Hominoid Evolution and Climatic Change in Europe: Volume 1, The Evolution of Neogene Terrestrial Ecosystems in Europe

Europe has changed greatly in terms of climate and environment in the past 20 million years. Once, there were sub-tropical forests, but by the end of the Miocene, 5 million years ago, these had all gone. This unique book provides evidence for the past climatic history of Europe and the Mediterranean in relation to hominoid evolution. Many diVerent lines of evidence are brought together including studies speciWcally on past climates and the application of climate modelling, the reconstruction of past geographical events, and the eVects they had on European environments and the plants and animals living in them. Together, they form a coherent and consistent image of environmental and climatic change in Europe from 18 to 1.6 million years ago, for all those interested in mammalian and human evolution. is Director of the Institute of Paleontology, M. Crusafont, in Sabadell, Spain. He specialises in the evolution of the Neogene and Quaternary small mammalian faunas in relation to environmental changes.

JORGE AGUSTI

is a researcher in the Department of Earth Sciences at the University of Florence, working on fossil primates and carnivora and on Neogene/Quaternary biochronology.

LORENZO ROOK

is a research scientist at the Natural History Museum in London, where he works on fossil primates, taphonomic and palaeoecological issues relating to the early stages of human evolution.

PETER ANDREWS

XXXXXX

H OM I N OI D EV OLU TI ON A N D C L I M AT I C C H AN GE I N E UR O P E VOL UME 1

Th e E v ol u ti o n o f Ne og e ne Te rr e s t r i al E c os y s te ms i n Eu r op e

XXXXXX

HOM I NO I D EVO L UT I ON AN D CLI M AT I C CH AN GE I N EURO P E VO L UM E 1

The Evolution of Neogene Terrestrial Ecosystems in Europe Edited by

JORGE AGUSTI LORENZO ROOK and

PETER ANDREWS

   Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge  , United Kingdom Published in the United States by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521640978 © Cambridge University Press 1999 This book is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 1999 ISBN-13 ISBN-10

978-0-511-06619-1 eBook (NetLibrary) 0-511-06619-8 eBook (NetLibrary)

ISBN-13 978-0-521-64097-8 hardback ISBN-10 0-521-64097-0 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of s for external or third-party internet websites referred to in this book, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

C o n t en t s

List of contributors Acknowledgements: The European Science Foundation

page x xvii

1 Introduction Jorge Agustı´, Lorenzo Rook and Peter Andrews

1

PART I. Palaeogeography of the circum-Mediterranean region

7

2 Mediterranean and Paratethys palaeogeography during the Oligocene and Miocene Fred Ro¨gl 3 Pliocene tephra correlations between East African hominid localities, the Gulf of Aden, and the Arabian Sea Peter B. deMenocal and Francis H. Brown 4 Climatic perspectives for Neogene environmental reconstructions Eileen M. O’Brien and Charles R. Peters PART II. Miocene mammalian successions 5 A critical re-evaluation of the Miocene mammal units in Western Europe: dispersal events and problems of correlation Jorge Agustı´ 6 Large mammals from the Vallesian of Spain Jorge Morales, Manuel Nieto, Meike Kholer and Salvador Moya`-Sola` 7 Trends in rodent assemblages from the Aragonian (early–middle Miocene) of the Calatayud-Daroca Basin, Aragon, Spain Remmert Daams, Albert J. van der Meulen, Pablo Pelaez-Campomanes and Maria A. Alvarez-Sierra 8 The Late Miocene small mammal succession from France, with emphasis on the Rhoˆne Valley localities Pierre Mein 9 Late Miocene mammals from Central Europe Jens Lorenz Franzen and Gerhard Storch 10 An overview on the Italian Miocene land mammal faunas Lorenzo Rook, Laura Abbazzi and Burkhart Engesser 11 The Miocene large mammal succession in Greece Louis de Bonis and George D. Koufos

8

23 55 83

84

113

127 140 165 191 205

Contents

viii

12 Chronology and mammal faunas of the Miocene Sinap Formation, Turkey Juha Pekka Lunkka, Mikael Fortelius, John Kappelman and Sevket Sen 13 The Late Miocene small mammal succession in Ukraine Valentin A. Nesin and Vadim A. Topachevsky

238

PART III. Palaeoenvironments: non-mammalian evidence

273

265

14 Marine invertebrate (chieXy foraminiferal) evidence for the palaeogeography of the Oligocene–Miocene of western Eurasia, and consequences for terrestrial vertebrate migration Robert Wynn Jones 15 Palaeoclimatic implications of the energy hypothesis from Neogene corals of the Mediterranean region Brian R. Rosen 16 Contribution to the knowledge of Neogene climatic changes in western and central Europe by means of non-marine molluscs Daniela Esu 17 Sedimentary facies analysis in palaeoclimatic reconstructions. Examples from the Upper Miocene–Pliocene successions of south-central Tuscany (Italy) Marco Benvenuti, Mauro Papini and Giovanni Testa 18 Neogene vegetation changes in West European and West circum-Mediterranean areas Jean-Pierre Suc, Se´verine Fauquette, Mostefa Bessedik, Adele Bertini, Zhuo Zheng, Georges Clauzon, Danica Suballyova, Filomena Diniz, Pierre Que´zel, Najat Feddi, Martine Clet, the late Ezzedine Bessais, Naima Bachiri TaouWq, Henriette Meon and Nathalie Combourieu-Nebout

378

PART IV. Palaeoenvironments: mammalian evidence

389

19 Shrews (Mammalia, Insectivora, Soricidae) as paleoclimatic indicators in the European Neogene Jelle W. F. Reumer 20 Mammal turnover and global climate change in the late Miocene terrestrial record of the Valle`s-Penede`s basin (NE Spain) Jorge Agustı´, Lluı´s Cabrera, Miguel Garce´s and Manel Llenas

274 309

328

355

390

397

Contents

21 Palaeoenvironments of late Miocene primate localities in Macedonia, Greece Louis de Bonis, Genevieve Bouvrain and George D. Koufos 22 The paleoecology of the Pikermian Biome and the savanna myth Nikos Solounias, J. Michael Plavcan, Jay Quade and Lawrence Witmer 23 Vicariance biogeography and paleoecology of Eurasian Miocene hominoid primates Peter Andrews and Raymond L. Bernor Index

ix

413

436

454 488

Contributors

Laura Abbazzi Dipartimento di Scienze della Terra, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Jorge Agustı´ Institute of Paleontology, M. Crusafont, Escola Industrial 23, E-08201 Sabadell, Spain Maria A. Alvarez-Sierra Departamento de Paleontologia y UEI, Facultad de Ciencias Geolo ´ gicas, Universidad Complutense y CSIC, Ciudad Universitaria, 24040 Madrid, Spain Peter Andrews Department of Palaeontology, Natural History Museum, Cromwell Road, London SW7 5BD, UK Marco Benvenuti Dipartimento di Scienze della Terre, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Raymond L. Bernor Department of Anatomy, Laboratory of Paleobiology, Howard University, 520 W. Street, Washington DC 20059, USA Adele Bertini Dipartimento di Scienze della Terre, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Mostefa Bessedik Institute des Sciences de la Terre, Universite´ d’Oran Es-Se´nia, BP 1524, DZ 31100 Oran, Algeria the late Ezzedine Bessais Genevieve Bouvrain Lab. Pale´ontologie des Verte´bre´s et Pale´ontologie humaine, Universite´ Pierre et Marie Curie, 4 place Jussieu, 75005 Paris, France Francis H. Brown Department of Geology and Geophysics, University of Utah, Salt Lake City, UT 84112, USA

Contributors

Lluı´s Cabrera Grup de Geodinamica i Analisi de Conques, Department de EstratigraWa i Paleontologia, Universitat de Barcelona, Campus Pedralbes, E-08028 Barcelona, Spain Georges Clauzon CEREGM (UMR 6635 CNRS), Europo ˆ le de l’Arbois, BP 80, 13545 Aix-en-Provence Cedex 04, France Martine Clet Morphodynamique continentale et coˆtiere, Universite´ de Caen, 24 rue des Tilleuls, 14000 Caen, France Nathalie Combourieu-Nebout Pale´ontologie et Stratigraphie (ESA 7073 CNRS), case courrier 106, Universite´ Pierre et Marie Curie, 4 place Jussieu, 75252 Paris Cedex 05, France Remmert Daams Departamento de Paleontologia y UEI, Facultad de Ciencias Geolo ´ gicas, Universidad Complutense y CSIC, Ciudad Universitaria, 24040 Madrid, Spain Louis de Bonis Lab. Ge´obiologie, Biochronologie et Pale´ontologie humaine, Faculte´ de Sciences, Universite´ de Poitiers, 40 av du Recteur Pineau, F 86022 Poitiers Cedex, France Peter B. deMenocal Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY 10964, USA Filomena Diniz Departamento de Geologia, Universidade de Lisboa, rua Escola Polytecnica 58, 1294 Lisboa Codex, Portugal Burkhart Engesser Naturhistorisches Museum, Agustinergasse 2, CH 4001 Basel, Switzerland Daniela Esu Dipartimento di Scienza della Terra, Universita´ ‘La Sapienza’, P. le A. Moro, 5, 00185 Roma, Italy Se´verine Fauquette Lab. de Botanique Historique et Palynologie, Inst. Me´diterrane´en d’Ecologie et de Pale´oe´col., Faculte´ des Sciences de Saint Je´roˆme, 13397 Marseille Cedex 20, France

xi

Contributors

xii

Najat Feddi De´partment des Sciences de la Terre, Faculte´ des Sciences Semlalia, Universite´ cadi Ayyad, avenue Prince Moulay Abdellah, BP S15 Marrakech, Morocco Mikael Fortelius Finnish Mueseum of Natural History, University of Helsinki, PO Box 11, 0014 Helsinki, Finland Jens Lorenz Franzen Forschungsinstitut Senckenberg, Senckenberg-Anlage 25, D-60325 Frankfurt am Main, Germany Miguel Garce´s Grup de Geodinamica i Analisi de Conques, Department de EstratigraWa i Paleontologia, Universitat de Barcelona, Campus Pedralbes, E-08028 Barcelona, Spain Robert Wynn Jones BP, Exploration Operating Co. Ltd, Chertsey Road, Sunbury on Thames Middlesex TW16 7LN, UK John Kappelman Department of Anthropology, University of Texas at Austin, Austin, TX 78712-1086, USA Meike Kholer Institute of Pakontology, M. Crusafont, Escola Industrial 23, E-08201 Sabadell, Spain George D. Koufos Department of Geology & Physical Geography, School of Geology, Aristotle University of Thessaloniki, 540 06 Thessaloniki, Macedonia, Greece Manel Llenas Institute of Paleontology, M. Crusafont, Escola Industrial 23, E-08201 Sabadell, Spain Juha Pekka Lunkka Finnish Museum of Natural History, University of Helsinki, PO Box 11, 0014 Helsinki, Finland Pierre Mein Centre des Sciences de la Terre, Universite´ Lyon I, 43, Bd du 11 Novembre 1918, 69622 Villeurbanne-Cedex, France

Contributors

Henriette Meon UFR des Sciences de la Terre, Universite´ Claude Bernard – Lyon I, 27–43 boulevard du 11 Novembre, 69622 Villeurbanne Cedex, France Jorge Morales Museo Nacional de Ciencias Naturales, CSIC, Jose´ Gutierrez Abascal 2, E-28006, Madrid, Spain Salvador Moya`-Sola` Institute of Pakontology, M. Crusafont, Escola Industrial 23, E-08201 Sabadell, Spain Valentin A. Nesin Institute of Zoology, Ukrainian Academy of Sciences, 15 Bogdan Khmelnitsky Str, 252030 Kiev 30, Ukraine Manuel Nieto Museo Nacional de Ciencias Naturales, CSIC, Jose´ Gutierrez Abascal 2, E-28006, Madrid, Spain Eileen M. O’Brien School of Forestry, Wales, Bangor, Gwyned, LL57 2UW Mauro Papini Dipartimento di Scienze della Terre, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Pablo Pelaez-Campomanes Museo Nacional de Ciencias Naturales, Jose´ Gutierrez Abascal 2, E-28006, Madrid, Spain Charles R. Peters Institute of Ecology, University of Georgia, Athens, GA 30602, USA J. Michael Plavcan New York College of Osteopathic Medicine, Old Westbury, NY 11546, USA Jay Quade Department of Geosciences, University of Arizona, Tucson, AZ 85721, USA Pierre Que´zel Lab. de Botanique Historique et Palynologie, Inst. Me´diterrane´en d’Ecologie et de Pale´oe´col., Faculte´ des Sciences de Saint Je´roˆme, 13397 Marseille Cedex 20, France

xiii

Contributors

xiv

Jelle W. F. Reumer Natuurmuseum, Rotterdam, PO Box 23452, NL 3001 KL Rotterdam, The Netherlands Fred Ro¨gl Naturhistorisches Museum. Geol. Palaeont. Abt, Burgring 7, A-1014, Vienna, Austria Lorenzo Rook Dipartimento di Scienze della Terre, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Brian R. Rosen Department of Palaeontology, The Natural History Museum, Cromwell Road, London SW7 5BD, UK Sevket Sen Lab. de Paleontologie des Verte´bre´s, Universite´ Pierre et Marie Curie, 4 Place Jussieu, 75252 Paris, France Nikos Solounias New York College of Osteopathic Medicine, Old Westbury, NY 11546, USA Gerhard Storch Forschungsinstitut Senckenberg, Senckenberg-Anlage 25, D-60325 Frankfurt am Main, Germany Danica Suballyova UFR des Sciences de la Terre, Universite´ Claude Bernard – Lyon I, 27–43 Bd du 11 Novembre 1918, 69622 Villeurbanne Cedex, France Jean-Pierre Suc Centre de Pale´ontologie Stratigraphique et Pale´oe´cologie (UMR 5565), Universite´ Claude Bernard – Lyon I, 27–43 Bd du 11 Novembre 1918, 69622 Villeurbanne Cedex, France Naima Bachiri TaouWq De´partment de Ge´ologie, Faculte´ des Sciences Ben M’Sik, Universite´ Hassan II Mohammedia, BP 7955, Casablanca, Morocco Giovanni Testa Dipartimento di Scienze della Terre, Universita´ di Firenze, via G. la Pira 4, 50121 Firenze, Italy Vadim A. Topachevsky Institute of Zoololgy, Ukrainian Academy of Sciences, 15 Bogdan Khmelnitsky Str, 252030 Kiev 30, Ukraine

