environmental significance since Coleman and Trueman reported that horse mussel beds were occasionally exposed on very low spring tides when they might be affected by rain water. However it seems likely that aerial emersion would simply allow water to seep out of the mantle cavity and salinity effects would be negligible. The observations (Davenport, 197913)on “propped open” Mytilus edulis appear to be relevant here. It seems likely that as long as horse mussels keep their inhalent and exhalent apertures closed and do not actively pump water through the mantle cavity, exchange of fluid between the mantle cavity and exterior will be poor, even though the byssal aperture prevents effective valve closure, just as wedging the shell valves of M . edulis apart does not ensure exposure of their tissues to external salinites. Only during aerial emersion will the patent byssal opening of Modiolus modiolus be disadvantageous. In support of these comments the author offers some further personal observations. I n northern Norway, horse mussels do occur intertidally, but only in rock pools, not where they may be aerially exposed. Some M . modiolus were observed in very shallow pools so high on the shore that they were exposed to strong melt water influence at low tide during the subarctic spring; the water surrounding the bivalves was virtually fresh. This suggests that Shumway’s observations are perfectly valid in the field for M . modiohs in some parts of its geographical range. I n subsequent studies three more species have been studied under simulated estuarine conditions, the American oyster Crassostrea virginica (see Hand and Stickle, 1977), the West African blood clam Anadara senilis L. (see Djangmah et al., 1979)and the Atlantic ribbed mussel Modiolus demissus (see Shumway and Youngson, 1979). All three species are euryhaline and can close their shell valves most effectively; they showed similar responses to those of Mytilus edulis and Crassostrea gigas.

220

J. DAVENPORT

As already discussed in Section VI.B, Shumway and Davenport (1977) showed that the lugworm Arenicola marina became quiescent and stopped irrigating its burrow when the salinity of the water overlaying its substrate was low. From Fig. 45 the effectiveness of this behaviour may be appreciated. Worms held naked in water fluctuating between 32 and 9%, experienced great changes in

2oo

r

rn-osmole

500

r

Na+

r

K’

l2

-

0

0

c

Ca2+

r

Mg2’

6o

30

r

sop

l2

12

24

-

0

0

12

24

Time (h)

FIG.45. The changes in coelomic fluid osmolality and Na’, K + , Mg”, Ca2+ and SO, concentrations of burrowed and non-burrowed A . marina exposed to a 30% seawater minimum sinusoidal salinity regime. Each point is a mean of five animals. Error bars at t h e 95% confidence level are smaller than the actual points. From Shumway and Davenport (1977).

ENVIRONMENTAL SIMULATION EXPERIMENTS

22 1

coelomic fluid osmotic and ionic composition as indeed is to be expected from the work of Schlieper (1929) who showed that the lugworm was an osmoconformer. I n contrast, worms allowed to burrow in seawater-laden sand beneath the fluctuating salinity water showed no significant changes in coelomic fluid composition whatsoever. Finally Shumway (197th) monitored changes in haemolymph osmolarity in hermit crabs exposed to salinity fluctuations. Naked hermit crabs manifest an osmoconformer response similar to the queen scallop, and become isosmotic with low salinities during 12 h wavelength square wave 32 + IS%, salinity regimes. Specimens of Pagurus which had been allowed t o keep shells, into which they retreated at low salinities, maintained significantly higher haemolymph concentrations than naked animals, thus demonstrating that some degree of protection is afforded by the shell. However, as Davenport et al. (1980)showed, this protection is somewhat marginal and cannot compete in effectiveness with the efficient structural/ behavioural mechanisms of more euryhaline species such as M . edulis and A . marina. (c) Osmoregulators. Few laboratory salinity fluct.uation simulation studies have been performed upon osmoregulators, although Spaargaren (1974) predicted from his equations that a combination of damping and osmoregulation would restrict internal osmotic changes in the shore crab Carcinus maenas (L.) to one third of the external fluctuations during a sinusoidal tidal salinity cycle. Findley and Stickle (1978) studied an even more euryhaline crab, Callinectes sapidus which was exposed to laboratory 24.8 h wavelength salinity cycles. These cycles were either of 20 + 10 + 20%,, 30 + 10 -+ 30X0or 10 + 30 -+ lo%,. The results for haemolymph osmolarities are displayed in Fig. 46 and demonstrate that internal osmotic conditions are almost independent of external fluctuations in the blue crab. I n none of the three regimes did the haemolymph osmolarity alter by as much as 100mOsmoles/kg despite external changes five to six times greater. Generally speaking ionic concentrations were equally stable as may be seen from the values displayed in Table XVII. Of particular interest are the haemolymph magnesium concentrations. It has been known for some time that there is a close relationship between activity and blood magnesium concentration in decapod crustacea. This was discovered by Robertson (1953, 1960) who attributed the effects to the interference of magnesium with neuromuscular transmission. The phenomenon has been further

222

J. DAVENPORT

-

900 800700

20-10-20

=::v -

400

300 200

2 .-

0 0 E

= -8 .5

Hemolymphw

L

A

,

;

:

:

:

w

&Ambient Water

-

-

900 800700600500-

30-10-30

400-

300 200 900

-

10-30-10

!!! 50 0 400

2 300 00

0 2 4 6 8 10 12 14 16 18 202224 26

discussed by Lockwood (1962) and Potts and Parry (1964). High blood magnesium levels are associated with slow moving crustacea (e.g. Maia squinado Latreille) while low haemolymph magnesium concentrations characterize active crabs and prawns. In the nominal 30 10 -+ 30%, regime mean magnesium levels in the haemolymph of C. sapidus varied from 41.9 -+ 32.2 -+ 45.1 m eq./litre while those of the medium ranged from 92 + 32.3 -+ 91 m ey./litre. Thus, when the external magnesium levels fell by 65% the internal concentrations only dropped by 23%. However, despite this damping, one would still predict from Robertson’s observations that activity would be increased at low salinity; certainly Findley et al. (1978) -+

223

ENVIRONMENTAL SIMULATION EXPERIMENTS

TABLE

(HL)

Ion Na'

XVII. LEVELS OF IONS ( M EQ / LITRE ) IN S E A WATER (S.W.) AND HAEMULYMPH BLUECRABS(Cullinectes supidus) EXPOSED TO 30 -+ 10 + 30% DIUHNAL SALINLTY REGIME (MODIFIED FROM FINDLEY AND STIC'KLE, 1978)

OF

0

S.W. 321

HL387514 C1- S.W. 360 HL 339+12 K + S.W. 108 HL 11.2k0.5 Mg2+ S.W. 92.0 HL 11.9f2.7

4.17 191 361f8 234 338511 5.4 100f3 603 37.5k1.7

Time (hours) 8.33 12.42 137 34659 163 335510 3.4 8.2504 37.5 34.0k1.4

101 343+8 106 330+8 3.2 7.450.4 32.3 32.2f1.7

16.58 213 351f8 246 323f8 6.4 81k0.4 722 32.9f1.7

20.75

24.83

314 365516 355 304+18 10.6 9.9f0.4 l0.8f0.7 87.1 91.0 37.151.6 45.1f3.1

273 363f9 308 314+12 9.0

Haemolymph values are means predicted from regression analysis with 95% confidence intervals.

found that oxygen uptake was enhanced at the low salinity extreme of such tidal cycles, but no measurements of activity were performed and it should be pointed out that Robertson's work was performed at high constant salinity. The only other study performed upon an osmoregulator so far is that of Davenport and Vahl (1979) who worked upon the intertidal benthic teleost Blennius pholis. The blenny often lives in gullies on the shore fed by freshwater streams which produce marked salinity decreases associated with low tides. I n a square wave tidal salinity regime fluctuating between 34%, and fresh water the plasma osmolality did not change significantly, but remained at about 380 mOsmoles/kg. House (1963) had earlier shown that the blenny's blood concentration did fall significantly when the fish was left in a salinity of lo%,for 48 h; however House also showed that Blennius differed from other euryhaline teleosts such as the eel Anguitla anguilla L. (see Keys, 1933) and the trout Salmo gairdnerii Rich (see Houston, 1957) in that it exhibits a very rapid physiological response t o salinity changes, switching in less that 5min from pumping salts out across the gills in concentrated media to taking up actively ions from hypoosmotic solutions. This rapidity of response, which operates in both directions, means that passive loss of salts caused by exposure to fresh water will be partially offset by active salt uptake so that a decline in blood concentration will occur relatively slowly. I n Davenport and Vahl's experiments i t was clear

224

d DAVEK’PORT

that 6 h exposure to fresh water was insufficient to cause a significant decline in plasma osmoconcentration. More data for osmoregulators would obviously be desirable. However, part of the reason for the lack of study follows from the generally greater mobility of osmoregulators by comparison with osmoconformers; they do not need to take refuge during periods of low external salinity and their very mobility makes it difficult to be certain of the precise short term salinity fluctuations that they encounter in nature. 2. Volume regulation

A detailed discussion of all of the differences and relationships between ionic regulation, osmoregulation and volume regulation would be out of place in this review since these topics have been dealt with elsewhere (e.g. Florkin, 1962; Potts and Parry, 1964; Gilles, 1979). There are two types of volume regulation. First there is regulation of the volume of extracellular fluid; this is sometimes referred to as whole animal volume regulation since a great proportion of the total volume of many animals consists of extracellular fluid. Because it is usually difficult to measure the volume of extracellular fluid or the volume of a whole animal, i t is a normal experimental procedure to weigh animals repeatedly while they are exposed to salinity changes to assess volume regulatory capacity. In forms such as the euryhaline osmoregulating crab Carcinus maenus, changes in urine output in response to salinity stress are so rapid that weight changes are almost undetectable (Schwabe, 1933). I n volume regulating osmoconformers such as Pagurus bernhardus initial volume ( = weight) changes of considerable magnitude may occur which can take many hours or even days to reverse (Davenport, 1972b). I n osmoconformers which have weak or negligible volume regulatory capacity, volume changes induced by salinity stress are never fully reversed except by further salinity changes in the opposite direction. In salinity simulation studies certain problems have arisen. First, it is tempting to assume that an animal which does not change weight in a fluctuating salinity regime is a volume regulator, whereas an animal whose weight fluctuates is not. However, it must be remembered that the original definitions of volume regulators stemmed from direct transfer experiments. Thus if one takes an osmoconformer such as a starfish from full strength sea water and places it immediately in a dilute medium, water will be taken up rapidly by osmosis while solutes

ENVIRONMENTAL SIMULATION EXPERIMENTS

225

diffuse outwards less quickly; consequently the animal will swell. If the animal’s integument is especially permeable to solutes, however, and the salinity changes of the environment relatively slow, then outward solute diffusion may be rapid enough to prevent the build up of substantial osmotic gradients and therefore prevent excessive swelling. This would appear to be the case for the three echinoderm species studied by Stickle and Ahokas (1974). Conversely, Shumway ( 1 9 7 8 ~noted ) substantial fluctuations in the volume of individual hermit crabs exposed to simulated tidal salinity regimes. Volume regulation in this species is a relatively slow process accomplished largely by changes in urine output which are not triggered until substantial weight changes have occurred. A second problem is experimental in origin. I n several cases i t is technically difficult to weigh single animals repeatedly during exposure to a salinity regime. An alternative procedure is to expose many animals to the regime and withdraw some at intervals which are then weighed, freeze dried and reweighed to allow calculation of the proportion of the animals’ weights which consists of water. If the latter rises during the course of the experiment the animals’ overall volumes are assumed to have risen and vice versa. Particular difficulty arises with bivalve molluscs since they have t o be cut out of their heavy non-living shells to be wet weighed and it is difficult to avoid significant haemolymph loss. If haemolymph is lost then the investigator will tend to be measuring tissue hydration rather than whole body hydration and will inadvertently be studying cellular volume regulation (see below)! To further complicate this picture some workers have unfortunately and inaccurately used the terms “Yobody water” and “Yotissue water” interchangeably. Because of these problems it is difficult to do more than divide the animals so far studied into those which do exhibit significant total volume changes in fluctuating salinity regimes and those which do not (see Table XVIII). Obviously changing total volumes are generally undesirable, since they will interfere with various biological processes, primarily locomotion and respiration, but also diffusion generally. The second type of volume regulation occurs at the cellular level; it is commonly known as intracellular fluid isosmotic regulation (Florkin, 1962). The cells of multicellular animals, whether osmoconformers or osmoregulators, are isosmotic with the surrounding body fluids (Conway and McCormack, 1953). If the body fluid concentration changes as a result of external influences, the cells will tend to swell or shrink because of osmotic gain or loss of water.