Contributors

Albert J. van der Meulen Department of Stratigraphy & Paleontology, Institute of Earth Sciences, Budapestlaan 4, 3508 TA, Utrecht, The Netherlands Lawrence Witmer Department of Biological Sciences, Ohio University, Athens, OH 45701, USA Zhuo Zheng Faculty of Geology, Zhongshang University, 510275 Guangzhou, People’s Republic of China

xv

A c kn o wl ed geme nt s: T h e Eu r o p ea n Science Foundation

The European Science Foundation (ESF) acts as a catalyst for the development of science by bringing together leading scientists for funding agencies to debate, plan and implement pan-European scientiWc and science policy initiatives. ESF is the European association of more than 60 major national funding agencies devoted to basic scientiWc research in over 20 countries. It represents all scientiWc disciplines: physical and engineering sciences, life and environmental sciences, medical sciences, humanities and social sciences. The Foundation assists its Member Organisations in two main ways ways: by bringing scientists together in its scientiWc programmes, networks, exploratory workshops and European research conferences, to work on topics of common concern; and through the joint study of issues of strategic importance in European science policy. It maintains close relations with other scientiWc institutions within and outside Europe. By its activities, the ESF adds value by cooperation and coordination across national frontiers and endeavours, oVers expert scientiWc advice on strategic issues, and provides the European forum for fundamental science.

1 Introduction Jorge Agustı´ , Lorenzo Rook and Peter Andrews

The late Neogene (the period between − 14 and − 2.4 Ma) is one of the most interesting phases in understand the present conWguration of terrestrial ecosystems. It was during this time that the change took place from the middle Miocene dominant subtropical forests that stretched across southern Europe and western Asia to a more open but still wooded biotope that now prevails in warm–temperate areas. This change in vegetation, which strongly aVected the composition of mammalian faunas, seems to be linked to the rapid spread of grasses around 8–10 Ma ago. Moreover, in the late Neogene, climatic shifts and falling temperatures due to the spread of the Antarctic Ice, were followed by the Plio-Pleistocene Arctic glaciations in the Northern Hemisphere. Furthermore, during the late Neogene, important changes at the regional level gave rise to the present conWguration of the Old World land masses, and the successive drying of inland seas in the European area facilitated faunal interchange between Europe and Central Asia and Africa. At the same time, other tectonic processes like the Himalayan and the Tibetan uplifts and the opening up of the great eastern African basins and Red Sea, although working in opposite directions, favoured the processes of speciation and isolated evolution. All these phenomena must have had a strong inXuence on hominid evolution between 14 and 2 million years ago (Ma). The diversity of hominid species recorded in that period must be considered in relation to the changing environmental conditions and to constrictions imposed by the existence of signiWcant zoogeographical barriers. As a consequence, at around 10 Ma ago Eurasia displayed a large variety of apes, with gracile forest-dwellers (Dryopithecus, Sivapithecus), dry-adapted forms (Ankarapithecus) and large robust gorilla-like forms (Graecopithecus), and all this variety in a period when there is a huge information gap in Africa. In some ways, the Eurasian Miocene radiation of apes parallels the trends observed in Africa 5 Ma later that led to the appearance of the Wrst bipedal hominids. DiVering from the Pliocene hominid evolution in Africa, the changing environments did not result in Eurasia in the emergence of bipedal apes, but in the extinction of generalised morphotypes and the persistence of highly specialised forms in relic areas (Gigantopithecus, Pongo). However, knowledge of the ecosystems in which hominids of the late Miocene evolved is still incomplete, and very little integrated analysis of data has been attempted, in particular making use of modern techniques for palaeoenvironmental analysis such as isotopes, pollen, micromammals. In order to focus on these issues and to foster co-operation between scientists dealing with them, the scientiWc network ‘Hominoid evolution

Introduction

2

and environmental change on the Neogene of Europe’ was approved by the European Science Foundation in 1995. One of the goals of this network was to create a database on Neogene Mammals of Eurasia, that could be used for further analysis of the ecosystems where hominoids lived. The network also organised three workshops in order to analyse diVerent aspects of the late Neogene time and their relevance to hominoid evolution. The Wrst workshop was held in Sant Feliu de Guixols (Spain), from the 24th to the 27th of October 1997, and had as the main topic ‘The Vallesian’. For analysing this particular interval of the late Miocene, 24 scientists from 11 countries attended this Wrst workshop. The workshop was mainly devoted to analysing the faunal and environmental changes that took place during the period known as the Vallesian stage. However, this analysis was not strictly limited to that period but included also the intervals immediately preceding and succeeding the Vallesian (eg., late Aragonian and early Turolian). In this way, a number of presentations dealt with regional mammalian successions ranging from Spain to Southern Asia: Spain (Agustı´ et al., Morales et al.), France (Mein), Italy (Rook & Engesser), Central Europe (Franzen & Storch), the Aegean area (Koufos & De Bonis), Anatolia (Fortelius et al.) and Eastern Europe (Nesin & Topachersky). In the case of Eastern Spain, the communications presented by the Spanish team were complemented with a Weld-trip in the Valle`s-Penede`s Basin, which enabled the participants to have an accurate idea of the work developed in this area. The emphasis of the workshop was put on the comparison and close correlation of long sections bearing sequences of large and small mammal localities, instead of the usual correlation between isolated localities lacking an accurate geological context (for instance, in the case of karstic Wssure inWllings). In this way, comparison could be made between sections bearing a detailed magnetostratigraphic analysis, particularly the sections of the Potwar Plateau (Pakistan), the Valle`s-Penede`s (Spain) and Sinap (Turkey). Another signiWcant topic of the workshop was the so-called ‘Mid-Vallesian Crisis’, an extinction event that followed the change from the middle to the late Miocene. Results reviewed at the workshop suggested that a change similar to the Mid-Vallesian Crisis of Western Europe occurred at diVerent times in diVerent areas. The change was already established in the early Vallesian in the so-called Sub-Parathetyan Province (data from Maragheh by Bernor), but in Western Europe did not take place until about 9.6 Ma ago (data from the Valle`s-Penede`s by Agustı´ and co-workers). On the other hand, in Central Europe, several species indicating wet, forested conditions persisted well into the late Vallesian (data from Franzen & Storch). Finally,

Introduction

such a change was not seen in Southern Asia (Pakistan) until 8–7.2 Ma (data from the Potwar Plateau). The second workshop took place in Certosa di Portignano (Siena, Italy), devoted to the ‘Climatic and environmental change in the Neogene of Europe’; 25 scientists from 11 countries and diVerent palaeoenviromental disciplines had the rare opportunity to experience a real interfacing of data from the terrestrial ecosystems, the shallow marine realm and the deep sea. A Wrst set of contributions dealt with the palaeogeographic evolution of the Tethys area and the eVects of changes in the extent of the Tethys sea on faunal distributions and climate evolution. F. Ro¨gl (Vienna) presented a sketch of the evolution of this area from the early Tertiary to the late Miocene. With the contact between the Arabian plate and the Anatolian plate, land bridges formed between Africa and Eurasia opened and closed throughout the middle Miocene, beginning about 19 Ma. No hominoid primates are known this early in Europe, but of particular signiWcance to hominoid migrations was the end-Burdigalian regression of the sea at around 16 Ma, as this coincides with the earliest evidence of hominoids in central Europe. Contributions to climate modelling during the late Neogene were provided by P. B. DeMenocal & F. H. Brown, and by E. O’Brien. According to DeMenocal & Brown, marine records of African climatic variability document a shift toward prolonged and seasonally more arid conditions after 2.8 Ma. This is linked to cold North Atlantic sea-surface temperatures associated with onset of Arctic Ice sheets. Major changes in African faunas coincide with this climatic change suggesting that some speciation events may have been climatically mediated. E. O’Brien demonstrated that a large proportion (79%) of African woody plant species richness is accounted for by two aspects of climate, annual rainfall and an optimised function of energy (minimum monthly potential evapotranspiration). Finally, the palaeobotanical approach included the point of view of the palynological analysis (J. P. Suc, A. Bertini, G. Clauzon & D. Suballyova). Evidence for climate change was considered both from invertebrate and vertebrate evidence. Data presented by Rosen demonstrated that most coral reefs (and their associated z-corals) occur in the Mediterranean area in three major high sea-level phases, corresponding to the early (Aquitanian), middle (Langhian–Serravalian) and late Miocene (Tortonian–early Messinian). Small mammals were discussed by Reumer (Insectivores), Daams et al. (rodents from Central Spain) and Agustı´ et al. (mainly rodents from Eastern Spain). According to Reumer, ecological studies have shown that environmental moisture may be the ultimate determinant of within-habitat diversity and numerical abundance of soricids, which can be taken as a good

3

Introduction

4

indicator of relatively warm and humid palaeoclimate. Shrews (Soricidae, Insectivora) are among the most relatively sensitive mammals against climatic shifts because of their small size and high surface/volume ratio. Three main periods of faunal turnover, corresponding to humid and warm conditions, characterise the Neogene history of shrews: the early Miocene (19–20 Ma), the early–late Miocene (Vallesian; 9–11 Ma), and the latest Miocene to Pliocene (6 and 2.5 Ma). Climatic deterioration at around 2.7–2.3 Ma Wnally caused a severe reduction of the European shrew fauna. Among rodents, the analysis of the Calatayud-Daroca succession developed by Daams and coworkers also suggests, as for the shrews, that species diversity is more related to relative humidity than to temperature. Quantitative analysis of the rodent succession developed by Agustı´ and coworkers in the Valle`s-Penede`s Basin allows the recognition of an alternation of dry and humid phases during the Miocene and early Pliocene times. Early Miocene localities indicate forested, humid conditions, similar to those recorded in the early Miocene of Central Europe. Increasingly dry conditions are recorded across the late early and middle Miocene (early middle Aragonian), but a return to more humid conditions is observed in the late Aragonian (Serravallian) times. The middle/late Miocene boundary coincides with a relatively dry period, followed again by a humid peak in the early Vallesian (early Tortonian), and it is this time that coincides with the maximum abundance of hominoid remains in Western Europe. Again, as in the case of shrews, it appears that species diversity is more related to relative humidity than to temperature. Contributions dealing with large mammal associations were those of N. Solounias, M. Plavcan, L. Witmer & J. Quade, L. de Bonis & G. Koufos, L. Kordos, and R. Bernor. After a detailed analysis based on a variety of sources (dental and postcranial ecomorphology of bovids, palynology, isotopes), Solounias and co-workers arrived to the conclusion that the main habitat of the Pikermian mammals was not a savanna (‘the savanna myth’), but sclerophyllous evergreen woodland similar to today’s mixed monsoon forest and grassland glades of north Central India. Large mammals with lucky exaptations migrated into Africa from the Pikermian bioprovince. A similar topic was developed by de Bonis & Koufos, who found in Greek faunas evidence of a trend towards drier conditions in the late Miocene. On the other hand, the disappearance of hominoids and other forest elements in the Pannonian Basin at 9–7 Ma was attributed by Kordos to the regression of the Pannonian sea rather than to a general climatic trend. Finally, Bernor & Andrews discussed the patterns of hominoid immigration, dispersal and extinction. Hominids entered Europe in the middle Miocene because land crossings of the Tethys were possible at this time and the subtropical forest