226

J . DAVENPORT

TABLE XVIII. SPECTIESSHOWING CONSTANT VOLUMES (A) OR V.4RIAKI.E VO1,I'MEh (B) IN FL~K~TUATIN(: SALINITY REGIMES (DATA FROM VAKIOL~S SOIWES) (A) Constant volume species

1. Echinoderms Pisaster ochraceus Cucumaria rniniata Strongyloeentrotus drobachiensis 2. Polychaetes Arenicoh marina (burrowed) 3. Molluscs Thais larnellosa

Source of information Stickle and Ahokas (1974) Stickle and Ahokas (1974) Stickle and Ahokas ( 1974) Shumway and Davenport (1977) Stickle and Ahokas (1975)

(B) Variable volume species 1 . Echinoderms Shumway (1977b) Asterias rubens Shumway (197713) Solaster papposus Shumway (1977b) Henricia sanguinolenta Shumway (1977b) Astropecten irregularis 2. Polychaetes Arenicola marina (non-burrowed) Shumway and Davenport (1977) 3. Molluscs (gastropods) Tucker (1970a) Scutus breviculus Stickle and Howey (1975) Thais haemastoma 4. Molluscs (amphineurans) Stickle and Ahokas (1975) Mopalia mucosa Stickle and Ahokas (1975) Katherina tunicata 5. Molluscs (bivalves) Hand and Stickle (1977) Crassostrea virginica 6. Crustaceans Shumwav (197%) Pagurus bernhardus

Experiments with isolated cells have shown that both swelling and shrinking do occur, but are reversed in cells from euryhaline species by regulatory processes which are much more rapid in action when reducing swelling than in resisting shrinking (Gainer and Grundfest, 1968; Gerard and Gilles, 1971; Pierce, 1971; Gilles, 1975); such processes are not exhibited by cells of stenohaline species. Over the past twenty years it has been demonstrated that a variety of intracellular free amino acids and other amino compounds (e.g. glycine-betaine) are involved as osmotically active compounds in cellular volume regulation in both osmoconformers and osmoregu-

ENVIRONMENTAL SIMULATION EXPERIMENTS

227

lators (for reviews see Florkin and Schoffeniels, 1969; Schoffeniels and Gilles, 1970, 1972; Schoffeniels, 1976). I n response to hypoosmotic stress, which tends to cause tissue swelling, intracellular amino compound concentrations fall (whether by extrusion or incorporation into proteins) in order to reduce intracellular osmotic pressure; hyperosmotic stress has the reverse effect. In bivalve molluscs for example an almost linear relationship between tissue free amino acid levels (measured as ninhydrin positive substancesN.P.S.-levels) and environmental salinities has been recorded for the osmoconformers Mytilus edulis and Crassostrea virginica (Lange, 1963; Lynch and Wood, 1966). Many studies have been devoted to confirming the role of the intracellular free amino acid pool in intracellular isosmotic regulation in a variety of animals; equal effort has gone into evaluating the importance of individual amino acids and the mechanisms for regulating the size of the pool (for a brief review of the relevant literature see Livingstone et al., 1979). All of this work featured either direct transfer experiments or was performed on animals taken from a variety of natural habitats where salinity was constant (e.g. Lange, 1963); not until the work of Shumway ( 1 9 7 7 ~and ) Shumway et al. (1977) were measurements made of cellular volume regulation in animals exposed to simulated estuarine conditions. Shumway ( 1977c) measured tissue water levels in eight bivalve species (Chlamys opercularis, Modiolus modiolus, M y a arenaria, Scrobicularia plana, Mytilus edulis, Cerastoderrna edule L., Mercenaria mercenaria L. and Crassostrea gigas). Unfortunately many of her data were from “propped open” animals and perhaps ought to be reinterpreted given the results of the subsequent study by Davenport (1979b). However, if normal animals alone are referred to i t is clear that all species except the quahog Mercenaria mercenaria showed significant changes in tissue hydration during square wave or sinusoidal salinity cycles with hydration levels rising at low salinity and falling when salinities rose again (see Fig. 47). “Propped open” quahogs exhibited similar tissue water changes, as did normal specimens of the ribbed mussel Modiolus demissus Dillwyn in a later study (Shumway and Youngson, 1979). In none of these species did the hydration level return to a higher level than the initial full seawater control level when the animals were returned to full sea water after a period of exposure to low salinities; taken with the observed fluctuating in tissue water content this indicated that the bivalves’ cells were behaving like simple osmometers with no volume control, little solute loss and passive ebb and flow of water. However, for Mytilus edulis alone, Shumway performed further measurements

228

J . DAVENPORT

HOURS

FIG.47. Changes in tissue water content of normal ( 0 ) and wedged open ( 0) M. edulis exposed to gradual and abrupt salinity fluctuations. Each point is a mean of three animals. Error bars represent 95% confidence limits. From Shumway ( 1 9 7 7 ~ ) .

upon mussels which had been held in a fluctuating salinity regime for 1 week. After this period she found that the tissue water content of the mussels remained almost constant throughout the salinity cycles and showed none of the changes exhibited by animals during the first 24 h of exposure. The corresponding N.P.S. determinations performed by Shumway etal. (1977)and Shumway and Youngson (1979) were most interesting. With nine species in all investigated, and both N.P.S. (ninhydrin positive substances) and individual amino acid determinations performed upon all of them in several salinity regimes, a great deal of information was collected, but a few generalizations may be made. It is clear that tissue N.P.S. levels do not simply fall and rise with salinity as might be predicted from the hypothesis of Florkin and Schoffeniels (1965, 1969). I n several cases, including M . edulis, falling salinity was associated with rising N.P.S. levels; Shumway et al. speculated that amino acid accumulation might result from anaerobic processes during shell valve closure, but unfortunately glycine, which formed a major part of the enlarged N.P.S. pool, is not produced during anaerobiosis in the mussel (De Zwaan et al., 1976). Specimens of M . edulis held for 1 week in a fluctuating salinity regime showed no significant N.P.S. concen-

ENVIRONMENTAL SIMlTLATION EXPERIMEXTS

229

tration changes during the tidal cycle; N.P.S. values simply remained steady at a value roughly midway between the extremes of N.P.S. changes shown by mussels during the first 24 h of exposure to the same regime. Shumway et al. interpreted all of this information to mean that mussels and other bivalves living in estuaries did not continually build up and lose intracellular free amino acids, but eventually assumed a constant tissue hydration and N.P.S. level which ensured a minimum expenditure of energy and material by the animals. Livingstone et aE. (1979) extended this approach with specimens of M . edulis exposed to salinity regimes which did not stress the animals enough to induce shell valve closure. The N.P.S. response during the first few tidal cycles was again equivocal, but over 48 cycles the mean N.P.S. level averaged over each cycle fell significantly indicating that adaptation in the form of isosmotic intracellular regulation had taken place. I n the only other simulation study involving tissue N.P.S. measurements Shumway and Davenport ( 1977) demonstrated that tissue N.P.S. concentrations in lugworms buried in sand did not change when the salinity of the overlying water fluctuated. On the other hand, naked specimens of Arenicola did show tissue N.P.S. changes in both sinusoidal and square wave salinity regimes; the patterns were not especially coherent although there was a general tendency for a fall in N.P.S. level as time elapsed. I n any case these data probably have no relevance for lugworms in nature. Stickle and his co-workers (Stickle and Howey, 1975; Hand and Stickle, 1977; Findley and Stickle, 1978) have also carried out work on invertebrates which is relevant to cellular osmoregulation but have monitored only N.P.S. levels in haemolymph? not tissue concentrations. This approach seems to stem from the observations of Pierce and Greenberg (1973)who showed that the isolated hearts of Modiolus extruded free amino acids during exposure to low salinity. Thus a rise in haemolymph N.P.S.would indicate a loss of free amino acids from the intracellular pool in response to reduced extracellular fluid concentrations. Stickle and Howey (1975) found that haemolymph N.P.S. rose in oyster drills as surrounding salinities fell, and Hand and Stickle (1977) reported similar data for the oyster Crassostrea virginica. These results support the cellular volume regulation hypotheses of Florkin and Schoffeniels. However, Findley and Stickle (1978) also reported some haemolymph N.P.S. fluctuations in the blue crab Callinectes sapidus. Although these fluctuations were described as “minor” by Findley and Stickle, changes of as much as 100% occurred during a 2 0 4 10+20%,

230

J. DAVENPORT

salinity cycle, which seems very strange in a crab with such pronounced control over extracellular fluid concentration. However Findley and Stickle’s haemolymph N.P.S. values were far below those reported for the crab by other workers, possibly due to starvation, so the results may not have been applicable to normal animals. It should also be pointed out that relying solely upon haemolymph N.P.S. concentrations for information about cellular volume regulation is somewhat unsafe. It is by no means clear that amino acids are always extruded intact from the cells, particularly in crustaceans in which Florkin and Schoffeniels (1965, 1969) proposed that intracellular free amino acids were degraded into keto acids and ammonia. I n summary it would appear that regulation of the intracellular amino acid pool is not effective in preventing cellular volume changes induced by tidal salinity fluctuations. However, in animals exposed to such fluctuations over long periods of time the size of the pool is altered t o minimize volume changes.

G. Oxygen consumption The cost in energy terms of existence in variable as opposed to stable habitats is obviously of interest to both physiologists and ecologists. As far as salinity fluctuations are concerned, the available evidence is confusing. Dehnel (1962) suggested that oxygen consumption would be enhanced at low salinity because of osmoregulatory (and volume regulatory?) work. However calculations based on thermodynamic criteria by Potts and Parry (1964) suggest that osmoregulation of the extracellular fluids should cost very little. I n any case the majority of intertidal and estuarine species are osmoconformers. No data appear to be available about the minimal cost of isosmotic intracellular osmoregulation. On the other hand, as discussed in earlier sections, fluctuating external salinities have profound effects on the behaviour and physiology of intertidal and estuarine organisms which are likely to alter respiration rates in a variety of ways independently of the cost of osmoregulatory work. Before the development of simulation equipment, studies of the effects of salinity on oxygen uptake were technically unsatisfactory because animals were physically disturbed whenever they were transferred from one salinity to another-a procedure bound to alter respiration at least temporarily. Independently Davenport (see Shurnway, 1978a) and Findley et al. (1978) developed through-flow respirometers which allowed continuous measurement of oxygen

ENVIRONMENTAL SIMULATION EXPERIMENTS

23 1

uptake during salinity fluctuations without otherwise disturbing the animals under investigation. A further refinement of such apparatus described by Vahl and Davenport (1979) allows feeding during exposure to experimental regimes. The first worker t o use such equipment was Bettison (unpublished data) who monitored oxygen uptake in several estuarine bivalve species. All showed zero oxygen uptake when isolated from deleterious external salinities by shell valve closure or siphon retraction. Usually oxygen uptake was apparently enhanced after such a period of isolation but some of this increase was probably an artifact caused by expulsion of deoxygenated water from the bivalve’s mantle cavity. Similar data was later collected by Shumway and Youngson (1979) for the Atlantic ribbed mussel Modiolus demissus. Shumway (1978a,b, c, 1979) also monitored oxygen consumption in a number of other animals which respond behaviourally to salinity fluctuations. These were the sea anemone Metridium senile, the sea squirt Ciona intestinalis, the hermit crab Pagurus bernhardus and a number (11 species) of gastropod molluscs. Except for P. bernhardus all behaved in similar fashion with low ( M . senile) or negligible (all other species) oxygen uptake occurring at low salinities when the animals reduced contact with the environment and ceased producing respiratory currents. Oxygen consumption results for P. bernhardus (see Shumway, 197th) are difficult to understand since no alterations in oxygen uptake were observed in salinity regimes (32 --+ 16 --+ 32%,) except for brief increases which occurred at 24%,and appeared to be associated with transient bursts of activity. I n a later study Davenport et al. (1980)showed quite clearly that there was no oxygen uptake by hermit crabs which had withdrawn into their shells in response to low salinity. Obviously if the scaphognathites continue to direct a respiratory current through the branchial chambers water would necessarily be taken up as well as oxygen and the animal would gain no osmotic benefit from withdrawal into the shell. Since Shumway demonstrated osmotic benefit the continuation of oxygen uptake seems inconsistent. Oxygen uptake measurements have been performed by Findley et al. (1978)and by Davenport and Vahl(l979)on animals which do not isolate themselves at low salinity. Findley et al. (1978) investigated the osmoconforming predatory whelk Thais haemastoma and the osmoregulating crab Callinectes sapidus collected from the same habitat and exposed to both steady and fluctuating salinities. They showed dissimilar respiratory patterns; T . haemastoma acclimated to

232

J . DAVENPORT

steady salinities showed higher respiration rates at 30%,than at lo%, and a reduction in oxygen consumption during salinity fluctuations which deviated in either direction away from an acclimation salinity. In contrast, oxygen uptake by the blue crab was higher at constant, lo%, or 20%, than at 30%,, and during salinity fluctuations was always greatest at the lowest salinities. The results for C. sapidus can be interpreted in terms of heightened oxygen uptake which could be caused by increased activity at low salinity resulting from lowered haemolymph magnesium concentrations or from the cost of active inward transport of ions. Findley et al. suggested that the results for T . haemastoma followed from its incomplete volume regulation capacity. Presumably a general reduction in physiological efficiency resulted from either osmotic swelling or shrinkage, causing oxygen uptake t o fall in response to salinity changes in either direction. Davenport and Vahl (1979) investigated the intertidal teleost Blennius pholis in square wave and sinusoidal salinity regimes fluctuating between 0 and 33-5%,.The results indicated heightened oxygen consumption at low and rising salinities with minimum uptake rates at high salinity. This again suggests a correlation between oxygen consumption and osmotic work, but the changes in oxygen consumption during salinity cycles were relatively small ( c . 25%). Kinne (196410) and Holliday (1971) have stressed the well known fact that the oxygen content of water depends upon salinity and Holliday stated that since oxygen uptake of teleost fish is to a large extent determined by the oxygen concentration of the surrounding water it is difficult to assess whether a change in respiration rate is related to salinity or oxygen concentration. True, in Davenport and Vahl’s experiments, as with those of Findley et al., salinity fluctuations were accompanied by changes in oxygen content (but not oxygen tension which remained constant) between 6.4 ml 02/1at 33Ym and 8.0ml 02/1in fresh water. The change in oxygen content (c. 25%) is comparable with the changes in oxygen uptake displayed by the blenny. However, it should be pointed out that Holliday’s statement that uptake is a function of concentration is dubious since oxygen tension-the equivalent of its thermodynamic po ten tial-and not its concentration, controls oxygen ex change across the gills. Also, the uptake changes in the blenny were asymmetrical whereas the oxygen content of their environment fluctuated symmetrically with salinity. Whatever the underlying cause the energy cost to the blenny is higher in a fluctuating salinity regime than at constant high salinity. Subsequently Vahl and Davenport (1979) measured the increase in metabolic rate of the

ENVIRONMENTAL SIMULATION EXPERIMENTS

233

blenny, expressed as enhanced oxygen consumption, following a meal. This increase, known as “specific dynamic action (S.D.A.) of the food consumed” (Kleiber, 1961) may be as great as 60% and reaches a peak 2 4 h after a meal. Vahl and Davenport calculated that S.D.A. and the energetic cost of living in fluctuating salinity environments could interact to reduce the blennies’ scope for activity seriously if large meals were eaten at the “wrong” time during a tidal cycle (see Fig. 48).Of course the blenny tends to be a browser (Qasim, 1957), thus spreading the influence of S.D.A., but it does take large meals in the wild on occasion (Grove, personal communication). VII. OXYGENTENSION STUDIES I n the open sea oxygen tensions at the surface are usually reasonably high and apparently do not limit animals in any way. Abnormally low or high oxygen tensions only occur in a few specialized areas, and are usually associated with equally unusual temperatures or pH, carbon dioxide or sulphide levels. Some habitats are usually anaerobic, for example beneath the surface of mud flats or some deeper areas of enclosed seas (e.g. Black Sea). On the other hand, turbulent open coast areas or waters characterized by dense kelp beds may feature intermittent oxygen supersaturation (Shelford and Powers, 1915). Some specialized habitats feature pronounced oxygen tension fluctuations; such occurrences in rockpools have already been discussed in Section I1 (2)but similar changes between zero and 2-300% air saturation have also been reported in Zostera beds (Powers, 1920; Broekhuyser, 1935). Organic pollution, usually by sewage, together with resultant eutrophication can result in low or fluctuating oxygen tensions in estuaries or enclosed areas of the sea. Rather less research has been directed at determining oxygen tension limits to survival and reproduction in marine animals than for either temperature or salinity; the most recent review of the field appears to be that of Vernberg (1972). Many physiological and biochemical studies have been performed on the dependence of oxygen consumption on external oxygen tension, the importance of blood pigments as oxygen carriers and the workings of anaerobic metabolism. The literature devoted to these topics is too large and diffuse to be reviewed here and in any case, as for the tolerance studies, no work has been performed under conditions of fluctuating oxygen tension. Cycles of oxygen tensions were first used in an experimental study

234

J . DAVENPORT

40

A

- scope for activity

feeding

60

salinity 30

40 20

20 10 0

0

feeTng

eeding

-J

v

40

I

x .= .1

0

60 c

a,

a

30 40

x

z

.r m

.5

20

2

5 E"

E .-

0

20 10

X

L

8 0

40

C

0

feerng

60

30 40

20 20

10 0

0

Hours

FIG.48. Blennius pholis. Combined effects of salinity and S.D.A. upon scope for activity; models for various feeding times. From Vahl and Davenport (1979).

by Davenport and Fletcher (1978) on the mussel Mytilus edulis. Shumway (1977a) showed that, in a sinusoidal salinity regime, the osmolarity of the mantle fluid of mussels closely followed that of the external environment until shell valve closure occurred; the mantle fluid concentration then remained virtually constant at the

ENVIRONMENTAL SIMULATION EXPERIMENTS

235

equivalent of about SO%, until the shell valves reopened in response to rising salinities (see Fig. 42). I n a separate study Bettison and Davenport (in preparation) showed that the oxygen content of the mantle fluid of N y t i l u s fell dramatically when the she11 valves closed in response to a n abrupt salinity change, from a mean value of 5-6ml O,/litre just before closure to 0.9 ml O,/litre 15 min later. After 6 h of valve closure the oxygen content was still 0.6 ml O,/litre: evidently the mantle fluid does not become completely anoxic. To summarize, the mantle fluid of Mytilus varies substantially in both salinity and oxygen tension during an estuarine salinity cycle. Davenport and Fletcher were interested in how these conditions affected the ciliary activity of the gills of M . edulis. I n their studies they used gill preparations and methods of the type developed by Ajana (1975)from the earlier techniques of Gray (1923, 1924). Salinity and oxygen tension regimes of the type shown in Fig. 49 were applied separately or in combination to the gill preparations. The results are summarized in Figs 50, 51 and 52. Reduced oxygen tensions or salinities acting alone reduced ciliary activity (expressed as particle transport rates) by about 40%. When both were combined (see Fig. 52) ciliary activity fell by about 40y0 during the period of gently falling salinity and normal oxygenation; when low oxygen tension was added to the salinity stress the ciliary activity fell by a further 25%. On a return to high salinities and oxygen tensions the cilia completely recovered and it seems certain that such regular fluctuations in ciliary activity occur under natural conditions. The only other study which has involved laboratory simulations of oxygen tension fluctuations is that of Ritz (1980)who investigated the effects of fluctuating conditions of salinity, oxygen tension and copper concentration on the intertidal amphipods Gammarus duebeni Liljeborg and Marinogammarus marinus (Leach). Both species survived tidal salinity fluctuations between full sea water and pure fresh water for 4 days, but when a tidal oxygen tension fluctuation between 100% and 4% air saturation was superimposed upon this salinity regime (with low salinity and low oxygen tension being simultaneous), 50% of M . marinus were dead in 2 days in square wave regimes; no mortality occurred in G. duebeni or in either species exposed to sinusoidal regimes. Unfortunately no data for oxygen tension acting alone were available but this study reinforces the conclusions derived from many multivariate steady-state investigations that separate stresses may interact and reinforce each other.

236

J. DAVENPORT

100

--

1

g

50

v)

0 I

,

I

I

,

I

05

0

5

50

-

-

c

a

01-

001

0-

0 I

0

I

6 H0”E

1

I

6

12

Hours

12

FIG.49 (left). Simulated mantle cavity conditions in estuarine salinity cycle. Upper graph indicates salinity levels with the solid line representing the mantle cavity concentrations and the circles representing the hypothetical environmental salinities. Lower graph shows the o saturation is equivalent to about 150 mm Hg. Arrows oxygen tension conditions: 1 0 0 ~ air indicate the onset and cessation of the stimulated shell valve closure period. From Davenport and Fletcher (1978). FIG.50 (right). Effects of salinity upon frontal ciliary activity. Upper graph represents the salinity regime. Lower graph shows ciliary activity; the symbols represent mean values for five gill preparations, with 95% confidence intervals. From Davenport and Fletcher (1978).

VIII. POLLUTANT STUDIES The literature devoted t o the effects of pollutants upon marine organisms is vast and unwieldy, but the reviews by Bryan (1971)and Phillips (1977a), together with the report of the symposium organized by Cole (1979) are relevant here. As described in earlier sections the mussel reacts to fluctuating external salinities by shell closure during periods of low salinity. Clearly in an estuarine mussel bed the tissues of M . edulis are not exposed to the full influence of the freshwater input. I n polluted estuaries pollutants are often freshwater borne and hence both pollutant and freshwater influences are closely linked.

2137

ENVIRONMENTAL SIMI'LATIOS EXPERIMENTS

100

is . 0

50 P

4

I

O J

,

I

6

12

00 J

0

OOJ I

0

I

I

Hours

0

6

12

Hours

FIG.51 (left).Effects of oxygen tension upon frontal ciliary activity. Upper graph represents the oxygen tension regime. Lower graph shows ciliary activity; the symbols represent mean values for five gill preparations, with 95% confidence intervals. Prom Davenport and Fletcher (1978). FIG.52 (right). Combined effects of salinity and oxygen tension upon frontal ciliary activity. Upper graph indicates the regimes used; solid line indicates salinity levels, dashed line represents oxygen tensions. Lower graph shows ciliary activity: the symbols represent mean values for five gill preparations, with 95% confidence intervals. From Davenport and Fletcher (1978).

Given the closure response to low salinities exhibited by Mytilus the closure mechanism may protect the bivalve against some of the effects of freshwater-borne pollutants; this possibility was independently suggested by Phillips (1977b) and Davenport (1977). Davenport (1 977) decided to use copper to test this hypothesis. The metal was chosen because there was considerable data concerned with copper toxicity in the antifouling literature while Bryan and Hummerstone (1971) showed that i t was common in estuaries associated with dumped mining wastes. First a control experiment was performed with continuous levels of either 0.5 ppm or 0.25 ppm copper being added to sea water ( 3 3 3 3 which was delivered to mussels. I n the former Concentration all animals were dead in 3 4 days, the median lethal time (M.L.T.) being about 2 days. In 0 2 5 p p m copper the M.L.T. was 4-5 days. However if copper was

238

J. DAVENPORT

delivered intermittently to mussels during a square wave tidal salinity cycle with copper being associated with low salinity and its absence occurring at high salinity, no mortality occurred in a 5-day experiment (see Fig. 53). On the other hand, if copper was added during the phase of high salinity, mortality was rapid (seeFig. 54).So far everything agreed with Davenport’s and Phillips’ hypotheses that shell valve closure induced by low salinity acted as a protection against pollutants as well as fresh water. However, a further experiment was performed with intermittent (6 h on, 6 h off) 0-5ppm copper being delivered to mussels maintained in full strength sea water. As may be seen from Fig. 55 mortality was again zero. It rapidly became clear that mussels could actually detect heightened external copper concentrations and close their shell valves to avoid

6

trppp; 0

1

2

3

4

12 Time (h)

5DavrO

24

1

2

3

4

5

12 Time (h)

24

FIG53 (left).M y t i l u s edulis. The effects of discontinuous 0 5 ppm added copper delivered under fluctuating salinity conditions. Copper on at low salinities. A, Percentage of animals able to form byssus during previous 24 h. B, Cumulative percentage of animals unable to maintain valve closure. C, Cumulative percentage of animals moribund. D, Cumulative percentage mortality. Squares = first experimental run. Triangles = second experimental run. F I G 54 (right).M y t i l w edulis. The effects of discontinuous 0 5 ppm added copper delivered under fluctuating salinity conditions. Copper on a t high salinities. A-D, as in Fig. 53 Squares = first experimental run. Triangles = second experimental run. From Davenport (1977).

0

0-

1

2

3

4

5DayrO

12

1

2

3

4

5

24

Time (h)

FIG.55. Mytalus edulis. The effects of discontinuous 0.5ppm added copper delivered under constant lOOyo S.W. conditions. A, percentage of animals able to form byssus during previous 24 h. B, Cumulative percentage of animals unable to maintain valve closure. C , cumulative percentage of animals moribund. D, cumulative percentage mortality. Squares = first experimental run. Triangles = second experimental run. From Davenport (1977).