Introduction

environments were suitable. By virtue of the taxonomic diversity and restricted biogeographic ranges alone, Eurasian Miocene hominids show strongly vicariant evolutionary patterns. This suggests a model whereby a founding species extends its range under favourable and speciWc environmental circumstances and then becomes geographically restricted to refugia by geographic ( = tectonic and palaeogeographic) and/or environmental events. A frequent byproduct of vicariance, exercised over millions of years time, is homoplasy, and it is evident that there was a great deal of homoplasy in Miocene hominids. The nature of the environments occupied by apes in Europe had many structural similarities with the environments in Africa with which they are associated at this time. Palaeoecological evidence suggests that African middle Miocene apes lived in seasonal woodlands and forests, for example at Fort Ternan and Maboko Island. The hominids at these sites were partly terrestrial and with their large thick-enamelled teeth were adapted for similar diets to some of the European apes. The earliest European apes were similar in being both partly terrestrial and with almost identical dietary adaptations. The community structure of the mammalian faunas in the African and European sites were extremely similar, and by inference the ecosystem they occupied was also similar. The similarities in locomotor and dietary adaptations of the African and European apes at this early stage indicate further that their position in their respective ecosystems was also very similar. In one sense, therefore, these middle Miocene taxa in Europe were not as distinct from their African relatives as taxonomic divisions and their geographic separation may appear to indicate. Towards the end of the middle Miocene, at 13–12 Ma, the trends of partial terrestriality and thick-enamelled frugivory continued in a group of fossil apes assigned either to the pongine clade or to a paraphyletic group unrelated to any living. They are associated with a range of open forest to woodland environments ranging from southeast Europe to China, and one genus at least may be related to the orang utan. At the same time, the more arboreal and suspensory Dryopithecus emerged in association with closed subtropical forest environments where they are sometimes found associated with Pliopithecus and Anapithecus. Similar environments and a similar, possibly heritage, adaptation for suspensory locomotion persisted in Oreopithecus, although it has recently been argued that this fossil ape may also have had adaptations for terrestrial bipedal mode of locomotion and the evolutionary relationships of Oreopithecus are still unclear. Later in the Miocene, the Alpine–Himalayan orogeny caused major changes in land–sea relations, global climatic circulation patterns and seasonality particularly in Central Asia. Regression of the Paratethys likewise

5

Introduction

6

caused a shift of habitats to greater seasonality and replacement of evergreen subtropical forests by deciduous woodlands and, progressively in the late Miocene, more seasonal warm temperate woodlands with progressively more open habitats. Hominid primate distribution tracked these changes closely during the 12–9 Ma interval, contracting their range from both west and east and Wnding temporary refuge in southeastern Europe, where favourable subtropical conditions persisted for a time after being lost elsewhere. Hominids disappeared from this region Wnally during MN11, although they persisted until MN12 in local insular habitats in Italy and the latest Miocene of China.

PART I

Palaeogeography of the circum-Mediterranean region

2 Mediterranean and Paratethys Palaeogeography during the Oligocene and Miocene Fred Ro¨ gl

Introduction The Cenozoic conWguration of continents and oceans is strongly inXuenced by plate tectonic movements. Opening and closing pathways for mammal migrations and marine exchanges are one of the triggering forces for faunal events and evolution. The impacts of a vanishing Tethys Ocean in the mid-Cenozoic are not only important for the marine and continental biotas of Eurasia and the Mediterranean, but also inXuenced the environmental conditions worldwide. The dispersal of continents in the Southern Hemisphere with the northward movement of the Indian and Australian continents, together with the counterclockwise rotation of Africa, closed down the Tethys Ocean. Parallel to these movements the Atlantic Ocean opened. The Mesozoic oceanic circulation patterns changed to varying conditions with decreasing temperatures in the Cenozoic. Based on oxygen isotopes (Kennett, 1995), bottom water temperatures were highest in the late Palaeocene–early Eocene (c. 55 Ma). Distinct steps to colder conditions followed around the Eocene/Oligocene boundary (33–35 Ma), in the middle Miocene (15 Ma), and in the Pliocene (3 Ma). DiVerent palaeogeographic reconstructions over the past two decades have attempted to solve the development of the Cenozoic Mediterranean and Paratethys areas. In many cases these reconstructions depend on the present sediment distribution and do not consider palinspastic reconstructions based on plate tectonic movements (e.g. Vinogradov, 1967–69; Senes & Marinescu, 1974; Steininger et al., 1985a; Hamor & Halmai, 1988; Cahuzac et al., 1992; Popov et al., 1993). Other reconstructions include too long a time span within one time slice to present a distinct time level within a strongly changing environment (e.g. Biju-Duval et al., 1977; Dercourt et al., 1985, 1993). The best reconstructions, based on tectonic, sedimentological and stratigraphic investigations are currently available for the Western Mediterranean (Boccaletti et al., 1986, 1990). The sketches for the Neogene, produced by Ro¨gl & Steininger (1983) and Steininger & Ro¨gl (1984), were based on plate tectonic hypotheses, sediment distribution, Eurasian mammal migrations, and marine faunal similarities between the Mediterranean and Paratethys. At that time the information on faunal development and stratigraphic correlation was poor for the Eastern Paratethys. Therefore a revision of those interpretations, including the early history of the Paratethys was discussed by Ro¨gl (1998a,b). The sketches presented here are based on plate

Mediterranean and Paratethys palaeogeography

distributions at three levels from the Eocene to the late Miocene (Scotese et al., 1988). The goal of the present palaeogeographic sketches is to explain mammal migration possibilities in and between Eurasia and Africa. Some marine connections are still being debated (Adams, 1998), and certain seaways proposed here are highly speculative. The open questions will require further tectonic and palaeontological studies. A postulate for any time slice is the exact stratigraphic correlation of the diVerent basins included in the reconstructions. The stratigraphic correlation chart (Table 2.1) is based on the most recent time tables of Berggren et al. (1995) and Steininger et al. (1996), correlated to the Paratethys (Popov et al., 1993; Jones & Simmons, 1996; Ro¨gl, 1998b).

Birth of the Paratethys In the late Eocene the Indian Plate collided with Eurasia. The Tethys Ocean vanished, leaving as relics the Mediterranean Sea at its western end, and to the north the intercontinental Paratethys Sea in Eurasia (Fig. 2.1). In the late Eocene, a pelagic Globigerina marl facies developed from the western Mediterranean to the Alpine–Carpathian belt and the inner-Asian Transcaspian Basin; north–south gradients in planktic faunal assemblages were evident. In the shallower parts of the basin tropical, larger foraminifera and mollusc faunas thrived. In the Eastern Paratethys the molluscs had a latitudinal distribution, with the most tropical assemblages in Transcaucasia (Akhaltsikhe depression), whereas in northern Ustyurt a northern fauna of low diversity was recorded (Krasheninnikov, 1974; Popov et al., 1993; Popov, 1994). From the eastern part of the later Paratethys, the tropical Tethys Ocean communicated with the Polar Sea via the shallow Turgai Strait in western Siberia during the middle–late Eocene (Vinogradov, 1967–69; Popov et al., 1993). This marine barrier prevented a continental faunal exchange between Asia and Europe. During the Eocene, western and middle Europe existed as an archipelago, with a distinct reduction of mammal diversity during the Priabonian. Northern Europe formed a landmass connected by the De Geer Route (over Svalbard) with North America (Ko¨nigswald, 1981). Around the Eocene/Oligocene boundary, continental collision and movements along the Alpine–Himalayan tectonic belt closed oV the Paratethys. The Turgai Strait vanished, and the continentalisation of the European archipelago formed new conWgurations (Ziegler, 1990). The mammal immigration wave, coming from the east out of North America and Asia, reached western Europe during nannoplankton zone NP 22 (32–33 Ma) and resulted in the so-called ‘Grande Coupure’ (Stehlin, 1909; Tobien, 1987).

9

Palaeogeography of the circum-Mediterranean region

10

Table 2.1. Stratigraphic correlation chart of the Central and Eastern Paratethys regional stage systems (Berggren et al., 1995; Popov et al., 1993; Ro¨gl, 1996, 1998a,b; Steininger et al., 1996).

Mediterranean and Paratethys palaeogeography

11

[Figure 2.1] Palaeogeographic sketch of the western end of the Tethys Ocean in the late Eocene. The western Tethys communicated with the Polar Sea via the Turgai Strait and prevented mammal migrations between Asia and Europe (Ro¨gl, 1998a,b).

Sixteen new mammal families are mentioned by Savage (1990). New important immigrants in mammal zone MP 21 include: Anthracotherium, Eusmilus, Rhinocerothidae, Lagomorpha, Eomyidae, and Cricetidae. This faunal migration was made possible by the closure of the Turgai Strait and the continuing northwestward movement of North America, closing in for a Wrst Bering landbridge. The Paratethys extended as an orogenic foredeep and intercontinental sea from the Western Alps to Central Asia (early Kiscellian/Pshekian). Marine connections existed in the far west with the Mediterranean, and to the northwest along the Danish–Polish Strait with the North Sea Basin (only in LatdorWan, NP 21).

First Paratethys isolation Cold water inXux, dysaerobic bottom conditions, aberrant microXoras and microfaunas, and northern molluscs characterised the early to middle Oligocene (Baldi, 1979, 1984; Krhovsky et al., 1991; Popov et al., 1993; Rusu et

Palaeogeography of the circum-Mediterranean region

12

[Figure 2.2] The isolated Paratethys Basin in the early Oligocene. Only narrow seaways remained open in the west. Salinity decreased and endemic conditions developed in the entire basin. The closure of the Turgai Strait established the conditions for the Eurasian faunal exchange known as the ‘Grande Coupure’ (Ro¨gl, 1998a,b).

al., 1996). The isolation of the Paratethys culminated (Fig. 2.2) during nannoplankton zone NP 23 (Solenovian stage) with reduced salinity conditions and endemic bivalves (‘Cardium lipoldi’ – Janschinella fauna). In contrast, the Mediterranean Tethys remained open between the Indo-PaciWc and Atlantic Ocean, and prevented a continental faunal exchange between Eurasia and Africa. Most of the Oligocene was characterised by a continuous faunal exchange within Eurasia, rather than a distinct migration wave, e.g. the appearance of Melissiodon and Paracricetodon in MP 24, and of Zapodidae in MP 26.

Tropical marine excursion A worldwide excursion of the tropical belt in the marine realm occurred in the late Oligocene–early Miocene. Horizons of tropical, larger foraminifera and molluscs are reported from the Mediterranean, the Paratethys, and the Middle East (Adams, 1973, 1976, 1983; Baldi & Senes, 1975; McGowran,

Mediterranean and Paratethys palaeogeography

13

[Figure 2.3] In the late Oligocene–early Miocene a transgression spread from the Middle East to the Mediterranean and Paratethys. Larger foraminifera, molluscs, and corals fringed the eastern shelf along the Lut Block from Makran to Qum Basin, Lake Urmia, and Transcaucasia. In the Western Paratethys the Rhinegraben connection and the seaway along the Alpine Foredeep closed. In the Western Mediterranean the Balearic Basin started to spread, with an eastward move of the Apennines (Ro¨gl, 1998a,b).

1979a,b; Popov et al., 1993; McCall et al., 1994). The Mediterranean–Indian Ocean gateway was open between the Anatolian and Arabian Plates (Fig. 2.3). A tropical-subtropical southeastern connection for the Paratethys newly opened across the Iranian plate (Transcaucasia–Qom Basin–Makran). In the south, the rotation of Arabia opened the graben of the Red Sea in the late Oligocene (Jones & Racey, 1994). The Central Paratethys was connected to the Mediterranean along fault structures in the Alps and between the Alps and Dinarides. The marine straits in the western Alpine foredeep and the Rhinegrabenwere closed intermittently. The Alpine foredeep re-opened for a seaway to the Rhoˆne Basin in the early Burdigalian (Eggenburgian). In the late Oligocene (mammal zone MP 28–30) new forms appeared in Europe successively: Heterocricetodon, Rhizospalax, and the second invasion of Lagomorpha. In the early Miocene (Aquitanian), important new forms include the suoids Palaeochoerus and Hyotherium in the European mammal zone MN 1 and Bunolistriodon in Dera Bughti (Pakistan). In the lower Burdigalian (MN 3) the North American immigrant Anchitherium reached Europe (Thenius, 1972; Steininger et al., 1985b; Made, 1990). A

Palaeogeography of the circum-Mediterranean region

14

Eurasian–African faunal exchange was still impossible, and no relation existed to the Wrst East African mammal fauna of Meswa Bridge (23.5 Ma).

The Gomphotherium Landbridge During the middle Burdigalian (upper Eggenburgian–Ottnangian), strong movements of the Savic tectonic phase changed the palaeogeographic patterns in the circum-Mediterranean (Fig. 2.4). The rotation of Africa Wnally closed the gap to Eurasia. The Arabian peninsula collided with the Anatolian Plate. The ‘Gomphotherium Landbridge’ was established. Continental faunal exchanges in both directions started in MN 4, around 19 Ma. Successive arrivals of gomphotheres, deinotheres, and primates are observed in Eurasia, while rhinos, tayassuids, or Amphicyon arrived in Africa. The Wrst Gomphotherium are recorded in the Siwaliks (Kamlial Formation) at  18 Ma,

[Figure 2.4] The Gomphotherium Landbridge. The Eastern Mediterranean seaway closed, and a landbridge opened between the Anatolian and Arabian/African Plates, enabling a remarkable faunal exchange. The Eastern Paratethys became entirely isolated, with reduced salinity and endemic faunas. The Mediterranean communicated with the Atlantic Ocean and fed the Western and Central Paratethys. The renewed seaway in the Alpine Foredeep was again connected to the North Sea by the Rhinegraben (Ro¨gl, 1998a,b).

Mediterranean and Paratethys palaeogeography

and in Poland in the coal mine Belchatow C also at this time (Thenius, 1979; Steininger et al., 1985b; Goldsmith et al., 1988; Barry & Flynn, 1989; Pickford, 1989; Kowalski & Kubiak, 1993; Fortelius et al., 1996; Bernor et al., 1996). The Paratethys Sea was divided in two separate realms. The Eastern Paratethys formed an enclosed basin with endemisms under reduced salinity (Kotsakhurian stage). The Western and Central Paratethys remained connected to the Mediterranean and via the Rhinegraben, to the North Sea. Cooler conditions were observed in the marine faunas. By the end of Ottnangian the Alpine foredeep became dry land. The Mediterranean itself was open to the Atlantic Ocean.

Re-opening of the Indo-Pacific gateway The exact process by which the seaway between the Indian Ocean and the Mediterranean reopened is still being discussed and remains controversial,

[Figure 2.5] Indo-Pacific recurrence. For a short time the Mediterranean–Indo-Pacific seaway reopened around the early–middle Miocene boundary. From Eastern Anatolia a new transgression flooded the Paratethys. The intramountain basins and the Carpathian foredeep in the Central Paratethys were covered by tropical–subtropical waters. A connection is proposed along the Rhodopes and Pontides, south of the Black Sea Plate. The Eastern Paratethys stayed in reduced communication (Ro¨gl, 1998a,b).

15

Palaeogeography of the circum-Mediterranean region

16

based on the interpretation of the distribution of larger foraminifera (Adams et al., 1983; Adams, 1998). At least for the lower Langhian, such a short connection along the Bitlis Zone between the Anatolian and Arabian Plates is required (Fig. 2.5). Another connection must have existed in eastern Anatolia linking the Paratethys with the Indian Ocean for an early Badenian transgression. During the time of open seaways the Eurasian/African mammal migrations were interrupted, which seems to be demonstrated by migration waves, e.g. arrivals of primates: Pliopithecus in MN 5, Griphopithecus and Plesiopliopithecus in MN 6, and Dryopithecus in MN 8 (Andrews et al., 1996; Begun, 1996). Final closure between Eurasia and Africa The Serravallian regression coincided with the re-establishment of the ‘Gomphotherium Landbridge’. Evaporitic sedimentation and continentalisation took place in the area around the Persian Gulf (Jones & Racey,

[Figure 2.6] The Paratethys closure. The end of open marine connections transformed the Paratethys environment into an endemic bioprovince under reduced salinity conditions. Since early Oligocene times, the most uniform biofacies was observed from the Vienna to the Caspian Basin in the Sarmatian. Intermittent narrow marine straits opened from the Mediterranean to the Eastern Paratethys along the East Anatolian fault zone. The Persian Gulf was characterised by evaporitic and continental sedimentation of the Fars and Gachsaran Formations (Ro¨lg, 1998a,b).

Mediterranean and Paratethys palaeogeography

1994). The Wnal closure of the circum-equatorial oceanic current system caused worldwide cooling and an increased accumulation of the East Antarctic ice sheet around 15 Ma (Kennett, 1995). Short-lived connections between the Indo-PaciWc, the Mediterranean and the Eastern Paratethys (Fig. 2.6) existed in eastern Anatolia in the upper middle Miocene in the Konkian/late Badenian and Sarmatian, but also in the Pliocene (Nevesskaya et al., 1984; Chepalyga, 1995; Iljina, 1995). These channels may have brieXy acted as barriers for mammal exchanges. The regression of the sea from the Greek mainland and the Aegean landmass at the end of the Burdigalian prevented marine connections with the Paratethys in the area of the Dardanelles. The earliest rodent faunas are reported from the Aegean in Orleanian times (Sen, 1982). Beginning with the Tortonian transgression, an Aegean seaway opened to the Black Sea Basin (Ro¨gl & Steininger, 1983; Marunteanu & Papaianopol, 1995). Since the Sarmatian the open ocean connections of the Paratethys were interrupted. The gigantic inland sea of the Paratethys turned into continuously shrinking basins. A reduced salinity and the alkaline chemistry of the aquatic realm led to strong endemisms and caused the stenohaline organisms in the Sarmatian sea to disappear (Pisera, 1996). The environments of the Eastern Paratethys Sea had higher salinities from the Sarmatian up to the Maeotian (late Miocene) and extended to the west to the Dacian Basin, whereas the enclosed Pannonian Basin turned to nearly freshwater conditions (Pannonian stage). The connecting facies between the Pannonian and Euxinian basins is expressed by the local Malvensian stage in the Eastern Carpathians and Dacian realm (Motas & Marinescu, 1969; Marinescu, 1985; Papaianopol et al., 1995). During MN 6 an increasing faunal exchange occurred between Eurasia and Africa (Bernor et al., 1996; Fortelius et al., 1996). The appearance of ‘Hipparion’ is one of the last Miocene immigration events from North America. A strong sea level drop around 11 Ma (Haq et al., 1988) once again opened the Bering landbridge. In a rapid wave the threetoed horse spread from China to the Eastern Paratethys and western Europe (Bernor et al., 1989). During the Turolian (late Miocene, MN 11–13) in southeastern Europe and southwestern Asia, the ‘Pikermian’ faunal province extended with a radiation in hipparionine horses and ruminants. Additionally, Asian ‘steppe elements’ such as Cricetus and Pliospalax appeared in Europe (Steininger et al., 1985b; Bernor, 1984; Bernor et al., 1996; Gentry & Heizmann, 1996). Finally, new but limited faunal routes with Africa were opened by the Messinian salinity crisis in the late Turolian (MN 13). Important representatives include Protatera, Hippopotamus, and Macaca (Agustı´, 1996). Modern conditions were established in the circum-Mediterranean by the Pliocene marine transgression coming from the Atlantic Ocean.

17

Palaeogeography of the circum-Mediterranean region

18

References Adams, C. G. 1973. Some Tertiary foraminifera. In Atlas of Palaeobiogeography, A. Hallam (ed.), pp. 453–68. Amsterdam, Elsevier. Adams, C. G. 1976. Larger foraminifera and the late Cenozoic history of the Mediterranean region. Palaeogeography, Palaeoclimatology, Palaeoecology, 20, 47–66. Adams, C. G. 1983. Speciation, phylogenesis, tectonism, climate and eustasy: Factors in the evolution of Cenozoic larger foraminiferal bioprovinces. In Evolution, Time and Space. The Emergence of the Biosphere, Price, J. H. & Whalley, P. E. S. (eds.), pp. 255–89. London-New York, Academic Press. Adams, C. G. 1998 (in press). The terminal Tethyan event: A critical review of the conXicting age determinations for the disconnection of the Mediterranean from the Indian Ocean. In Fossil Vertebrates of Arabia, Whybrow, P. J. & Hill, A. (eds.). New Haven, Yale University Press. Adams, C. G. , Gentry, A. W. & Whybrow, P. J. 1983. Dating the terminal Tethyan event. Utrecht Micropaleontological Bulletin, 50, 273–98. Agustı´, J. 1996. Late Miocene mammal events in Europe. Europal, 10, 14–15. Andrews, P., Harrison, T., Delson, E., Bernor, R. L. & Martin, L. 1996. Distribution and biochronology of European and Southwest Asian Miocene Catarrhines. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 168–207. New York, Columbia University Press. Baldi, T. 1979. Changes of Mediterranean/?IndopaciWc/ and boreal inXuences on Hungarian marine molluscfaunas since Kiscellian until Eggenburgian times; the stage Kiscellian. Annales Geologiques Pays Helleniques, tom hors serie. 1979, fasc.I, 39–49. Baldi, T. 1984. The terminal Eocene and early Oligocene events in Hungary and the separation of an anoxic, cold Paratethys. Eclogae Geologicae Helvetiae, 77, 1–27. Baldi, T. & Senes, J. 1975. OM Egerien. Die Egerer, Pouzdraner, Puchkirchener Schichtengruppe und die Bretkaer Formation. Chronostratigraphie und Neostratotypen, 5, 577 pp. Bratislava, Slovenskej akademie vied. Barry, J. C. & Flynn, L. J. 1989. Key biostratigraphic events in the Siwalik sequence. In European Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.). NATO ASI Series, ser. A, 180, 557–71. New York-London, Plenum Press. Begun, D. R. 1996. Events in European Hominoid evolution. Europal, 10, 16–20. Berggren, W. A., Kent, D. V., Swisher, C. C., III & Aubry, M.-P. 1995. A revised Cenozoic geochronology and chronostratigraphy. SEPM (Society of Sedimentary Geology), Special Publication, 54, 129–212. Bernor, R. L. 1984. A zoogeographic theater and biochronologic play: The time/biofacies phenomena of Eurasian and African Miocene mammal provinces. Pale´obiologie Continentale, 14, 121–42. Bernor, R. L., Fahlbusch, V., Andrews, P., De Bruijn, H., Fortelius, M., Ro¨gl, F., Steininger, F. F. & Werdelin, L. 1996. The evolution of Western Eurasian Neogene mammal faunas: A chronologic, systematic, biogeographic, and paleoenvironmental synthesis. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp.

Mediterranean and Paratethys palaeogeography

449–69. New York, Columbia University Press. Bernor, R. L. Tobien, H. & Woodburne, M. O. 1989. Patterns of Old World hipparionine evolutionary diversiWcation and biogeographic extension. In European Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.). NATO ASI Series, ser. A, 180, 263–319. New York-London, Plenum Press. Biju-Duval, B., Dercourt, J. & Le Pichon, X. 1977. From the Tethys Ocean to the Mediterranean Seas. A plate tectonic model of the evolution of the western Alpine system. In Structural History of the Mediterranean Basins, Biju-Duval, B. & Montadert, L. (eds.), pp. 143–64. Paris, Editions Technip. Boccaletti, M., Calamita, F., Centamore, E., Chiocchini, U., Deiana, G., Micarelli, A., Moratti, G. & Potetti, M. 1986. Evoluzione dell’Appennino tosco-umbro-marchigiano durante il Neogene. Giornale di Geologia, ser. 3, 48, 227–33. Boccaletti, M., CiaranW, N., Cosentino, D., Deiana, G., Gelati, R., Lentini, F., Massari, F., Moratti, G., Pescatore, T., Ricci Lucchi, F. & Tortorici, L. 1990. Palinspastic restoration and paleogeographic reconstruction of the peri-Tyrrhenian area during the Neogene. Palaeogeography, Palaeoclimatology, Palaeoecology, 77, 41–50. Cahuzac, B., Alvinerie, J., Lauriat-Rage, A., Montenat, C. & Pujol, C. 1992. Palaeogeographic maps of the northeastern Atlantic Neogene and relation with the Mediterranean Sea. Paleontologia i Evolucio, Inst. Paleont. ‘Miguel Crusafont’, 24–25, 279–93. Chepalyga, A. L. 1995. East Paratethys–Tethys marine connections along Euphrat passage during Neogene. Abstracts 10th Congress RCMNS, Romanian Journal of Stratigraphy, 76, suppl. nr. 7, 149–50. Dercourt, J., Ricou, L. E. & Vrielynck, B. (eds.) 1993. Atlas Tethys Palaeoenvironmental Maps, pp. 307, 14 maps. Paris, Gauthier-Villars. Dercourt, J., Zonenshain, L. P., Ricou, L.-E., Kazmin, V. G., Le Pichon, X., Knipper, A. L., Grandjacquet, C., Sborshchikov, I. M., Boulin, J., Sorokhtin, O., Geyssant, J., Lepvrier, C., Biju-Duval, B., Sibuet, J.-C., Savostin, L. A., Westphal, M. & Lauer, J. P. 1985. Pre´sentation de 9 cartes pale´oge´ographiques au 1/20.000.000 s’etendant de l’Atlantique au Pamir pour la pe´riode du Lias a` l’Actuel. Bulletin de la Socie´te´ Ge´ologique de France, 8e ser., 1, 637–52. Fortelius, M., Van der Made, J. & Bernor, R. L. 1996. Middle and late Miocene Suoidea of Central Europe and the Eastern Mediterranean: Evolution, biogeography, and paleoecology. In The Evolution of Western Eurasian Neogene Mammal Faunas Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 348–77. New York, Columbia University Press. Gentry, A. W. & Heizmann, E. P. J. 1996. Miocene ruminants of the Central and Eastern Tethys and Paratethys. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V.& Mittmann, H.-W. (eds.), pp. 378–91. New York, Columbia University Press. Goldsmith, N. F., Hirsch, F., Friedman, G. M., Tchernov, E., Derin, B., Gerry, E., Horowitz, A. & Weinberger, G. 1988. Rotem mammals and Yeroham crassostreids. Stratigraphy of the Hazeva formation (Israel) and the paleogeography of Miocene Africa. Newsletters on Stratigraphy, 20, 73–90. Hamor, G. & Halmai, J. (eds.) 1988. Neogene Palaeogeographic Atlas of Central and Eastern Europe, 7 maps. Budapest: Hungarian Geological Institute.