Valves open

FIG.56. Strain-gauge traces of four successive periods of shell valve opening in a mussel exposed to a discontinuous added copper regime in lOOyo S.W. From Davenport (1977).

240

.J. DAVENPORT

damage; some stress gauge traces which confirm this are shown in Fig. 56. In subsequent studies by Davenport and Manley (1978). Manley and Davenport (1978) and Manley (in preparation) the concentration threshold which initiates changes in behaviour in Mytilus (about 0.02 ppm total copper in 33%, Menai Strait sea water) has been determined. The phenomenon has been demonstrated in several other bivalves (Crassostrea gigas, Modiolus demissus, Modiolus modiolus and Anadara senilis), and is now known to be more clear cut for copper than for other heavy metals. It is still debateable whether the response t o copper is ever of significance in nature, since fluctuations in copper content of estuarine waters with mussel populations have yet to be demonstrated. However, the results do cast a measure of doubt upon the value of mussels as biological indicators of pollution. The common bivalves meet many of the criteria for biological monitors proposed by Butler et al. (197l ) , but the usefulness of a biological monitoring system depends upon its providing an accurate integration of all changes in environment pollutant levels. If it fails to register transient or recurrent short term slugs of highly polluted water because of its ability to isolate its tissues from them, the integration could be misleading. Ritz (1980), employing the same equipment and techniques, imposed simultaneous salinity, oxygen tension and copper fluctuations upon Marinogammarus marinus and Gammarus duebeni; the multiple variables make it difficult to abstract conclusions about single factors. However it is clear that G. duebeni survives intermittent exposure to 1 ppm added copper in sea water whereas continuous exposure is rapidly lethal. Presumably periodic access to clean sea water allows removal of the accumulated pollutant in the urine (Bryan, 1971). In fresh water the situation is quite different, both intermittent and continuous 1 pprn copper regimes are equally lethal. However, in low salinities copper is known to be especially toxic (Jones, 1975; Jones et al., 1976). Moreover, G. duebeni is faced with maintaining active ion uptake and producing hypo-osmotic urine both of which can reduce resistance to pollutants (Inman and Lockwood, 1977 ). IX. CONCLUSIONS The value of environmental simulation experiments lies in their ability to reveal responses and mechanisms not predictable from steady state experiments. Many of the studies reviewed here meet

EKVIHONMESTAL SIMULATIOX EXPERIMENTS

24 1

this criterion at least to some extent, though in some cases, of course, the results obtained were not entirely unexpected. For example, given the excellent field work of Milne (1940),the data obtained by Shumway (1977a) concerning mantle fluid and haemolymph concentrations in Mytilus exposed to fluctuating salinities were predictable except in detail. On the other hand many of the thermal and salinity tolerance limits found for species exposed to cyclic regimes were dramatically different from those previously accepted. Similarly exciting were, for example, the observation by Diaz (1975) that damage to oyster larvae by brief exposure to high temperature may not be expressed for several days, a feature of great significance to oyster fisheries operating near power stations. The remarkable behavioural osmotic control exerted by burrowed lugworms (Shumway and Davenport, 1977),and the subtle isolation behaviour of Mytilus by closing the exhalant aperture (Davenport, 1979a) could not have been foreseen. Throughout the simulation study literature, are scattered criticisms of steady-state experiments. I n most cases these comments are valid though perhaps expressed with the overstatement of the enthusiast! However it seems clear that simulation studies with varying factors must represent a supplementary approach to the conventional methods rather than a replacement. Some potential pitfalls are evident in fluctuating factor experiments. Once simulation equipment is available it is tempting to use it indiscriminately. There is a fine dividing line between using say a sublittoral species for comparative purposes in a study primarily aimed at understanding the biology of its intertidal and estuarine relatives, and using the species simply because it is available. Unless there are good scientific reasons which indicate otherwise (e.g. to elucidate basic mechanisms), it seems logical that only animals which have been proved t o encounter physico-chemical fluctuations in nature should be exposed to idealized simulations in the laboratory; here we need much more information. I n this respect especial care should be taken with highly mobile marine animals (e.g. fish, crustacea) which may live in variable habitats but avoid fluctuations by swimming, crawling or making appropriate vertical migrations. Despite these various problems, simulation studies appear to have a bright future. Likely trends in equipment trends and techniques have been referred to in Section 111. Experimental possibilities are probably endless and certainly unpredictable, but a few tentative suggestions may be made. First, it seems probable that

242

J. DAVENPORT

fluctuations in environmental factors such as pressure, pH or food concentration (for filter feeders) may be added to the present repertoire. Pollutants other than heavy metals might also be investigated (e.g. oil, dispersants). Secondly, it appears likely that studies involving fluctuations in salinity or temperature may tackle more fundamental questions of ion transport or enzyme behaviour. Finally, and of particular interest, is the idea that animals may live and grow better if they are exposed to variable rather than constant conditions. Most evidence for this derives from behavioural or psychological studies on terrestrial vertebrates. However, 30 years ago Allee et al. (1949) demonstrated that aquatic organisms might survive variable thermal conditions better than they do constant temperatures. Most of the factor fluctuation studies performed so far have concentrated upon the effects of extreme stresses, but what consequences would low amplitude factor fluctuations have for development in marine organisms? The equipment and methods reviewed here could provide the answers.

X. ACKNOWLEDGEMENTS I am grateful to Professor D. J. Crisp, C.B.E., F.R.S. for critically reading the manuscript. I also wish to thank Mrs D. Clapton and Mrs E . Hughes who patiently typed it. XI. REFERENCES Ahokas, R. A. and Sorg, G. (1977). The effect of salinity and temperature on intracellular osmoregulation and muscle free amino acids in Fundulus diaphanus. Comparative Biochemistry and Physiology 56A,101-105. Ajana, A. M. (1975).Effect of temperature and salinity on the ciliary activity of bivalves. M.Sc. Thesis, University of Wales. Alderdice, D. F. (1963). Some effects of simultaneous variation in salinity, temperature and dissolved oxygen on the resistance of young Coho salmon to a toxic substance. Journal of the Fisheries Research Board of Canada 20 (2), 525-550. Alderdice, D. F. (1 972). Responses of marine poikilotherms to environmental factors acting in concert. In “Marine Ecology” (0. Kinne, ed.), Vol. I , Part 3, pp. 1659-1722. Wiley Interscience, New York. Alderdice, D. F. and Velsen, F. P. J. (1971).Some effects of salinity and temperature on early development of Pacific herring (Clupea pallasi).Journal of the Fisheries Research Board of Canada 28, 1545-1562. Alderdice, D. F., Wickett, W. P. and Brett, J. R. (1958).Some effects of temporary exposure to low dissolved oxygen levels on Pacific Salmon eggs. Journal of the Fisheries Research Board of Canada 15 (2), 229-249.

ENVIRONMENTAL SIMITLATION EN PEKIMENTS

243

Allee, W. C., Emerson, A. E., Park, T., Park, 0. and Schmidt, K. P. (1949). “Principles of Animal Ecology.” W. B. Saunders, New York. Andrews, J. D. (1954).Setting of oysters in Virginia. National Shellfish Association Proceedings 45, 3 8 4 5 . Atkins, B. and Ritz, D. A. (1977).New instrument for supplying water of fluctuating salinity. Laboratory Practice 26 (2), 99-100. Avens, A. C. and Sleigh, M. A. (1965).Osmotic balance in gastfopod molluscs. I . Some marine and littoral gastropods. Comparative Biochemistry and Phyaiology 16, 121-141. Baggerman, B. (1960). Salinity preference, thyroid activity and the seaward migration of four species of Pacific salmon. Journal of the Fisheries Research Board of Canada 17,296-322. Barnes, H.(1953).The effect of lowered salinity on some barnacle nauplii. Journal of Animal Ecology 22, 328-330. Barnes, H. and Barnes, M. (1958). Note on the opening response of Balanus balanoides (L.) in relation to salinity and certain inorganir ions. Veroffentlichungen des Instituts f u r Meeresforschung in Bremerhaven 5, 160-1 64. Bayne, B. L. (1976). “Marine Mussels: Their Ecology and Physiology”, 506 pp. Cambridge University Press. Bedford, W. B. and Anderson, J. W. (1972). The physiological response of the estuarine clam, Rangia cuneata (Gray) to salinity. I . Osmoregulation. Physiological Zoology 45, 255-260. Bhlehridek, J. (1935). “Temperature and living matter.” Borntraeger, Berlin (Protoplasma Monographs 8). Bettison, J. C. and Davenport, J. (1976). Salinity preference in gammarid amphipods with special reference to Marinogammarus marinus (Leach).Journal of the Marine Biological Association of the United Kingdom 56, 135-142. Bhatnagar, K . and Crisp, D. J. (1965).The salinity tolerance of nauplius larvae of cirripedes. Journal of Animal Ecology 34, 41M28. Bousfield, E. L. (1955). Ecological control of the occurrence of barnacles in the Miramichi estuary. Bulletin of the National Museum of Canada, 137, 1-69. Box, G. E. P. (1954).The exploration and exploitation of response surfaces; some general considerations and examples. Bimetrics 10 ( l ) ,16-60. Box, G. E. P. (1956).The determination of optimum conditions, I n “Design and Analysis of Industrial Experiments” (0.L. Davies, ed.), pp 495-578. Oliver and Boyd, London. Box, G. E. P. and Wilson, K . B. (1951).On the experimental attainment of optimum conditions. Journal of the Royal Statistical Society (Series B ) 13 ( l ) , 1 4 5 . Brand, A. R. and Roberts, D. (1973). The cardiac responses of the scallop Pecten maximus to respiratory stress. Journal of Experimental Biology and Ecology 13, 2M3. Briand, F. J.-P. (1975). Effects of power plant cooling~. systems on marine phytoplankton. Marine Biology 33,-135-146. Brett, J. R. (1944). Some lethal temperature relatives of Algonquin Park fishes. University of Torcmto Studies in Biology Series 52; Publications of the Ontario Fisheries Research Laboratory 63, 1 4 9 . Brett, J. R. (1970).I n “Marine Ecology” (0.Kinne, ed.), Vol. I, Part 1, pp. 515-560. Wiley Interscience, New York. Broekhuysen, G. R. J r (1935).The extremes in percentages of dissolved oxygen t o

244

J. DAVENPORT

which the fauna of a Zostera field in the tidal zone a t Nieuwdiep can be exposed. Archives nierlandaises de Zoologie 1, 339-346. Bryan, G. W. (1971).The effects of heavy metals (other than mercury) on marine and estuarine organisms. Proceedings of the Royal Society (Series 3)177, 38-10. Bryan, G. W. and Hummerstone, L. G. (1971).Adaptation of the polychaete Nereis diversicolor to estuarine sediments containing high concentrations of heavy metals. I . General observations and adaptations to copper. Journal of the Marine Biological Association of the United Kingdom 51, 845-863. Bull, H . 0. (1938). Studies on conditioned responses in fishes. Part VIII. Discrimination of salinity changes by marine teleosts. Reports of the Dove Marine Laboratory 5, 1-19. Butler, P. A., Andrew, L., Bonde, G. J . , Jernelow, A. and Reish, D. J . (1971). Monitoring organisms. F.A.O. Fisheries Reports No. 99 (Suppl. l ) , 101-1 12. Calabrese, A . (1969).Individual and combined effects of salinity and temperature on embryos and larvae of the coot clam MuEinia lateralis (Say). Biological Bulletin. Marine Biological Laboratory, Woods Hole, Mass. 137, 417-428. Cawthorne, D. F. (1978).Tolerance of some cirripede nauplii to fluctuating salinities. Marine Biology 46, 321-325. Cawthorne, D. F. (1979a).Some effects of fluctuating temperature and salinity upon cirripedes. Ph.D. Thesis, University of Wales. Cawthorne, D. F. (1979b). A comparative study of the closure responses of some cirripede species exposed to falling seawater concentrations. Journal of the Marine Biological Association of the United Kingdom 59, 811-817. Cawthorne, D. F. (1980). Tolerances of nauplii of Balanus balanoides (L.) and Elminius modestus (Darwin) t o cyclical temperature fluctuations. Journal of Thermal Biology 5 , 253-255. Cawthorne, D. F. and Davenport, J. (1980).The effects of fluctuating temperature, salinity and aerial exposure upon larval release in Balanus balanoides and Elminius modestus. Journal of the Marine Biological Association of the United Kingdom 60, 367-377. Chapman, G. and Newell, G. E. (1956).The role of the body fluid in the movement of soft-bodied invertebrates. 11. The extension of the siphons of M y a arenaria and Scrobicularia plana. Proceedings of the Royal Society (Series B ) 145, 564-580. Christiansen, M. E . and Costlow, J. D. J r (1975). The effect of salinity and cyclic temperature on larval development of mud-crab Rhithropanopeus harrisii (Brachyura : Xanthidae) reared in the laboratory. Marine Biology 32, 215-221. Christiansen, M. E., Costlow, J. D. and Monroe, R. J . (1977a).Effects ofthe juvenile hormone mimic ZR-515 (Altosid@')on larval development of the mud-crab Rhithropanopeus harrisii in various salinities and cyclic temperatures. Marine Biology 39, 269-279. Christiansen, M. E., Costlow, J . D. Jr and Monroe, R. J. (197713). Effects of the juvenile hormone mimic ZR-512 (Altozar@)on larval development of the mudcrab Rhithropanopeus harrisii at various cyclic temperatures. Marine Biology 39, 281-288. Cole, H . A . (1979). The assessment of sublittoral effects of pollutants in the sea-a discussion organised by H. A. Cole in collaboration with the marine pollution subcommittee of the British National Committee on Oceanic Research. Philosophical Transactions of the Royal Society of London (Series B ) 286,397-633. Coleman, N . and Trueman, E. R. (1971). The effects of aerial exposure on the activity .of the mussels Mytilus edulis and Modiolus modiolus. Journal of Experimental Marine Biology and Ecology 7,295-304.