19

Palaeogeography of the circum-Mediterranean region

20

Haq, B. U., Hardenbol, J. & Vail, P. R. 1988. Mesozoic and Cenozoic chronostratigraphy and cycles of sea-level changes. In Sea-level Changes – an Integrated Approach, C. K. Wilgus et al. (eds.). SEMP Special Publication, 42, 71–108. Iljina, L. B. 1995. Connections of Eastern Paratethys paleobasins with Tethyan seas in middle and late Miocene. Abstracts 10th Congress RCMNS, Romanian Journal of Stratigraphy, 76, suppl. nr. 7, 157. Jones, R. W. & Racey, A. 1994. Cenozoic stratigraphy of the Arabian Peninsula and Gulf. In Micropalaeontology and Hydrocarbon Exploration in the Middle East, M. D. Simmons (ed.), pp. 273–307. London, Chapman & Hall. Jones, R. W. & Simmons, M. D. 1996. A review of the stratigraphy of Eastern Paratethys (Oligocene–Holocene). Bulletin of the Natural History Museum London (Geology), 52, 25–49. Kennett, J. P. 1995. A review of polar climatic evolution during the Neogene, based on the marine sediment record. In Paleoclimate and Evolution – with Emphasis on Human Origins, E. Vrba et al. (eds.), pp. 49–64. New Haven, Yale University Press. Ko¨nigswald, W. 1981. Pala¨ogeographische Beziehungen der Wirbeltierfauna aus der alttertia¨ren Fossillagersta¨tte Messel bei Darmstadt. Geologisches Jahrbuch Hessen, 109, 85–102. Kowalski, K. & Kubiak, H. 1993. Gomphotherium angustidens (Cuvier, 1806) (Proboscidea, Mammalia) from the Miocene of Belchatow and the Proboscidean datum in Poland. Acta Zoologica Cracoviensis, 36, 275–80. Krasheninnikov, V. A. 1974. Some species of planktonic foraminifera from the Eocene and Oligocene deposits of South Armenia. Voprosyi Mikropaleontogia, Akademia Nauk SSSR, 17, 95–135. Krhovsky, J., Adamova´, M., Hladı´kova´a, J. & Maslowska´, H. 1991. Paleoenvironmental changes across the Eocene/Oligocene boundary in the Zda´nice and Pouzdrany Units (Western Carpathians, Czechoslovakia). The long-term trend and orbitally forced changes in calcareous nannofossil assemblages. In Nannoplankton Research, Proceedings of the 4th International Nannoplankton Association Conference, II, Hamrsmı´d, B. & Young, J. R. (eds.). Knihovnie`ka Zemnı´ Plyn Nafta, 14b, 105–87. Made, J. V. D. 1990. A range-chart for European Suidae and Tayassuidae. Paleontologia i Evolucio, 23 (1989–1990), 99–104. Marinescu, F. 1985. Der o¨stliche Teil des Pannonischen Beckens (Ruma¨nischer Sektor): Das Pannonien s. str. (Malvensien). In M6 Pannonien (Slavonien und Serbien), Papp, A., Jambor, A. & Steininger, F. F. (eds.). Chronostratigraphie und Neostratotypen, 7, 144–8. Budapest, Akademiai Kiado. Marunteanu, M. & Papaianopol, I. 1995. The connection between the Dacic and Mediterranean basins based on calcareous nannoplankton assemblages. Abstracts 10th Congress RCMNS, Romanian Journal of Stratigraphy, 76, suppl. nr. 7, 169–70. McCall, J., Rosen, B. & Darrell, J. 1994. Carbonate deposition in accretionary prism settings. Early Miocene coral limestones and corals of the Makran Mountain range in Southern Iran. Facies, 31, 141–78. McGowran, B. 1979a. Some Miocene conWgurations from an Australian standpoint. Annales Geologiques Pays Helleniques, tom hors serie 1979, fasc. III, 767–79. McGowran, B. 1979b. The Tertiary of Australia: Foraminiferal overview. Marine

Mediterranean and Paratethys palaeogeography

Micropaleontology, 4, 235–64. Motas, I. & Marinescu, F. 1969. L’evolution et les subdivisions du Sarmatien dans le Bassin Dacique. Fo¨ldtani Ko¨zlo¨ny, 101, 240–3. Nevesskaya, L. A., Voronina, A. A., Goncharova, I. A., Iljina, L. B., Paramonova, N. P., Popov, S. V., Chepalyga, A. L. & Babaak, E. V. 1984. Istoriya Paratetisa. 27th International Geological Congress Moscow, Paleookeanologiya, Koll. 03, Doklady, 3, 91–101. Papaianopol, I., Jipa, D., Marinescu, F., Ticleanu, N. & Macalet, R. 1995. Guide to excursion B2 (Post-Congress). Upper Neogene from the Dacic Basin. Romanian Journal of Stratigraphy, 76, suppl. nr. 1, 1–43. Pickford, M. 1989. Dynamics of Old World biogeographic realms during the Neogene: Implications for biostratigraphy. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.). NATO ASI Series, ser. A, 180, 413–42. New York-London, Plenum Press. Pisera, A. 1996. Miocene reefs of the Paratethys. A review. In SEPM Concepts in Sedimentology and Paleontology, pp. 97–104. SEPM (Society of Sedimentary Geology). Popov, S. V. 1994. Zoogeography of the Late Eocene basins of Western Eurasia based on bivalve mollusks. Stratigraphy and Geological Correlation, 2, 581–95. Popov, S. V., Akhmet’ev, M. A., Zaporozhets, N. I., Voronina, A. A. & Stolyarov, A. S. 1993. Evolution of Eastern Paratethys in the late Eocene–early Miocene. Stratigraphy and Geological Correlation, 1, 572–600. Ro¨gl, F. 1996. Migration pathways between Africa and Eurasia. Oligocene–Miocene palaeogeography. Europal, 10, 23–6. Ro¨gl, F. 1998a (in press). Oligocene–Miocene palaeogeography and stratigraphy of the Circum-Mediterranean region. In Fossil Vertebrates of Arabia, Whybrow, P. J. & Hill, A. (eds.). New Haven, Yale University Press. Ro¨gl, F. 1998b (in press). Palaeogeographic considerations for Mediterranean and Paratethys seaways (Oligocene to Miocene). Annalen des Naturhistorischen Museums in Wien, 99A. Ro¨gl, F. & Steininger, F. F. 1983. Vom Zerfall der Tethys zu Mediterran und Paratethys. Die neogene Pala¨ogeographie und Palinspastik des zirkum-mediterranen Raumes. Annalen des Naturhistorischen Museums in Wien, 85A, 135–63. Rusu, A., Popescu, Gh. & Melinte, M. 1996. Oligocene–Miocene transition and main geological events in Romania, 28 August–2 September 1996. A. Excursion guide. Romanian Journal of Paleontology, 76, suppl. 1, 3–47. Savage, R. J. G. 1990. The African dimension in European early Miocene mammal faunas. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.). NATO ASI Series, ser. A, 180, 587–99. New York-London, Plenum Press. Schlunegger, F., Burbank, D. W., Matter, A., Engesser, B. & Mo¨dden, C. 1996. Magnetostratigraphic calibration of the Oligocene to Middle Miocene (30–15 Ma) mammal biozones and depositional sequences of the Swiss Molasse Basin. Eclogae Geologicae Helvetiae, 89, 753–88. Scotese, Ch. R., Gahagan, L. M. & Larson, R. L. 1988. Plate tectonic reconstructions of the Cretaceous and Cenozoic ocean basins. Tectonophysics, 155, 27–48. Sen, S. 1982. Bioge´ographie et biostratigraphie du Ne´oge`ne continental de la re´gion ´ge´enne. Apports de rongeurs. Geobios, mem. spec. 6, 465–72. E

21

Palaeogeography of the circum-Mediterranean region

22

Senes, J. & Marinescu, Fl. 1974. Cartes pale´oge´ographiques due Ne´oge`ne de la Parate´thys Centrale. Memoires BRGM, 78, 785–92. Stehlin, H. G. 1909. Remarques sur les faunules de mammife`res des couches e´oce`nes et oligoce`nes du bassin de Paris. Bulletin de Socie´te´ Ge´ologiques de France, ser. 4, 9, 488–520. Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Miocene and Pliocene) marine–continental chronologic correlations of European mammal units. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 7–46. New York, Columbia University Press. Steininger, F. F., Rabeder, G. & Ro¨gl, F. 1985b. Land mammal distribution in the Mediterranean Neogene. A consequence of geokinematic and climatic events. In Geological Evolution of the Mediterranean Basin, Stanley, D. J. & Wezel, F.-C. (eds.), pp. 559–71. New York, Springer Verlag. Steininger, F. F. & Ro¨gl, F. 1984. Paleogeography and palinspastic reconstruction of the Neogene of the Mediterranean and Paratethys. In The Geological Evolution of the Eastern Mediterranean, Dixon, J. E. & Robertson, A. H. F. (eds.), pp. 659–68. The Geological Society, Oxford, Blackwell ScientiWc Publishers. Steininger, F. F., Senes, J., Kleemann, K. & Ro¨gl, F. (eds.) 1985a. Neogene of the Mediterranean and Paratethys. Stratigraphic Correlation Tables and Sediment Distribution Maps, vol. 1, XIV + 189 pp., vol. 2, XXVI + 436 pp. Vienna, Institute of Paleontology, University Vienna. Thenius, E. 1972. Grundzu ¨ ge der Verbreitungsgeschichte der Sa¨ugetiere. X + 345 pp. Jena, VEB G. Fischer Verlag. Thenius, E. 1979. Afrikanische Elemente in der miozaenen Saeugetierfauna Europas (African elements in the Miocene mammalian fauna of Europe). Annales Ge´ologiques Pays Helleniques, tom hors ser., fasc. III, 1201–8. Tobien, H. 1987. The position of the ‘Grande Coupure’ in the Paleogene of the Upper Rhine Graben and the Mainz Basin. In International Symposium on Mammalian Biostratigraphy and Paleoecology of the European Paleogene–Mainz, February 18th–21st 1987, N. Schmidt-Kittler (ed.). Mu ¨ nchner Geowissenschaftliche Abhandlungen, ser. A, 10, 197–202. Vinogradov, A. P. 1967–69. Atlas of the Lithological–Paleogeographical Maps of the USSR, 4 pts., 255 maps. Moscow, Ministry of Geology USSR & Academy of Sciences USSR. Ziegler, P. A. 1990. Geological Atlas of Western and Central Europe. Second edition, vol. 1, 239 pp.; vol. 2, 56 encl. Den Haag: Shell Int. Petroleum Maatschappij B. V.