ENVIRONMENTAL SIMULATION EXPERIMENTS

245

Conway, E. J . (1956). Fundamental problems in hormonal control of water and salt electrolyte metabolism. I n “The Hormonal Control of Water and SaltElectrolyte Metabolism in Vertebrates” (I. C. Jones and P. Eckstein, eds). Memoirs of the Society for Experimental Endocrinology Vol. 5 , Part 11, pp. 3-22. Cambridge University Press, London. Conway, E. J. (1960). Principles underlying the exchange of sodium and potassium ions across cell membrane. Journal of General Physiology 43,Suppl. 1 , 17-48. Conway, E. J. and McCormack, J. I. (1953). The total intracellular concentration of mammalian tissues compared with that of the extracellular fluid. Journal of Physiology, London 120,1-14. Costa, C. J. and Pritchard, A. W. (1978). The response of Mytilus edulis to short) duration hypo osmotic stress. Comparative Biochemistry and Physiology 61A. 14S155. Costlow, J . D. Jr (1967). The effect of salinity and temperature on survival and metamorphosis of megalopa of the blue crab Callinectes sapidus. Helgolander wissenschaftliche Meeresuntersuchungen 15,84-97. Costlow, J.D. and Bookhout, C. G. (1971).The effect ofcyclic temperatureson larval development in the mud crab Rhithropanopeus harrisii. I n “Fourth European Marine Biology Symposium” (D. J . Crisp, ed.), pp. 21 1-220. Cambridge University Press, Cambridge. Costlow, J . D. J r , Bookhout, C. G. and Monroe, R. (1960).The effect of salinity and temperature on larval development of Sesarma cinereum (Bosc) reared in the laboratory. Biological Bulletin. Marine Biological Laboratory, Woods Hole, Mass. 188,183-202. Costlow, J . D. Jr, Bookhout, C. G. and Monroe, R. (1962). Salinity temperature effects on the larval development of the crab, Panopeus herbstii Milne-Edwards, reared in the laboratory. Physiological Zoology 35,79-93. Costlow, J . D. Jr, Bookhout, C. G. and Monroe, R. (1966). Studies on the larval development of the crab, Rhithropanopeus harrisii (Gould). I. The effect of salinity and temperature on larval development. Physiological Zoology 39, 81-100. Crisp, D. J. (1956). A substance promoting hatching and liberation of young in cirripedes. Nature, London 28, 119-132. Crisp, D.J. and Costlow, J. D. (1963).The tolerance ofdeveloping cirripede embryos to salinity and temperature. Oikos 14,22-34. Crisp, D. J. and Ritz, D. A. (1967). Changes in temperature tolerance of Balanus during its life cycle. Helgolander wissenschaftliche balanoides Meeresuntersuchungen 15,98-1 15. Crisp, D. J. and Southward, A. J. (1961). Different types of cirral activity in barnacles. Philosophical Transactions of the Royal Society (Series B ) 243,27 1-308. Crisp, D. J. and Spencer, C. P. (1958). The control of the hatching process in barnacles. Proceedings of the Royal Society, London (Series B ) 148,278-299. Cronklin, R . E . and Krogh, A. (1938).A note on the osmotic behaviour of Eriocheir in concentrated and Mytilus in diluted seawater. Zeitschrift f u r vergleichende Physiologie 26,239-241. Dakin, W. J. and Edmonds, E. (1931).The regulation ofthe salt content of the blood of aquatic animals, and the problem of permeability of the bounding membranes of aquatic invertebrates. Australian Journal of Experimental Biology and Medical Science 8, 169-187. Daniel, M. J . and Boyden, C. R. (1975). Diurnal variations in physico-chemical conditions within intertidal rockpools. Field Studies 4 ( 2 ) , 161-176.

246

J . DAVENPORT

Davenport, J. (1972a). Salinity tolerance and preference in the porcelain crabs Porcellana platycheles and Porcellana lorqicornis. Marine Behaviour and Physiology 1, 123-138. Davenport, J. (1972b).Volume changes shown by some littoral anomuran crustacea. Journal of the Marine Biological Association of the United Kingdom 52, 863-877. Davenport, J. (1972~). Effects of size upon salinity tolerance and volume regulation in the hermit crab Pagurus bernhardus. Marine Biology 17, 222-227. Davenport, J. (1972d). Study of the importance of the soft abdomen of the hermit crab Pagurus bernhardus in minimising the mechanical effects of osmotic uptake of water. Marine Biology 17, 304-307. Davenport, J. (1976). A comparative study of the behaviour of some balanomorph barnacles exposed to fluctuating seawater concentrations. Journal of the Marine Biological Association of the United Kingdom 56, 889-907. Davenport, J. (1977). A study of the effects of copper applied continuously and discontinuously to specimens of Mytilus edulis (L.) exposed to steady and fluctuating salinity levels. Journal of the Marine Biological Association of the United Kingdom 57, 63-74. Davenport, J. (1979a).The isolation response of mussels (Mytilus edulis L.) exposed to falling sea-water concentrations. Journal of the Marine Biological Association of the United Kingdom 59, 123-132. Davenport, J. (1979b). Is Mytilus edulis a short term osmoregulator? Comparative Biochemistry and Physiology MA, 91-95. Cold resistancein GamrnarusduebeniLiljeborg. Astarte 12,21Davenport, J. (1979~). 26. Davenport, J. and Cawthorne, D. F. (1978). An apparatus for the laboratory simulation of environmental fluctuations in temperature. salinity and oxygen tension. Laboratory Practice 27, 281-284. Davenport, J. and Fletcher, J. 8. (1978).The effects of simulated estuarine mantle cavity conditions upon the activity of the frontal gill cilia of Mytilus edulis. Journal of the Marine Biological Association of the United Kingdom 58, 67 1-681. Davenport, J. and Manley, A. (1978).The detection of heightened sea-water copper concentrations by the mussel Mytilus edulis. Journal of the Marine Biological Association of the United Kingdom 58, 843-850. Davenport, J. and Vahl, 0. (1979). Responses of the fish Blennius pholis to fluctuating salinities. Marine Ecology Progress Series 1, 109-1 13. Davenport, J . , Gruffydd, L1. D. and Beaumont, A. R . (1975). An apparatus to supply water of fluctuating salinity and its use in a study of the salinity tolerances of larvae of the scallop Pecten maximus L. Journal of the Marine Biological Association of the United Kingdom 55, 391-409. Davenport, J., Vahl, 0. and Lonning, S. (1979). Cold resistance in the eggs of the capelin Mallotus villosus. Journal of the Marine Biological Association of the United Kingdom 59, 443-453. Davenport, J., Busschots, P. M. C. F. and Cawthorne, D. F. (1980).The influence of salinity upon the distribution, behaviour and oxygen uptake of the hermit crab Pagurus bernhardus L. Journal of the Marine Biological Association of the United Kingdom 60, 127-134. Dehnel, P. A. (1962). Aspects of osmoregulation in two species of intertidal crabs. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass 122, 208-227. Diaz, R . J. (1973). Effects of brief temperature increases on larvae of the American

ENVIRONMENTAL SIMULATION EXPERIMENTS

247

oyster (CrasFostreavirginica). Journal of the Fisheries Research Board of Canada

30,991-993. Diaz, R . J . (1975). Effects of brief temperature increases on larvae of Mu.l%nia lateralis (Mollusca : Bivalvia). Journal of the Fisheries Research Board of Canada 32, 187Ck-1872.

Dickson, D. M. J. (1978). The effects of fluctuating salinity on the tissue sap of Fucus serratus. M.Sc. Thesis, University of Wales. Djangmah, J. S., Shumway, S. E. and Davenport, J. (1979). Effects of fluctuating salinity on the behaviour of the West African blood clam, Anadara senilis and on the osmotic pressure and ionic concentrations of the haemolymph. Marine Biology 30, 209-213. Edney, E. B. (1964). Acclimation to temperature in terrestrial isopods. I . Lethal temperatures. Physiological Zoology 37, 364-377. Feldmeth, C. R., Stone, E. A. and Brown, J. H . (1974). An increased scope for thermal tolerance upon acclimating pupfish (Cyprinodun) to cycling temperature. Journal of Comparative Physiology 89, 3 9 4 4 . Findley, A . M. and Stickle, W. B. (1978). Effects of salinity fluctuation on the haemolymph composition of the blue crab Callinectes sapidus. Marine Biology46, 9-15.

Findley, A. M., Belisle, B. W. and Stickle, W. B. (1978). Effects of salinity fluctuations on the respiration rate of the southern oyster drill Thais haemastoma and the blue crab Callinectes sapidus. Marine Biology 49, 59-67. Florkin, M. (1962). La regulation isosmotique intracellulaire chez les invertkbrbs marins euryhalins. Bulletin de l’dcadimie r . de Belgique. Classe des Sciences. Florkin, M. and Schoffeniels, E. (1965). Euryhalinity and the concept of physiological radiation. (K. A. Munday, ed.), pp. W 0 . I n “Studies in (ktmparative Biochemistry”. Pergamon Press, Oxford. Florkin, M. and Schoffeniels (1969). “Molecular Approaches to Ecology”, 203 pp. Academic Press, London and New York. Forrester, C. R. and Alderdice, D. F . (1966). Effects of salinity and temperature on embryonic development of the Pacific cod (Gadus macrocephalus). Journal of the Fisheries Research Board of Canada 23, 319-340. Foster, B. A. (1970). Response and acclimation to salinity in the adults of some balanomorph barnacles. Philosophical Transactions of the Royal Society (Series B ) 256, 377400.

Frederighi, H. (1931). Studies on the oyster drill (Urosalpinx cinerea Say). Bulletin of the Bureau of Fisheries, Washington 47, 83-1 15. Freeman, R. F. H. and Rigler, F. H. (1957). The responses ofScrobiculariaplana (Da Costa) to osmotic pressure changes. Journal of the Marine Biological Association, of the United Kingdom 36, 553-567. Fry, F. E. J. (1947). Effects of the environment on animal activity. University of Toronto Studies in Biology, Series 55 (Publicatiuns of the Ontario Fisheries Research Laboratory 68) 1-62. Fry, F. E. J. (1951). A fractionating column to provide water of various dissolved oxygen content. Canadian Journal of Technology 29 (3), 144-146. Fry, F. E. J. (1957). The lethal temperature as a tool in taxonomy. Annie biologique 33, 205-219.

Fry, F. E. J. (1967). Responses of vertebrate poikilotherms to temperature. (A. H . Rose, ed.), pp 3 7 5 4 0 9 . I n ‘Thermobiology” 653 pp. Academic Press, London and New York.

Fyhn, H. J. (1976). Holoeuryhalinity and its mechanisms in a cirripede crustacean. Balanus improvisus. Comparative Biochemistry and Physiology 53A, 19-30. Gainer, H. and Grundfest, H . (1968). Permeability of alkali metal cations in lobster muscle. A comparison of electrophysiological and osmometric analyses. Journal of General Physiology 51, 3 9 9 4 2 5 . Galtsoff, P. A,, Prycherch, H . R. and Engle, J. B. (1937). Natural history and methods of controlling the common oyster drills (lirosalpinx cinerea Say and Europleura candata Say). Circular of the Bureau of Sports Fisheries and Wildlife 25, 1-24. Ganning, B. (1970).Population dynamics and salinity tolerance of Hyodesia fusca (Lohman) (Acarina, Sarcoptiformes) from brackish water rock pools, with notes on the microenvironment inside Enteromorpha tubes. Oecologia, Berlin 5, 127-137. Ganning, B. (1971). Studies on chemical, physical and biological conditions in Swedish rockpool ecosystems. Ophelia 2 ( l ) , 51-105. Ganning, B. and Wulff, F. (1969).The effects of bird droppings on the chemical and biological dynamics in brackish water rock pools. Oikos 20, 274-286. Gerard, J . F. and Gilles, R . (1971).Osmoregulation and amino acid efflux in the estuarine crab Callinectes sapidus. In “ Second National Coastal and Shallow Water Research Conference” (D. S. Gosline, ed.), Abstract volume p. 82. University Press, University of Southern California, Los Angeles. Gilles, R . ( 1972). Osmoregulation in three molluscs: Acanthochitona discrepans, Glycymeris g ~ y ~ y m e rand i s Mytilus edulis. Biological Bulletin, Marine Biological Laboratory, Woods Hole 142, 25-35. Gilles, R . (1975). Mechanisms of ion and osmoregulation. I n “Marine Ecology”. (0. Kinne, ed.), Vol. 11, Part 1, pp. 259-347. Wiley Interscience, New York. Gilles, R. (1979). “Mechanisms of Osmoregulation in Animals.” (R. Gilles, ed.). 667 pp. Wiley Interscience, New York. Grainger, J. N. R. (1956). Effects of changes of temperature on the respiration of certain crustacea. Nature, London 178, 93Ck931. Grainger, J . N. R. (1958). First stages in the adaptation of poikilotherms t o temperature change. In “Physiological Adaptation. A symposium . . . held in Woods Hole, Mass., Sept. 5 4 , 1957” (C. L. Prosser, ed.), pp. 7 S 9 1 . Society of General Physiologists, Washington D.C., U.8.A.. Grainger, J. N. R . (1959). The effect of constant and varying temperatures on the developing eggs of Rana temporaria. Zoologische Anzeiger 163, 267. Gray, J. (1923). Mechanism of ciliary movement. 111. Effect of temperature. Proceedings of the Royal Society ( B )95, 6-15. Gray, J . (1924). Mechanism of ciliary movement. IV. Relation of ciliary activity t o oxygen consumption. Proceedings of the Royal Society ( B )96, 95-1 14. Gray, J. (1929). The growth of fish. 111. The effect of temperature on the development of the eggs of Salmo fario. Journal of Experimental Biology 6, 125-1 30. Gross, W. J. (1957). A behavioural mechanism for osmotic regulation in a semi terrestrial crab. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 115, 114-125. Gruffydd, L1. D. and Beaumont, A. R . (1970). Determination of the optimum concentration of eggs and spermatozoa for the production of normal larvae in Pecten maximus (Mollusca, Lamellibranchia). Helgolander wissenschaftliche Meeresuntersuchungen 20, 4 8 6 4 9 7 .

ENVIRONMENTAL SIMIJLATION EXPERIMENTS

249

Hand, A. C. and Stickle, W. B. (1977). Effects of tidal fluctuations of salinity or pericardial fluid composition of the American oyster Crassostrea virginica. Marine Biology 42,25S27 1. Haskin, H. H. (1964). The distribution of oyster larvae. Proceedings of the Symposium of Experimental Marine Ecology. Occasional Publications No. 2, pp. 76-80. Graduate School of Oceanography, University of Rhode Island. Heath, W. G. (1963). Thermoperiodism in sea-run cutthroat trout (Salmo clarki clarki). Science 142,4 8 6 4 8 8 . Heinle, D. R. (1969). Temperature and Zooplankton. Chesapeake Science 10, 18G209. Hidu, H . and Haskin, H. H . (1978). Swimming speeds of oyster larvae Crassostrea virginica in different salinities and temperatures. Estuaries 1, 252-255. Hillman, N. S. (1964).Studies on the distribution and abundance of decapod larvae in Narragansett Bay, Rhode Island with a consideration of morphology and mortality. MAC. Thesis, University of Rhode Island. Hochachka, P. W. and Lewis, J. K. (1970).Enzyme variants in thermal acclimation: trout liver synthetases. Journal of Biological Chemistry 245,6567-6573. Hochachka, P. W. and Somero, G. N. (1968). The adaptation of enzymes to temperature. Comparative Biochemistry and Physiology 27, 659-668. Hokanson, K . E. F., Kleiner, C. F. and Thorslund, T. W. (1977).Effects of constant temperatures and die1 temperature fluctuations on specific growth and mortality rates and yield of juvenile rainbow trout, Salmo gairdneri. Journal of the Fisheries Research Board of Canada 34, 63-48, Holliday, F. G. T. (1971).I n “Marine Ecology” (0.Kinne, ed.), Vol. I, Part 2. pp. 997-1083. Wiley Interscience, New York. Hoss, D. E., Coston, L. C., Baptist, J. P. and Engle, D. W. (1975). Effects of temperature, copper and chlorine during simulated entrainment in powerplant condensor cooling systems, pp. 519-527. Vienna, International Atomic Energy Agency (Ref. IAEA-SM-189/19). House, C. R. (1963). Osmotic regulation in the brackish water teleost Blennius pholis. Journal of Experimental Biology 40, 87-104. Houston, A. H. (1957). Responses of juvenile chum, pink and coho salmon to sharp seawater gradients Canadian Journal of Zoology 35,37 1-383. Hoyaux, J.,Gilles, R. and Jeuniaux, C. (1976). Osmoregulation in molluscs of the intertidal zone. Comparative Biochemistry and Physiology 53A,361-365. Hubbs, C. (1964). Effects of thermal fluctuations on the relative survival of greenthroat darter young from stenothermal and eurythermal waters. Ecology 45 (2), 37G379. Inman, C. B. E. and Lockwood, A. P. M. (1977).Some effects of methylmercury and lindane on sodium regulation in the amphipod Gammarus duebeni during changes in the salinity of its medium. Comparative Biochemistry and Physiology 58C, 67-75. Jansson, B. 0. (1962).Salinity resistance and salinity preference of two oligochaetes Akerdrilus monospermatecus Knoller and Marionina preclittochaeta N.sp. from the intertidal fauna of marine sandy beaches. Oikos 13,293-305. Jones, L.H . , Jones, N . V. and Radlett, A. J . (1976). Some effects of salinity on the toxicity of copper to the polychaete Nereis diversicolor. Estuarine and Coastal Marine Science 4,107-111. Jones, M. B. (1975).Effects of copper on survival and osmoregulation in marine and brackish water isopods (Crustacea).I n “Proceedings of the 9th European Marine

250

J . DAVENPORT

Biological Symposium” (H. Barnes, ed.), pp. 419-431. Aberdeen University Press, Scotland. Keys, A. B. (1933).The mechanism of adaptation to varying salinity in the common eel and the general problem of osmotic regulation. Proceedings of the Royal A‘ociety (Series B ) 112, 184-199. Khan, M. F. (1965). The effect of constant and varying temperatures on the development of Acanthocyclops viridius (Jurine). Proceedings of the Royal Irish Academy 64B,117-130. Khlebovich, V. V. (1969).Aspects of animal evolution related to critical salinity and internal state. Marine Biology 2, 338-345. Kinne, 0. (1964a). Non genetic adaptation to temperature and salinity. Helgolander wissenschaftliche Meeresuntersuchungen 9, 433-457. Kinne, 0. (1964b). The effect,s of temperature and salinity on marine and brackish water animals. 11. Salinity and temperaturesalinity combinations. Oceanography and Marine Biology Annual Review 2, 281-339. Kinne, 0. (1970).I n “Marine Ecology” (0.Kinne, ed.), Vol 1, Part 1, pp. 407-514. Wiley Interscience, New York. Kinne, 0. (1971). I n “Marine Ecology” (0.Kinne, ed.), Vol 1 , P a r t f , pp. 821-995. Wiley Interscience, New York. Kleiber, M. (1961).“The Fire of Life, An Introduction to Animal Energetics.” Wiley Interscience, New York. Knight, D. W. and West, J. R. (1975).Preliminary report on conditions in tJheConwy Estuary Dept of Civil Engineering. University of Birmingham. Koizumi, T. (1935). Studies on the exchange and the equilibrium of water and electrolytes in the holothurian Cawdina chilensis-IIIa. On the velocity of permeation of K’, Na’, C a + + and Mg++ through the isolated body wall of Caudina. Tohoku University, Tohoku daigaku Scienti$c Reports Series 4 Biology 10, 26S275. Korringa, P. (1952).Recent advances in oyster biology. Quarterly Review of Biology 27, 266-308, 339-365. Krogh, A. (1939). “Osmotic Regulation in Aquatic Animals”, 242 pp. Cambridge Univemity Press, London. Lange, R. (1963).The osmotic function of amino acids and taurine in the mussel, Mytilus edulis. Comparative Biochemistry and Physiology 10, 173-179. Lange, R. (1970). Isosmotic intracellular regulation and euryhalinity in marine bivalves. Journal of Experimental Marine Biology and Ecology 5 , 17(r179. Lange, R . and Mostad, A. (1967). Cell volume regulation in osmotically adjusting marine animals. Journal of Experimental Marine Biology and Ecology 1, 209-2 19. Lewis, J. B. (1960).The fauna of rocky shores of Barbados, West Indies. Canadian Journal of ZooZogy 38, 391435. Lewis, J . B. (1963). Environmental and tissue temperatures of some tropical intertidal marine animals. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 124, 277-284. Ling, S. W. (1969). Methods of rearing andcylturing Macrobrachium rosenbergii (de Man). F A 0 Fisheries Reports’No. 57, Vol. 3, pp. 607-619. Livingstone, D. R., Widdows, J. and’-Fieth, P. (1979). Aspects of nitrogen metabolism of the common mussel MytiZus edulis: adaptation to abrupt and fluctuating changes in salinity. Marine Biology 53, 41-55.

ENVIRONMENTAL SIMl’LA‘lYOK E X P E R I M E N T S

25 1

Lockwood, A. P . M. (1962).The osmoregulation of Crustacea. Biological Reviews 37, 257-305. Loosanoff, V. L. (1945a). Precocious gonad development in oysters induced in midwinter by higher temperatures. Science 102, 124-125. Loosanoff, V. L. (1945b). Effect of sea water of reduced salinities upon the starfish Asterias forbesi of Long Island Sound. Transactions of the Connecticut Academy o j Arts and Sciences 36, 813-833. Loosanoff, V. L. (1949).Vertical distribution of oyster larvae ofdifferent ages during the tidal cycle. Abstract, Anatomical Record 105, 591. Loosanoff, V. L. and Davis, H . C. (1950).Conditioning V . mercenaria for spawning in winter and breeding its larvae in the laboratory. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 98, 6(f65. Lough, R. G. (1975). A re-evaluation of the combined effects of temperature and salinity on survival and growth of bivalve larvae using response surface techniques. Fisheries Bulletin of the United States 73, 86-94. Lowe, C. H. and Vance, V. J. (1955).Acclimation of the critical thermal maximum of the reptile Urosaurus ornatus. Science 122, 73-74. Lucas, J. S. and Costlow, J. D. J r (1979).Effects of various temperature cycles on the larval development of the gastropod mollusc, Crepidula fornicata. Marina Biology 51, 11 1-1 17. Lynch, M. P . and Wood, L. (1966). Effects of environmental salinity on free amino acids of Crassostrea virginica Gamelin. Comparative Biochemistry and Physiology 19, 783-790. MacGinitie, G. E. and MacGinitie, N. (1968).“Natural History of Marine Animals”, 523 pp. McGraw-Hill, New York. McLean, R . I . (1973). Chlorine and temperature stress in esharine invertebrates. Journal of the Water Pollution Control Federation 45, 837-841. McLeese, D. W. (1956). Effects of temperature, salinity and oxygen on the survival of the American lobster. Journal of the Fisheries Research Board of Can,ada 13, 247-272. McLusky, D. A. (1970).Salinity preference in Corophium volutator. Journal of the Marine Biological Association of the United Kingdom 50, 747-752. Manley, A. R. (1980).An apparatus for the preparation of varying concentrations of chemicals for toxicity tests with aquatic organisms. Water Research 14, 1023-1 027. Manley, A. R. and Davenport, J. (1979). The behavioural response of some marine bivalves to heightened seawater copper concentrations. Bulletin, of Environmental Contamination and Toxicology 22, 739-744. Manzi, J. J. (1970).Combined effects of salinity and temperature on the feeding and survival rates of Eupleura candata (Say) and Urosa,lpinx cinerea (Say) (Prosobranchia :Muricidae). Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 138, 3 5 4 6 . Matthiessen, G. C. (1960).Observations on the ecology of the soft clam Mya arenaria in a salt pond. Limnology and Oceanography 5, 291-300. Middaugh, D. P., Dean, J. M., Domey, R. G. and Floyd, G. (1978).Effect of thermal stress and total residual chlorination on early life stages of the mummichog Fundulus heterolitus. Marine Biology 46, 1-8. Mihursky, J. A. and Kennedy, U. S . (1967). Water temperature criteria to protect aquatic life. Special Publications of the American Fisheries Society 4. 2&32.