3 Pliocene tephra correlations between East African hominid localities, the Gulf of Aden, and the Arabian Sea Peter B. deMenocal and Francis H. Brown

Introduction The Pliocene–Pleistocene chronology of hominid and other vertebrate evolution in East Africa is largely constrained by isotopic dating and regional intercorrelation of volcanic ash layers. Some eruptions were of suYcient magnitude or duration that their widespread tephra dispersal deWnes a series of dated marker horizons throughout the fossil-bearing sedimentary deposits of Tanzania, Uganda, Kenya, and Ethiopia (Brown, 1982; Feibel et al., 1989; Haileab & Brown, 1992, 1994; Pickford et al., 1991; WoldeGabriel et al., 1994). Although many of the larger eruptive events have been dated directly the ages of many tephra are only constrained by indirect stratigraphic interpolation between dated levels. The geochemical compositions of volcanic glasses from each eruption are unique, providing a deWnitive means to establish broad tephrostratigraphic correlations linking the regional climatic, tectonic, and biologic histories of this distinctive archive of Earth history. This same tephrostratigraphic approach has been used to extend the East African tephra correlations into the continuous and well-dated marine sediment record of regional and global paleoclimate variability (Brown et al., 1992; Sarna-Wojcicki et al., 1985). These authors identiWed several megascopic volcanic ash layers within Deep-Sea Drilling Project (DSDP) sites from the Gulf of Aden, nearly 1000 km northeast of hominid localities in Ethiopia and Kenya. Major element chemistries of volcanic glass shards extracted from these marine sediments were used to establish precise tephrostratigraphic correlations into the fossil-bearing East African sedimentary sequences. Moreover, controversy concerning the ages of speciWc eruptive events which then deWned key temporal junctures in early hominid evolution could be tested using the independent marine sediment chronostratigraphic framework (Brown et al., 1992; Sarna-Wojcicki et al., 1985). A more comprehensive eVort to establish more Pliocene–Pleistocene tephra correlations between terrestrial and marine sequences was hampered by incomplete and often disturbed core recovery at these early DSDP sites in the Gulf of Aden. Subsequent advances in deep-sea drilling technology and the return of scientiWc ocean drilling to the Arabian Sea present a new opportunity to establish tephrostratigraphic links between East African terrestrial and adjacent marine sedimentary sequences. Leg 117 of the Ocean Drilling Program drilled twelve sites oV the Omani margin and Arabian Sea (Fig. 3.1;

Palaeogeography of the circum-Mediterranean region

24

[Figure 3.1] Locations of Ocean Drilling Program (ODP) drill sites in the Arabian Sea.

Prell & Niitsuma, 1988; Prell et al., 1988) in an eVort to reconstruct the late Cenozoic history of the Asian monsoon and associated Arabian Sea upwelling. Coring of Pliocene–Pleistocene sediments was accomplished using an advanced hydraulic piston coring system which recovered complete and undisturbed sediments. In a marked improvement over earlier drilling efforts in the Gulf of Aden multiple holes were drilled at each site permitting the construction of complete, composite sedimentary sequences extending to the latest Miocene (deMenocal & Bloemendal, 1995; deMenocal et al., 1991; Murray & Prell, 1991). Subsequent study of Sites 721 and 722 from the Arabian Sea has produced long and continuous orbitally tuned time series of Pliocene–Pleistocene variations in regional aridity (Clemens & Prell, 1990, 1991; deMenocal, 1995; deMenocal & Bloemendal, 1995) and the strength of the Asian monsoon (Clemens & Prell, 1990, 1991; Clemens et al., 1996; Prell & Kutzbach, 1992). Although these sites are some 2000 km distant from East African source volcanoes, numerous discrete levels with enhanced ash shard abundances have been identiWed which are temporally correlative with known East African tephra layers (deMenocal & Bloemendal, 1995). The present study provides a comprehensive survey of tephra-

Pliocene tephra correlations

bearing levels within the Pliocene–Pleistocene sediments at Sites 721 and 722.

Research strategy The Site 721 and 722 sediments present an unique opportunity to place the East African sedimentary sequences, and the fossils they contain, within the context of the continuous and very detailed marine paleoclimatic record and chronostratigraphic framework. The research strategy is to use the oxide compositional signatures of volcanic glass shards extracted from Sites 721 and 722 to deWne these terrestrial–marine correlations. We employ a similar approach to that described by Sarna-Wojcicki & Davis (1991) which correlates tephra using glass chemistry, with conWrmation by sequence (i.e., the known temporal succession of eruptions). The challenge is that ash shards are extremely rare in these Site 721 and 722 sediments which are so distant from East African source volcanoes. However, once Wrmly established these tephrostratigraphic linkages present several new research opportunities. The correlations provide a direct way to compare the radioisotopic ages of East African tuVs with the orbitally tuned ages derived for the marine sediment record (e.g. McDougall et al., 1992; Renne et al., 1993; Walter, 1994). Additionally, the correlations link speciWc East African Xuviolacustrine stratigraphic sections to the marine paleoclimatic and paleoceanographic records, providing a direct means to assess past climatic linkages between the two regions. Finally, a full suite of tephra correlations will provide the means to directly test hypotheses linking Pliocene–Pleistocene changes in hominid and other African vertebrate evolution to regional changes in climate (Behrensmeyer et al., 1997; deMenocal, 1995; Grine, 1986; Vrba, 1985, 1995). In this paper we present electron microprobe major element oxide data for Wve tephra horizons at Sites 721 and 722 between 4.0 and 3.5 Ma, and establish correlations to East African tuVs and Gulf of Aden tephra. Approximately 25 tephra-rich layers can be analysed within the Pliocene–Pleistocene interval at Sites 721 and 722.

Site locations and regional climatology Sites 721 (16° 40.6
N, 59° 51.9'E, 1944 m) and 722 (16° 37.3
N, 59° 47.7'E, 2028 m) were both drilled near the crest of the Owen Ridge in the Arabian Sea (Fig. 3.1), separated by approximately 20 km. The ridge setting was selected speciWcally to recover records of pelagic sedimentation in the

25

Palaeogeography of the circum-Mediterranean region

26

Arabian Sea to record past variations in regional climate and surface ocean changes associated with the late Cenozoic evolution of the Indian monsoon. The relative shallowness of these sites places them above the calcite lysocline and precludes any signiWcant contribution from downslope processes associated with the Indus fan. Pliocene–Pleistocene sedimentation rates average ~ 3–4 cm/kyr based on preliminary biostratigraphic age control (Prell & Niitsuma, 1988) and sediments are predominantly composed of nannofossil ooze with varying concentrations of terrigenous clay and silt. Monsoonal circulation, such as that associated with the African, Indian, and Southeast Asian monsoonal regions, results as a consequence of the diVering heat capacities of land and water. Sensible heating warms land surfaces much more rapidly than an ocean mixed layer (Hastenrath, 1985). During the winter months the South Asian landmass cools more eYciently than the adjacent North Indian Ocean and a broad high pressure cell develops over Siberia and the Tibetan Plateau. Dry and variable northeast trade winds develop over the southeast Asian region, including the Arabian Sea from October to April (Fig. 3.2a). Sensible heating during the Northern Hemisphere summer promotes the development of a strong low pressure cell over South Asia that establishes very strong regional cyclonic circulation over South Asia from May to September. Strong (15 m/s) moisture-laden southwest winds (Somali Jet) parallel the East African, Arabian, and Omani coasts during the summer, bringing monsoon rains to southern Asia (Fig. 3.2b). As a dynamical response to the surface wind stress Weld, cool, nutrient-rich waters upwell oV Arabia and Oman and support high surface ocean productivity in the Arabian Sea. Atmospheric sampling, satellite images, and sediment trap studies have demonstrated that wind-borne mineral grains (eolian dust) are entrained from Mesopotamian, Arabian, and northeast African sources during the peak months of the summer monsoon: June, July, and August (Clemens et al., 1996; Nair et al., 1989; Pye, 1987; Sirocko, 1989; Sirocko & Sarnthein, 1989). Sediment trap data from the Arabian Sea indicate that 80% of the annual terrigenous Xux to the western Arabian Sea occurs during the summer months (Nair et al., 1989). Geochemical and mineralogic data indicate that the Mesopotamian Xoodplains are by far the largest source of eolian dust to the Arabian Sea, although minor contributions from East Africa are also indicated (Sirocko, 1989; Sirocko et al., 1996; Sirocko & Sarnthein, 1989). Given their trajectory, strength, and persistence, the summer SW monsoon winds are an eYcient vector for transporting East African volcanic ash shards to the Arabian Sea (Fig. 3.2b).

Pliocene tephra correlations

27

[Figure 3.2] Boreal winter (a) and summer (b) surface winds over northeast Africa and the Arabian Sea.

Pliocene–Pleistocene variability of regional climate and the Indian monsoon The sediments of Site 721 and 722 have been the focus of much research into the late Cenozoic evolution of the Indian monsoon and regional climate. Initial work at Site 722 demonstrated that large changes in regional aridity were linked to the Pleistocene succession of glacial–interglacial cycles (Clemens & Prell, 1990, 1991). Eolian dust Xuxes to Site 722 during glacial maxima were three to Wve times higher than observed for interglacial periods; these low-latitude aridity cycles were observed to be directly in-phase with the oxygen isotopic record of high-latitude glacial–interglacial cycles. Parallel analysis of the terrigenous (eolian) grain size variations at Site 722, which monitor variations in the strength of the summer SW monsoon winds, further demonstrated that the Indian monsoon intensity was closely linked to

Palaeogeography of the circum-Mediterranean region

28

variations in summer insolation resulting from orbital precession (at the 23–19 kyr periodicity). This Wrst-order relation between sensible heating variations (forcing due to orbital variations in seasonal radiation distribution) and the strength of the summer monsoon circulation (the climate response) has been a focus of many atmospheric climate model studies (e.g. Kutzbach, 1981; Kutzbach & Guetter, 1986; Prell & Kutzbach, 1987). Analysis of the full Pliocene–Pleistocene interval (last c. 5 Ma) at Sites 721 and 722 has produced new perspectives on the evolution of regional climate associated with the initial onset and subsequent growth of high-latitude glacial cycles after c. 2.8 Ma. Study of eolian dust variations at Sites 721 and 722, as well as analysis of six other sites in the Gulf of Aden and oV subtropical West Africa have demonstrated that the onset of large amplitude regional aridity cycles was closely linked to the development of high-latitude glacial cycles (Bloemendal & deMenocal, 1989; Clemens et al., 1996; deMenocal, 1995; deMenocal & Bloemendal, 1995; deMenocal et al., 1991, 1993; Tiedemann et al., 1994). Several features of the Pliocene–Pleistocene evolution of subtropical African aridity variations are common to all of these sites (deMenocal, 1995). Prior to 2.8 Ma subtropical African aridity varied at the 23–19 kyr period associated with low-latitude radiation forcing of monsoonal climate, whereas after 2.8 Ma African aridity followed the longer 41 kyr and then 100 kyr periods associated with glacial–interglacial cycles of the late Pliocene and Pleistocene, documenting the post-2.8 Ma regulation of this low-latitude climate system by high-latitude glacial climates. Distinct shifts in African eolian variability were observed at 2.8 Ma, 1.7 Ma, and 1.0 Ma, each tied to coeval shifts in high-latitude climate. This sensitivity of African climate to high-latitude glacial boundary conditions has been documented using general circulation model experiments (Clemens et al., 1991; deMenocal & Rind, 1993; Kutzbach & Guetter, 1986; Prell & Kutzbach, 1987). Based on analysis of the terrigenous (eolian) grain size record over the last 3.5 Ma at Site 722, Clemens et al. (1996) documented discrete shifts in the intensity and phase of the Indian monsoon at 2.6 Ma, 1.7 Ma, and 1.2 Ma, and 0.6 Ma, further emphasizing the importance of high- and lowlatitude climate linkages throughout the Pliocene and Pleistocene.

Analytical methods A primary objective of the present study is to document a continuous Pliocene–Pleistocene history of explosive volcanism in East Africa through the stratigraphic record of tephra deposition in the Arabian Sea. Consequently, we employed a sampling strategy whereby the full Site 721/722

Pliocene tephra correlations

composite sequence spanning the last 4.5 Ma (c. 135 m; see Table 3.1) was continuously scrape sampled across the core diameter at 30 cm intervals using a TeXon spatula, yielding approximately 2–3 g of dry sediment. Using an average Pliocene–Pleistocene sedimentation rate of 3.2 cm/kyr; (Clemens et al., 1996; Murray & Prell, 1991), this sample interval is equivalent to ~ 10 kyr, our nominal temporal resolution in this study. Samples were freeze-dried, weighed, and then decarbonated using 100 ml 1M acetic acid and placed on a shaker table for one hour. After settling, samples were decanted and washed once with deionized water, and then wet sieved at 38
m. The  38
m fraction was saved, freeze-dried, and weighed. Grain mount slides were prepared for all 380 samples using Permount medium and large area (50 ; 25 mm) coverslips. Optically isotropic ash shards were identiWed and counted to obtain semi-quantitative estimates for shard abundance, which ranged between 0–1800 shards per slide. Tephra-bearing layers were never visibly evident from the split core and shards were generally extremely rare except for six levels where shards were very abundant. Median shard sizes were approximately 40–60
m which proved to be too small for manual extraction. Shard morphology was dominated by clear, arcuate fragments with the larger shards having welldeWned bubble-wall junctions and rare cylindrical vesicles. Samples with highest shard abundances were further concentrated using a sodium polytungstate (Na2WO4) heavy-liquid separation step ( = 2.2 g/cc) to separate ash shards and other detrital mineral grains (  2.5 g/cc) from lighter but very abundant diatoms and radiolaria ( = 1.9–2.2 g/cc). Samples were added to a 2.2 g/cc Na2WO4 solution, sonicated, then centrifuged for 3 minutes. The basal 1 cm of the centrifuge tube (now containing the shards and other detrital mineral grains) was frozen in liquid nitrogen and remaining supernatant (containing diatoms and radiolaria) was decanted and the tube walls were rinsed with deionized water. After thawing the shardbearing Na2WO4 slurry was then captured and rinsed using a 0.2
m Wlter. This technique proved to be very eVective in concentrating the shard abundances and removing the biogenic components; the Wnal concentrate typically comprised  1% of the original dry bulk sediment mass. After drying these samples were then mounted for microprobe analysis. Samples were analysed on a Cameca SX-50 electron microprobe equipped with four wavelength-dispersive spectrometers. The accelerating voltage was 15 kV, the beam current 25 nA, and the beam diameter between 5 and 25
m. Elemental concentrations were calculated from relative peak intensities using the (z) algorithm (Pouchou & Pichoir, 1991). The standard for O, Si, Al, and K was natural obsidian. The remaining elements were standardized using minerals and synthetic oxides. In order to assure an

29

Palaeogeography of the circum-Mediterranean region

30

Table 3.1. Site 721 and 722 composite section

From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To From To

Site

Hole

Core

Type

Sect.