252

J . DAVENPORT

Milne, A. (1940).Some ecological aspects of the intertidal area of the estuary of the Aberdeenshire Dee. Transactions ofthe Royal Society of Edinburgh 60, Part 1, No. 4, 107-139. Mobius, K . and Heincke, F. (1883). Die fische der Ostsee. Jahresbericht (Bericht) Cmmision zur Wissenschnftlichen Untersuchung der Deutschen Meere in Riel 4 , 48-296. Moon, T. and Hochachka, P. W. (1971). Temperature and enzyme activity in poikilotherms; isocitrate dehydrogenase in rainbow trout. Biochemistry Journal 123, 695-706. Morgan, R. P. and Stross, R. G. (1969).Destruction ofphytoplankton in t,he cooling water supply of a steam electric station. Chesapeake Science 10, 165-171. Muchmore, D. and Epel, D. (1973). The effects of chlorination waste water on fertilisation in some marine invertebrates. Marine Biology 19, 93-95. Nassonov, D. N. and Alexandrov, V. Y . (1943) Uber die Ursachen der kolloidalen Verinderungen des Protoplasmas und der erhiihten Affinitat des Letzteren drn Farbstoffen gegeniiber unter dem Einfluss schadigender Agentien. Actu Zoologica. Stockholm 24, 189-230. Naylor, E. (1965). Effects of heated effluents upon marine and estuarine organisms. Advances in Marine Biology 3, 63-103. Nelson, J. (1912). Report of the Biological Department of the New Jersey Agricultural Experiment Station for the year 1911. N. J. Agric. Expt. Sta., New Brunswick, N.J. Newell, R. C. (1969). Effect of fluctuations in temperature on the metabolism of intertidal invertebrates. American Zoologist 9, 293-307. Newell, R . C. and Branch, G. M. (1980). The influence of temperature on the maintenance of metabolic energy balance in marine invertebrates. Advumes in Marine Biology 17, 329-396. Newman, W. A. (1967). On physiology and behaviour of estuarine barnacles. In “Proceedings of the Symposium on Crustacea. Ernakulam, 1965”, pp. 1038-1066. Marine Biological Association of India, India. Ong, K . and Costlow, J . D. Jr (1970).The effect of salinity and temperature on the larval development of the stone crab, Menippe mercenaria (Say), reared in the laboratory. Chesapeake Science 11, 16-29. Orr, A. P. and Moorhouse, F. W. (1933). Great Barrier Expedition 1928-29, Scientific Reports British Museum (Natural History). Vol. 11. pp. 87-1 10. Otto, R. G. (1974). The effects of acclimation to cyclic thermal regimes on heat tolerance of the western mosquitofish. Transactions of the Am.erican Fisheries Society 103, 331-335. Peterson, R. H., Spinney, H . C. E. and Sreedharan, A. (1977). Development, of Atlantic salmon (Salmo salar) eggs and alevins under varied temperature regimes. Journal of the Fisheries Research Board of Canada 34, 3&43. Phillips, D. J. H. (1977a).The use of biological indicator organisms to monitor trare metal pollution in marine and estuarine environments-a review. Environmental Pollution 13, 281-317. Phillips, D. J. H . (1977b).Effects of salinity on the net uptake of zinc by the common mussel Mytilus edulis. Marine Biology 41, 79-88. Pierce, 6. K . J r (1970) The water balance of Modiolus (Mollusca: Bivalvia: Mytilidae) : osmotic concentrations in changing salinities. Cmkpurative Biochemistry and Physiology 36, 521-533.

ENVIROSMENTAL SIMULATION EXPERIMENTS

253

Pierce, S. K . Jr (1971).Volume regulation and valve movements by marine mussels. Comparative Biochemistry and Physiology 39A, 103-1 17. Pierce, S. K. and Greenberg, M. J. (1972).The nature of cellular volume regulation in marine bivalves. Journal of Experimental Biology 57, 681-692. Pierce, 8. K . and Greenberg, M. J. (1973).The initiation and control of free amino acid regulation of cell volume in salinity stressed marine bivalves. Journal of Experimental Biology 59, 435-446. Potts, W. T. W. and Parry, G. (1964). “Osmotic and Ionic Regulation in Animals”. Pergamon Press, Oxford. Powers, E. B. (1920). The variation of the condition of sea-water, especially the hydrogen ion concentration, and its relation to marine organisms. Publications of the Puget Sound Marine Biological Station 2, 369-385. Precht, H . (1958). Concepts of the temperature adaptation of unchanging reaction systems of cold blooded animals. I n “Physiological Adaptation” (C. L. Prosser, ed.), pp. 50-78. Ronald Press, New York. Precht, H., Christophersen, J. and Hensel, H. (1955). “Temperatur und Leben”. Springer-Verlag, Berlin. Pritchard, D. W. (1953). Distribution of oyster larvae in relation to hydrographic conditions. Proceedings of the Guyand Caribbean Fisheries Institute, 1952, pp. 123132. Prosser, C. L. (1973). “Comparative Animal Physiology”, 966 pp. W. B. Saunders Co., Philadelphia. Pyefinch, K. A. (1943).The intertidal ecology of Bardsey Island, North Wales, with special reference to the recolonization of rock surfaces, and the rock pool environment. Journal of Animal Ecology 12, 82-108. Qasim, S. Z. (1957).The biology of Blenniuspholis (L.) (Teleostei).Proceedinys of t h P Zoological Society of London 128, 161-208. Read, K . R. H. (1969).Thermal tolerance of the bivalve mollusc Modio1.u.s modiolns L. American Zoologist 9, 279-282. Regnault, M. and Costlow, J. D. (1970). Influence de la temperature et de la salinitG sur le developpement larvaire de Crangon septemspinosa Say (Decapoda, Caridea), Vie et Milieu (Sir A . , Biol. mar.) 21, 453-466. Ritz, D. A. (1980). Tolerance of intertidal amphipods to fluctuating conditions of salinity, oxygen and copper. Journal of the Marine Biological Association of the United Kingdom 60, 489-498. Robertson, J. D. (1953).Further studies on ionic regulation in marine invertebrates. Journal of Experimental Biology 26, 182-200. Robertson, J. D. (1960). Osmotic and ionic regulation. I n “The Physiology of Crustacea” (T. H. Waterman, ed.), Vol. 1. pp. 317-340. Academic Press, London and New York. Rosenberg, R. and Costlow, J. D. Jr (1976). Synergistic effects of cadmium and salinity combined with constant and cycling temperatures on the larval development of two estuarine crab species. Marine Biology 38, 291-303. Sanders, H. L., Mangelsdord, P. C., J r and Hampson, G. R. (1965). Salinity and faunal distribution in the Pocasset River, Massachusetts. Limnoloyy and Oceanography 10, Suppl. R216-R229. Sandoz, M. and Rogers, R. (1944).The effect of environmental factors on hatching. moulting and survival of zoea larvae of the blue crab, Callinectes sapidus Rathbun. Ecology 25, 216-228.

254

J . DAVENPORT

Sastry, A. N. (1978). Physiological adaptation of Cancer irr0ratw.s larvae to cyclic temperatures. Zn “Physiology and Behaviour of Marine Organisms”, Proceedings of the 12th European Symposium on Marine Biology, Stirling, Scotland. September 1977 (D. S. McLusky and A. J. Berry, eds), pp. 57-65. Pergamon Press, Oxford. Schlieper, C. (1929). Ueber die Einwirkung niederer Salzkonzentrationen auf marine ‘Organismen. Zeitschrift f u r vergleichende Physiologie 9,478-514. Schoffeniels, E. (1976). Adaptations with respect to salinity. Biochemical Society Symposiw.m 41, 179-204. Schoffeniels, E. and Gilles, R. (1970). Osmoregulation in aquatic arthropods. I n “Chemical Zoology” (M. Florkin and B. T. Scheer, eds), Vol. V, Part A , pp. 255-286. Academic Press, London and New York. Schoffeniels, E. and Gilles, R. (1972).Ion regulation and osmoregulation in Mollusca. I n “Chemical Zoology” (M. Florkin and B. T. Scheer, eds),Vol. VII, pp. 393-420. Academic Press, London and New York. Schwabe, E. (1933). Uber die Osmoregulation verschiedener Krebse (Malocostracen). Zeitschrift f u r vergleichende Physiologie 19, 183-236. Segal, E. and Dehnel, P . A. (1962). Osmotic behaviour in an intertidal limpet. Acmaea limatula. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 122,417430. Shelford, V.E. and Powers, E. B. (1915). An experimental study of the movements of herring and other marine fishes. Biological Bulletin, Marine Biological Laboratory, Woods Hole, Mass. 28,315-334. Sherberger, F.F., Benfield, E. F., Dickson, K . L. and Cairns, J . Jr (1977).Effects of thermal shocks on drifting aquatic insects: a laboratory simulation. Journal of the Fisheries Research Board of Canada 34,52S536. Shumway, S. E . (1977a). Effect of salinity fluctuation on the osmotic pressure and Na+ , Ca2+and Mgz+ ion concentrations in the haemolymph of bivalve molluscs. Marine Biology 41,153-177. Shumway, 6. E. (1977b). The effects of fluctuating salinities on four species of asteroid echinoderms. Comparative Biochemistry and Physiology, 58A, 177-179. Shumway, S. E . (1977~). The effect offluctuatingsalinity on the tissue water content of eight species of bivalve molluscs. Journal of Comparative Physiology 116, 26S285. Shumway, S. E. (1978a).Activity and respiration in the anemone, Metridium senile (L.)exposed to salinity fluctuations. Journal of Experimental Marine Biology and Ecology 33,85-92. Shumway, 8. E. (1978b). Respiration, pumping activity and heart rate in Ciona intestinalis exposed to fluctuating salinities. Marine Biology 48, 235-242. Shumway, S. E. (1978~).Osmotic balance and respiration in the hermit crab, Pagurus bernhardus, exposed to fluctuating salinities. Journal of the Marine Biological Association of the United Kingdom 58,869-876. Shumway, S.E . (1979).The effects of fluctuating salinity on respiration in gastropod molluscs. Comparative Biochemistry and Physiology MA, 274-283. Shumway, S. E. and Davenport, J. (1977). Some aspects of the physiology of Arenicola marina (Polychaeta) exposed to fluctuating salinities. Jownal of the Marine Biological Association of United Kingdom 57,907-924. Shumway, S.E. and Youngson, A. (1979). The effects of fluctuating salinity on the physiology of Modiolus demissus (Dillwyn). Journal of Experimental Marine Biology and Ecology 40,167-181.