Int.

mbsf

722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 722 721 721 722 722 722 722 722 722 722 722 721 721 722 722 721 721 721 721 722 722 721 721 722 722

B B A A B B A A B B A A B B A A B B A A B B A A A A A A B B A A B B C C B B C C C C A A C C A A

1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 6 6 6 6 6 6 7 7 8 8 8 8 9 9 10 10 10 10 11 11 12 12 11 11 13 13 12 12

H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H X X X X X X X X X X

1 4 3 3 1 6 2 4 1 6 2 4 1 6 3 4 1 7 3 4 1 7 3 7 4 5 1 4 1 6 3 4 1 7 2 2 1 7 3 5 1 4 3 6 1 5 3 5

5 50 60 140 30 45 95 35 40 20 65 100 70 90 15 55 65 15 70 125 95 15 75 20 75 45 20 85 50 105 10 105 70 55 5 145 35 35 10 80 50 95 70 135 140 100 10 130

0.05 5.00 3.60 4.40 5.80 13.45 12.25 14.65 15.50 22.80 21.55 24.90 25.40 33.10 32.15 34.05 34.95 43.45 42.30 44.35 44.85 53.05 51.95 57.40 53.75 54.95 57.70 62.85 63.40 71.45 70.30 72.75 73.20 82.05 81.85 83.25 83.05 92.05 93.00 96.70 100.00 104.95 99.90 105.05 110.50 116.10 109.00 113.20

adj (m) 0.00 0.00 1.45 1.45 0.10 0.10 1.35 1.35 0.55 0.55 1.85 1.85 1.40 1.40 2.40 2.40 1.55 1.55 2.75 2.75 2.30 2.30 3.45 3.45 7.15 7.15 4.45 4.45 3.95 3.95 5.15 5.15 4.75 4.75 5.05 5.05 5.30 5.30 4.40 4.40 1.15 1.15 6.25 6.25 0.85 0.85 8.00 8.00

mcd 0.05 5.00 5.05 5.85 5.90 13.55 13.60 16.00 16.05 23.35 23.40 26.75 26.80 34.50 34.55 36.45 36.50 45.00 45.05 47.10 47.15 55.35 55.40 60.85 60.90 62.10 62.15 67.30 67.35 75.40 75.45 77.90 77.95 86.80 86.90 88.30 88.35 97.35 97.40 101.10 101.15 106.10 106.15 111.30 111.35 116.95 117.00 121.20

Pliocene tephra correlations

Table 3.1. (cont)

From To From To From To

31

Site

Hole

Core

Type

Sect.

Int.

mbsf

adj (m)

mcd

721 721 721 721 721 721

C C B B C C

14 14 14 14 15 15

X X X X X X

2 5 2 6 3 6

75 60 135 130 70 130

120.95 125.30 127.25 133.20 132.00 137.10

0.30 0.30 − 1.60 − 1.60 − 0.35 − 0.35

121.25 125.60 125.65 131.60 131.65 136.75

mbsf = meters below seafloor. adj = depth adjustment in meters required to build composite. mcd = meters composite depth.

equivalent thickness of the C coating, the obsidian standard and unknowns were coated simultaneously. Details of the analytical method are provided in Nash (1992). Raw elemental percentages were normalized to 100% to remove the eVects of variable hydration. Average and standard deviation chemical data for samples from Site 721/722 (Arabian Sea) samples, Site 231/232 (Gulf of Aden), and Turkana Basin tephra are presented in Tables 3.2, 3.3, and 3.4, respectively. The Gulf of Aden chemical data were taken directly from the primary reference (Sarna-Wojcicki et al., 1985), whereas the Turkana Basin chemical data (Table 3.4) were dervied from multiple shard averages of the sample (Brown et al., 1992; Haileab & Brown, 1992, 1994; Sarna-Wojcicki et al., 1985). Original totals (To) are given to indicate degree of shard hydration and to permit recalculation to original raw percentages. Analytical standard deviations for SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, and CaO using our internal standards were 0.5, 0.01, 0.2, 0.03, 0.005, 0.01, and 0.02 weight percent, respectively.

Results Thirty-nine samples contained suYcient shards for subsequent concentration, presenting between 25 discrete, analytically viable ash layers. Here, we focus on analyses of Wve tephra-bearing levels between 4.0–3.4 Ma. Several East African tuVs have been dated by K/Ar and 40Ar/39Ar methods within this interval to very high precision (McDougall et al., 1992; Renne et al., 1993; Walter, 1994; Walter & Aronson, 1993). Additionally, many tephra layers within this interval have been correlated between stratigraphic sequences in Kenya and Ethiopia and have been found also within the Gulf of Aden marine sediments. Finally, there exist several tephra horizons within this

N

SiO2

0.50 0.58 0.06 0.37 1.34 0.88 0.29

sd

0.14 0.16 0.21 0.19 0.23 0.30 0.24

TiO2

0.01 0.04 0.01 0.05 0.08 0.03 0.04

sd 12.74 11.48 11.11 10.99 10.42 10.57 10.22

Al2O3 0.20 0.23 0.03 0.10 0.24 0.24 0.30

sd 1.76 2.61 2.92 3.27 4.57 4.64 4.61

Fe2O3 0.03 0.07 0.14 0.06 0.20 0.13 0.47

sd 0.08 0.07 0.11 0.09 0.13 0.17 0.17

MnO 0.01 0.04 0.02 0.04 0.03 0.04 0.03

sd 0.06 0.03 0.01 0.04 0.03 0.03 0.02

MgO 0.01 0.01 0.00 0.02 0.01 0.03 0.02

sd 0.33 0.19 0.23 0.22 0.22 0.28 0.21

CaO 0.02 0.01 0.00 0.05 0.01 0.06 0.02

sd 94.49 95.67 96.07 95.03 96.42 96.18 90.67

Total

N

30 15 15 14 15 15

Sample

231-19-2 (33-34) 231-21-2-(15-17) 231-22-1-(82-85) 232A-1-4 (42-450 231-20-2 (25-28) 232A-1-5 (25-27)

76.43 76.87 76.68 75.73 76.65 75.39

SiO2

0.6 0.68 0.54 0.66 0.66 0.79

sd

0.14 0.13 0.19 0.20 0.23 0.24

TiO2

0.01 0.02 0.03 0.05 0.03 0.04

sd 12.50 11.44 11.28 10.61 10.62 10.53

Al2O3 0.23 0.24 0.25 0.57 0.35 0.45

sd 1.57 2.42 2.73 3.97 4.06 4.14

Fe2O3 0.06 0.06 0.06 0.18 0.35 0.37

sd 0.31 0.18 0.21 0.19 0.19 0.19

MnO

0.06 0.01 0.01 0.02 0.02 0.02

sd

0.05 0.07 0.08 0.16 0.16 0.16

MgO

0.01 0.01 0.01 0.02 0.02 0.02

sd

0.31 0.18 0.21 0.19 0.19 0.19

CaO

0.06 0.01 0.01 0.02 0.02 0.02

sd

94.31 93.20 92.94 92.89 92.48 92.68

Total

Table 3.3. Electron probe analysis of volcanic glass shards separated from Gulf of Aden Sites 231 and 232 (Sarna-Wojcicki et al., 1985, 1059)

722A 11-1 (30-60) 1 76.97 721C 13-3 (90-120) 6 76.12 722A 12-4 (90-120) 2 76.01 722A 11-4 (60-90) 14 75.62 722A 11-4 (60-90) 6 75.99 722A 12-4 (90-120) 11 75.83 722A 11-5 (60-90) 20 74.84

Sample

Table 3.2. Electron probe analysis of volcanic glass shards separated from Arabian Sea Sites 721 and 722

N

5 27 15 15 11 20 11

Sample

82-869 K81-485 81-602 80-295 80-295 83-1ANU 82-742

76.80 77.91 77.03 77.43 77.33 74.48 77.85

SiO2

0.21 0.43 0.55 0.80 0.61 0.62 0.68

sd

0.16 0.14 0.16 0.18 0.21 0.33 0.25

TiO2

0.06 0.02 0.04 0.04 0.05 0.06 0.03

sd 12.73 11.42 11.19 11.12 10.41 11.11 10.62

Al2O3 0.10 0.26 0.22 0.19 0.23 0.21 0.24

sd 1.69 2.48 2.81 3.19 4.43 4.56 4.62

Fe2O3 0.06 0.06 0.06 0.10 0.06 0.33 0.12

sd 0.05 0.08 0.10 0.11 0.15 0.17 0.19

MnO 0.01 0.03 0.02 0.01 0.03 0.03 0.02

sd

Table 3.4. Electron probe analysis of volcanic glass shards separated from Turkana Basin tuVs

0.06 0.03 0.00 0.03 0.02 0.00 0.02

MgO 0.01 0.02 0.00 0.02 0.01 0.02 0.01

sd 0.30 0.20 0.22 0.19 0.20 0.30 0.21

CaO 0.02 0.02 0.01 0.02 0.02 0.02 0.03

sd

93.09 92.49 93.79 91.45 90.60 95.31 95.48

Total

Palaeogeography of the circum-Mediterranean region

34

interval that have not been dated directly and reported ages of these key marker horizons are only constrained by stratigraphic interpolation. Thus, the several tephra layers within this interval provide an excellent opportunity to test the tephrostratigraphic and chronologic precision of these terrestrial–marine correlations.

Tephra abundances and the Site 721/722 age model The volcanic ash shard abundance record from the Site 721/722 composite sequence spanning 0.5–4.5 Ma is shown in Fig. 3.3. Also shown in this Wgure is the eolian dust record (from these same sites) presented in deMenocal (1995). The temporal sampling resolution of the 721/722 eolian record is 1.5 kyr. We identify approximately 25 discrete ash layers (where shard abundances exceeded 20 shards per slide to maximum values reaching ~ 2000 shards per slide) between 0.5–4.5 Ma. There were many additional levels where ash shards were present (5–20 shards/slide) but these concentrations were judged from experience to be too low to recover suYcient material for analysis. Focusing on the 4.5–3.0 Ma interval (90–140 mcd) we Wnd approximately 11 discrete ash layers (Fig. 3.4). We can estimate the ages of these ash layers by referring to the orbitally tuned timescale previously developed for this site (deMenocal, 1995; deMenocal & Bloemendal, 1995) (Fig. 3.5). As a brief review, Site 721/722 eolian dust percentages were phase-locked to an orbital composite target signal comprised of 0.66 weighting of orbital precession and 0.34 weighting of orbital obliquity, both of which were aligned for maxima in boreal summer insolation (deMenocal, 1995; deMenocal & Bloemendal, 1995). The rationale for this approach was that early Pliocene dust and upwelling indices suggested that maxima in eolian dust and upwelling indices appeared to be indicating maxima in summer monsoon intensity which climate models indicate should coincide with orbital precession index (esin ()) minima, and orbital tilt index maxima (Berger & Loutre, 1991; deMenocal, 1995). The Site 721/722 eolian time series is shown adjacent to the orbital insolation tuning target for the 3.0–4.5 Ma interval in Fig. 3.5. Note the very close correspondence (correspondingly high coherency) between the relative amplitude variations of the orbital forcing and paleoclimate (aridity) response for this interval. As shown by Clemens et al. (1996) the phase of the Indian monsoon response to insolation forcing appears to have been non-stationary over the Pliocene–Pleistocene interval, showing clear phase jumps with respect to a ‘phase-stationary’ orbital target. These phase jumps require that either high-latitude glacial ice volume and/or the low-latitude monsoon shifted in

[Figure 3.3] Volcanic ash shard abundance record derived for the Site 721/722 composite sedimentary sequence between 0.5–4.5 Ma shown adjacent to the Site 721/722 eolian dust record (deMenocal, 1995). Approximately 25 discrete tephra-rich ash layers are identified for this 0.5–4.5 Ma interval. Depths are in meters, composite depth (mcd) based on the composite table shown in Table 3.1. The interval between 30 and 46 mcd (c. 0.8–1.2 Ma) was not sampled in this study.