ENVIRONMEXTAL SIMULATION EXPERIMENTS

255

Shumway, S. E., Gabbott, P . A. and Youngson, A. (1977).The effect of fluctuating salinity on the concentrations of free amino acids and ninhydrin positive substances in the adductor muscles ofeight species of bivalve molluscs. Journal of Experimental Marine Biology and Ecology 29, 131-150. Sleigh, M.A. (1962).“The Biology of Cilia and Flagella”. 242 pp. Pergamon Press, Oxford. Somero, G. N. (1969). Enzymatic mechanisms of temperature compensation: immediate and evolutionary effects of temperature on enzymes of aquatic poikilotherms. American Naturalist 103,517-530, Southward, A. J. (1958). Note on the temperature tolerance of some intertidal animals in relation t o environmental temperatures and geographical distribution. Journal of the Marine Biological Association of the United Kingdom 37, 49-66. Spaargaren, D. H. (1974). A study on the adaptation of marine organisms t o changing salinities with special reference t o the shore crab Carcinus muenas ( L .). Comparative Biochemistry and Physiology 47A,499-512. Spaargaren, D. H. (1978).A comparison of the blood osmotic composition of various marine and brackish water animals. Comparative Biochemistry and Physiology M A , 327-333. Stephenson, T. A , , Zoond, A. and Eyre, J. (1934).The liberation and utilization of oxygen by the population of rock pools. Journal of Experimental Biology 11, 162-1 7 2. Stickle, W. B. and Ahokas, R. (1974).The effects of tidal fluctuation of salinity on the perivisceral fluid composition of several echinoderms. Comparative Biochemistry and Physiology 47A,469-476. Stickle, W.B. and Ahokas, R . (1975).The effects of tidal fluctuation of salinity on the haemolymph composition of several molluscs. Comparative Biochemistry and Physiology 50A,291-296. Stickle, W.B. and Denoux, G. J. (1976).Effects of in situ tidal salinity fluctuations on osmotic and ionic composition of body fluid in south-eastern Alaska rocky intertidal fauna. Marine Biology 37, 125-135. Stickle, W. B. and Howey, T. W. (1975). Effects of tidal fluctuations of salinity haemolymph composition of the southern oyster drill Thais haemastoma. Marine Biology 33,309-322. Stomgren, T. ( 1975).Linear measurements ofgrowth of shells using laser diffraction. Limnology and Oceanography 20, 845-848. Tamura, T. (1944).The effect of changes in environment on fishes. 7 . The effect of sudden change of water temperature on heat and cold tolerance. Bulletin of the Japanese Society for 8cientijc Fisheries 12,204-208. Threde, H. (1965). Vergleichende experimentelle Unt,ersuchungen tiher die zelluliire Gefrierrcsistenz mariner Muschdn. Kielvr Meerr.sforsch?tnym~21. 153 166. Thorp, J. H . and Hoss, D. E. (1975).Effects of salinity and cyclic temperature on survival of two sympatric species of grass shrimp (Palaemonetes), and their relationship to natural distributions. Journal of Experimental Marine Biology and Ecology 18, 19-28. Trueman, E . R. (1967). Activity and heart rate of bivalve molluscs in their natural habitat. Nature, London 214,832-833. Tsukuda, H.(1960).Temperature adaptation in fishes. 4. Change in the heat and cold tolerances of the guppy in the process of temperature acclimatisation. Journal of the Institute of Polytechnics, Osaka City University (Ser. D ) , 11, 43-54.

256

.I. DAVENPORT

Tsukuda, H. and Ohsawa, W. (1958). The heat and cold coma temparatures in Lebistes reticulatus as determined by raising and lowering water temperatures a t constant rates. Journal of the Institute of Polytechnics, Osaka City University (Ser. D ) 9, 6+76. Tucker, L. E. (1970a).Effects of external salinity on Scutus breviculus (Gastropoda, Prosobranchia)-I. Body weight and blood composition. Comparative Biochemistry and Physiology 36,301-319. Tucker, L. E. (1970b). Effects of external salinity on Scutus breviculus (Gastropoda, Prosobranchiaf-11. Nerve conduction. Comparative Biochemistry and Physiology 37,467480. Turner, H . J. J r , Reynolds, D. M. and Redfield, A. C. (1948).Chlorine and sodium pentachlorophenate as fouling preventives in sea water conduits. Industrial and Engineering Chemistry 40, 450-453. Ushakov, B. P. (1968). Cellular resistance adaptation to temperature and thermostability of somatic cells with special reference to marine animals. Marine Biology 1, 153-160. Vahl, 0.and Davenport, J. (1979). Apparent specific dynamic action of food in the fish Blennius pholis. Marine Ecology Progress Series 1, 109-113. Vernberg, F. J. (1972). Dissolved gases-animals. f n “Marine Ecology” (0.Kinne, ed.), Vol. I, Part 3, pp. 1491-1526. Wiley-Interscience, New York. Waugh, G. D. (1964). Observations on the effects of chlorine on the larvae of oysters (Ostrea edulis L.) and barnacles (Elminiusmodestus Darwin). Annals of Applied Biology 54, 423440. Webb, D. A. (1940). Osmotic and ionic regulation in Carcinus maenas. Proceedings of the Royal Society (Series B ) 129, 107-136. Wells, G. P. (1949a). Respiratory movements of Arenicola marina L.: intermittent irrigation of the tube, and intermittent aerial respiration. Journal of the Marine Biological Association of the United Kingdom 28, 447464. Wells, G. P. (1949b). The behaviour of Arenicola marina L. in sand, and the rate of spontaneous activity cycles. Journal of the Marine Biological Association of the United Kingdom 28, 465478. Wells, G. P. and Ledingham, I. C. (1940). Simple methods for exposing isolated organs to sudden and gradual changes in the chemical environment. Journal of Experimental Biology 17, 35 1-352. Widdows, J. (1976). Physiological adaptation of Mytilus edulis to cyclic temperatures. Journal of Comparative Physiology 105, 115-128. Widdows, J. and Bayne, B. L. (1971). Temperature acclimation of Mytilus edulis with reference to its energy budget. Journal of the Marine Biological Association of the United Kingdom 51, 827-843. Wood, L. and Hargis, W. J. (1971). Transport of bivalve larvae in a tidal estuary, I n “Fourth European Marine Biology Symposium, Bangor, Wales, 1969” (D. J. Crisp, ed.), pp. 2 9 4 4 . Cambridge University Press. Zachary, A. and Haven, D. S. (1973). Survival and activity of the oyster drill Urosalpinx cinerea under conditions of fluctuating salinity. Marine Biology 22 4652. Zwaan, A. De, Kluytmans, J. H. F. and Zandee, D. I. (1976). Facultative anaerobiosis in molluscs. Biochemical Society Symposium 41. 133-168.

Adv . Mar . Bid.,Vol . 19. 1982. pp . 257-355

THE POPULATION BIOLOGY OF BLUE WHITING IN THE NORTH ATLANTIC R . S. BAILEY Department of Agriculture and Fisheries for Scotland. Marine Laboratory A b e d een. Scotland

.

I . Introduction . . . . . . . . . . . . . . . . I1. The Life History . . . . . . . . . . . . . A . The planktonic stages . . . . . . . . . . B . The immature phase . . . . . . . . . . . C. The adult phase . . . . . . . . . . . . 111. The Ecological Role of Blue Whiting . . . . . . . . . . . . . . . . . . A . Food and feeding B . Predators . . . . . . . . . . . . . C . Parasites and diseases . . . . . . . . . . D . Competition . . . . . . . . . . . . . . . . . . . . . . . . IV . Population Dynamics A . Introduction . . . . . . . . . . . . . B . Age determination . . . . . . . . . . . C. Growth . . . . . . . . . . . . . . . . D . Mortality . . . . . . . . . . . . . E . Fecundity . . . . . . . . . . . . . F . Condition . . . . . . . . . . . . . G . Stock discrimination . . . . . . . . . . . v . Distribution . . . . . . . . . . . . . . . .

258 . . . . . . . . . 259

.

.

. . . . . . . .

VII . VIII . IX. X. XI.

.

.

259

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . . . .

. . . . . . . . . . . . . . . .

VI .

.

. . . . . 266

.

.

.

. . . . .

. . . . .

. . . . .

. . .

.

.

.

.

A. Eggs and larvae B . Immatures . . . . . . . . . . . . . . . . C. Adult distribution and migrations . . . . . . . . . . D . Ecological correlates . . . . . . . . . . . . . . E . Depth distribution . . . . . . . . . . . . . . Abundance and Stock Size . . . . . . . . . . . . . . A . Trends . . . . . . . . . . . . . . . . . . . B . Absolute estimates of stock size . . . . . . . . . . Exploitation . . . . . . . . . . . . . . . . . . . The Southern Blue Whiting . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .

.

.

.

.

.

.

.

.

. .

.

. .

. . . . .

266 276 276 280 283 284 286 286 286 290 297 300 302 30'2 306 306 309 314 322 323 326 327 329 334 337 340 342 342

258

R. H. BAILEY

I. INTRODUCTION It is unusual in the volumes of “Advances in Marine Biology” t o find a review of the biology of a single species of fish, and i t is therefore appropriate to explain why the blue whiting Micromesistius poutussou (Risso) should have been singled out for such treatment. Until recently, little interest was shown in this species, either by commercial fishermen or by fisheries biologists, yet in 1979 over a million tonnes of blue whiting were caught in the north-east Atlantic (Anon, 1980). The desire within the last decade to find underexploited stocks of fish has dramatically changed the level of research on this species. As a result there already exists an extensive documentation of its distribution and biology which is almost unparalleled in a fish species prior to the development of exploitation. Unfortunately, however, the reports are largely to be found in fisheries journals obscure to many non-fishery biologists, or in unpublished documents which are not generally available. Since interest in the blue whiting, which may well be the most abundant fish of commercial importance in the north-east Atlantic, is likely t o grow, it seemed apposite to document what is known about the population biology of the species in the period before exploitation became a significant factor in its biology. The blue whiting is a member of the family Gadidae, and is now usually included in a separate genus containing only the northern species, with which this review is mainly concerned, and a closelyrelated species living in the southern hemisphere (Micrmesistius australis (Norman)).The blue whiting is a relatively small fish (the largest in most samples of adults is usually 3 5 4 0 cm in length) and is easily recognized by its slim shape and mauvish-blue coloration when taken in trawls. I n life it may have a different appearance since those caught by trawl are usually almost devoid of scales. When caught in good condition, it has a more silvery appearance. Details of its anatomical characteristics are given by Svetovidov (1948) and a review of its taxonomy is provided by Bigelow and Schroeder (1955). This article reviews the biology of the northern species of blue whiting throughout its range and a brief account is given of its southern hemisphere counterpart. Particular attention is also paid to the fact that data obtained so far are from an unexploited stock and an evaluation is made of which characteristics may be due to this unexploited condition. The recent development of research on blue whiting has largely been channelled by the need to evaluate the potential of the resource

POPULATION BIOLOGY OF BLIIE WHITIN(:

259

for commercial fisheries. It was as early as the first decade of this century, however, that Johannes Schmidt (1909) demonstrated its widespread distribution to the west of the European continental shelf. That this was a resource worth investigating was supported by the results of the Continuous Plankton Recorder Survey (Henderson, 1957),which showed that the larvae of blue whiting are abundant in the north-east Atlantic to the west of the British Isles. Under pressure of growing exploitation of many traditional resources, exploratory research, both to locate fishable concentrations and to estimate stock size, began in the Soviet Union and the United Kingdom in the 1960s. Following an incidental finding that the species was abundant in mid-water in the Norwegian Sea (Mohr, 1968), a major research effort was mounted in this area during the 1970s. Research into methods of catching blue whiting has also been carried out and an evaluation has been made of the potential of the species for human consumption. The results of this research, however, are not of direct relevance to the present review and are only drawn upon insofar as they are germane to the main subject. The only previous review of the biology of blue whiting is the excellent synopsis prepared by Raitt (1968a) for F.A.O. A detailed account of the distribution of blue whiting is given in Section V of this review. By way of introduction, however, the species is widely distributed along the continental margin in the north-east Atlantic between latitudes 26" and 82"N. Smaller populations also occur in the Mediterranean and in the north-west Atlantic. In the main part of its range the majority of the adult population makes an annual migration between spawning grounds along the edge of the continental shelf west of the British Isles and the feeding areas in the Norwegian Sea (see Fig. 13). Spawning takes place in the period February-April and the planktonic young drift to nursery areas in somewhat shallower water than that occupied by the adults (see Fig. 11).Localities and sea areas mentioned in the text are shown in Fig. 1 .

11. THE LIFE HISTORY A. The planktonic stages 1. Description of embryonic and larval development Interest in the eggs and larvae of blue whiting has focused mainly on the description of the spawning distribution and estimation of the size of the spawning populations. Nevertheless, an essential

~

FIG.1 . Map showing places referred to in the tables and text.

POPULATION BIOLOGY O F BLUE WHITING

26 1

prerequisite of these studies was the description of the early development. The first adequate description of the larvae (strictly the post-larvae after absorption of the yolk sac) was provided by Schmidt (1905, 1906) while Henderson (1957) added information on the younger post-larvae down to 2.7mm in length. The eggs and embryonic development were not described until much later from the results of artificial fertilizations (Fliichter and Rosenthal, 1965; Seaton and Bailey, 1971), although earlier there had been an erroneous inference about the egg by Henderson (1957), and measurements of ripe intraovarian eggs had been made by Polonsky (1968). An admirable summary of the planktonic stages is given by Russell (1976). The fullest account of the development, including the effects temperature has on i