Palaeogeography of the circum-Mediterranean region

[Figure 3.4] Volcanic ash abundances and eolian dust percentage at Sites 721/722 for the 3.0–4.5 Ma interval. The eolian timeseries shown was derived from the originally published orbitally tuned timescale (deMenocal, 1995; deMenocal & Bloemendal, 1995). Details of the orbital tuning precision for this interval are shown in Fig. 3.5.

their phase response relative to orbital radiation forcing by c. 5–15 kyr at key times (2.6 Ma, 1.7 Ma, and 1.2 Ma, and 0.6 Ma,) during the Pliocene– Pleistocene. This means that the absolute oxygen isotopic chronology may have a time-varying oVset of 5–15 kyr with respect to sideral time, an unresolved problem which potentially applies to our timeseries as well. This oVset is small with respect to our chronostratigraphic resolution issues here;

Pliocene tephra correlations

[Figure 3.5] The orbitally tuned timescale developed for the Site 721/722 eolian timeseries for the 3.0–4.5 Ma interval (deMenocal, 1995; deMenocal & Bloemendal, 1995). Note the close match between the relative amplitudes of the orbital radiation forcing target (Berger & Loutre, 1991) and the eolian (aridity) paleoclimate response over this interval. Although the precision of the tuning process is ± 5 kyr, remaining uncertainty surrounding the precise phase of the monsoon response to insolation forcing (Clemens et al., 1996) introduces an additional uncertainty of 5–15 kyr.

however, there are larger paleoclimatic and paleoceanographic implications (Clemens et al., 1996).

Chemical correlation results of selected tephra layers between 3.4–4.0 Ma Five tephra-bearing samples from the 3.4–4.0 Ma interval at Sites 721 and 722 were selected for subsequent concentration, mounting, and analysis by electron microprobe. Between six and twenty shards were analysed per sample with several determinations per shard. Averaged chemical data (normalized to 100% to correct for variable hydration), standard deviations, and original totals (To) for these samples are listed in Table 3.2. Oxide crossplots for these Wve Arabian Sea samples are shown in Figs. 3.6–3.10 along with oxide data from Turkana Basin tuVs and analyses on extracted Gulf of Aden tephra (Brown et al., 1992; Sarna-Wojcicki et al., 1985). We use the Similarity CoeYcient (SC) as a means to quantify the chemi-

37

Palaeogeography of the circum-Mediterranean region

38

Table 3.5. Similarity CoeYcients between Site 721/722, Turkana Basin, and Site 231/2 tephra chemistries (5-elements) Sample

SC

WRT tuff . . .

Tuff Name

SC

WRT Sites 231/2 . . .

722A 11-1 (30-60) 721C 13-3 (90-120) 722A 12-4 (90-120) 722A 11-4 (60-90) 722A 11-4 (60-90) 722A 12-4 (90-120) 722A 11-5 (60-90)

0.95 0.95 0.92 0.95 0.95 0.95 0.98

82-869 K81-485 83-1ANU 80-295 80-295 81-602 82-742

 Tulu Bor Wargolo Moiti, lo-Fe Lokochot, lo-Fe Lokochot, hi-Fe Moiti, hi-Fe Lomogol

0.96 0.92 0.94

231 19-2 (33-34) 231 21-2 (15-17) 231 22-1 (82-85)

0.94

231 20-2 (25-28)

0.95

232A 1-5 (25-27)

cal similarity between an analysed tephra and other candidate tephra matches (Borchardt et al., 1972). The SC is computed as: SC = 1/n ·   (xn − x0 )  where xn and x0 are the separate oxide compositions for the sample tephra and the comparison target tephra, and oxide diVerences are summed for n-oxides (Wve oxides in our case). We computed SC values using the following Wve oxides: SiO2, TiO2, Al2O3, Fe2O3, and CaO (Brown et al., 1992; SarnaWojcicki et al., 1985). The SC statistic equals unity for samples with identical chemical compositions. In practice the SC values can range between 1.00– 0.97 for replicate analyses whose compositions are within one standard deviation of the mean composition. In all cases, the tuV correlations discussed below represented the best compositional match to the speciWed East African tuV.

Sample 722A 12-4 (90–120 cm) – Moiti Tuff (SC = 0.92 and 0.95) The Wve oxide bivariate plots for sample 722A 12-4 (90–120 cm) are shown in Fig. 3.6, and these data are compared to published oxide compositions for other East African tuVs of this approximate age range, including the Moiti, Wargolo, Lomogol, Lokochot, and -Tulu Bor TuVs (Brown, 1982; Brown et al., 1992; Sarna-Wojcicki et al., 1985). We found that shards within this sample fell into bimodal high-Fe (n = 11) and low-Fe (n = 2) compositions. The Moiti TuV was the only tuV in this age range (c. 4.0 Ma) which had this dual low-Fe (K81-602, Turkana Basin) and high-Fe (83-1 ANU, Turkana Basin) composition (Tables 3.2–3.4) and closely matched our sample for all Wve oxides. The computed SC values for the low- and high-Fe Moiti compositions were 0.95 and 0.92, respectively (Table 3.5, Fig. 3.6). Only two shards were analysed for the low-Fe Moiti tephra within 722A 12-4 (90–120 cm) and this may result in a lower precision SC estimate (SC = 0.92). The Wve oxide

Pliocene tephra correlations

39

[Figure 3.6] Crossplots of major element oxide percentage compositions for high-Fe and low-Fe endmembers of Moiti Tuff volcanic glass shards from the Turkana Basin, Kenya (closed circles), Gulf of Aden (open circles), and the Arabian Sea (open triangles). All data were recalculated to 100% on a fluid-free basis (Tables 3.2–3.4); error bars represent one standard deviation.

Palaeogeography of the circum-Mediterranean region

40

[Figure 3.7] Crossplots of major element oxide percentage compositions for Wargolo Tuff volcanic glass shards from the Turkana Basin, Kenya (closed circles), Gulf of Aden (open circles), and the Arabian Sea (open triangles). All data were recalculated to 100% on a fluid-free basis (Tables 3.2–3.4); error bars represent one standard deviation.

Pliocene tephra correlations

41

[Figure 3.8] Crossplots of major element oxide percentage compositions for Lomogol Tuff volcanic glass shards from the Turkana Basin, Kenya (closed circles), Gulf of Aden (open circles), and the Arabian Sea (open triangles). All data were recalculated to 100% on a fluid-free basis (Tables 3.2–3.4); error bars represent one standard deviation.

Palaeogeography of the circum-Mediterranean region

42

[Figure 3.9] Crossplots of major element oxide percentage compositions for high-Fe and low-Fe endmembers of Lokochot Tuff volcanic glass shards from the Turkana Basin, Kenya (closed circles), Gulf of Aden (open circles), and the Arabian Sea (open triangles). All data were recalculated to 100% on a fluid-free basis (Tables 3.2–3.4); error bars represent one standard deviation.

Pliocene tephra correlations

43

[Figure 3.10] Crossplots of major element oxide percentage compositions for -Tulu Bor Tuff volcanic glass shards from the Turkana Basin, Kenya (closed circles), Gulf of Aden (open circles), and the Arabian Sea (open triangles). All data were recalculated to 100% on a fluid-free basis (Tables 3.2–3.4); error bars represent one standard deviation.

Palaeogeography of the circum-Mediterranean region

44

compositions for sample 722A 12-4 (90–120 cm) match those for the low-Fe (K81-602) and high-Fe (83-1 ANU) Moiti compositions within one-sigma analytical variability (Fig. 3.6). Additionally, the low-Fe Moiti composition was also identiWed within the Gulf of Aden samples at Site 231 (Sample 231 22-1 (82 cm); Sarna-Wojcicki et al., 1985). The Gulf of Aden oxide compositions are also plotted on Fig. 3.6, closely matching (SC = 0.94; Table 3.5) those from the Arabian Sea. The orbitally tuned age for the Moiti TuV in the Arabian Sea sequence is 3.96 ± 0.01 Ma (Fig. 3.5, Table 3.6).

Sample 721C 13-3 (90–120 cm) – Wargolo Tuff (SC = 0.95) The oxide plots for this sample are shown compared to the East African tuV compositions in Fig. 3.7. This sample is distinguished by its intermediate Fe, Si, and Al oxide composition and lower Ca and Ti oxide percentages. Based on an average of six shard analyses this sample most closely matches (SC = 0.95; Tables 3.2–3.5) the Wargolo TuV (K81-485; Tables 3.2–3.4). The Wargolo TuV was also identiWed within the Gulf of Aden sediments (231 21-2 (33–34 cm) and its oxide composition also closely matches (SC = 0.92; Table 3.5) the Arabian Sea data (Fig. 3.7). The Wargolo TuV oxide compositions are readily distinguished from the stratigraphically underlying (older) Cindery TuV (3.85 ± 0.08 Ma; White, 1993) which has characteristically much higher Al2O3, TiO2, CaO percentages relative to the Wargolo composition (Brown et al., 1992; Haileab & Brown, 1992). The orbitally tuned age for the Wargolo TuV in the Arabian Sea sequence is 3.80 ± 0.01 Ma (Fig. 3.5, Table 3.6).

Sample 722A 11-5 (60–90 cm) – Lomogol Tuff (SC = 0.98) Average oxide compositions for twenty shards from the Arabian Sea sample 722A 11-5 (60–90 cm) are shown compared to the East African TuV compositions in Fig. 3.8. The best compositional match for this sample is with the Lomogol TuV (SC = 0.98, K82-742; Tables 3.2–3.4), as evidenced by its distinctive high-Fe and low Ca and Al oxide composition (Table 3.5, Fig. 3.8). The Gulf of Aden oxide data shown in Fig. 3.8 have been correlated to the Lomogol TuV (Brown et al., 1992; Sarna-Wojcicki et al., 1985). This composition closely matches the Arabian Sea Lomogol-equivalent (SC = 0.95; Tables 3.2–3.5). The Arabian Sea and Gulf of Aden SiO2 data are lower than those for the Turkana Basin sample K82-742 but it should be noted that SiO2 data for this and other tuVs can be highly variable between samples due, in part, to higher relative percentage error and signiWcant intersample SiO2 variability. The Lomogol oxide composition is very close to the high-Fe Lokochot composition which commonly confounds eVorts to distinguish between

Pliocene tephra correlations

Table 3.6. Comparison of orbitally tuned and radiometric ages of East African tuVs between 3.4 Ma and 4.0 Ma 721/722 Orbitally tuned Age (Ma)

Kenyan/Ethiopian sites Radiometric Age (Ma)

Reference

-Tulu Bor/SHT

3.41 ± 0.01 Ma

Wargolo/VT-1

3.80 ± 0.01 Ma

Moiti/VT-3

3.96 ± 0.01 Ma

3.39 ± 0.04 Ma 3.41 ± 0.01 Ma 3.80 ± 0.05 Ma 3.75 ± 0.02 Ma O 4.10 ± 0.07 Ma 3.89 ± 0.02 Ma 3.94 ± 0.04 Ma

(White, 1993) (Walter and Aronson, 1993) (Haileab and Brown, 1992) (White, 1993) (McDougall, 1985) (White, 1993) (White et al., 1994)

Tuff

these two compositionally similar tuVs. However, the younger Lokochot TuV has a characteristic bimodal high-Fe and low-Fe composition which, when both compositions are present in a single sample, can be used diagnostically (see below). The orbitally tuned age for the Lomogol TuV in the Arabian Sea sequence is 3.62 ± 0.01 Ma (Fig. 36).

Sample 722A 11-4 (60–90 cm) – Lokochot Tuff (SC = 0.95) The oxide compositions for Arabian Sea sample 722A 11-4 (60–90 cm) are shown compared to the East African tuV compositions in Fig. 3.9. Oxide analyses from this sample identiWed separate low-Fe and high-Fe endmember compositions (Fig. 3.9, Table 3.2) These two compositions closely match the oxide compositions for the high-Fe (SC = 0.95, n = 6) and low-Fe (SC = 0.95, n = 14) compositions of the Lokochot TuV (Fig. 3.9, Tables 3.2– 3.5). As noted above, the high-Fe Lokochot is compositionally similar to the Lomogol TuV, but we correlate these Arabian Sea tephra to the Lokochot TuV because of the dual high-Fe and low-Fe compositions and its proper stratigraphic position above (younger than) the Lomogol TuV (Brown et al., 1992; Haileab & Brown, 1992). The Gulf of Aden data for the high-Fe end member of the Lokochot TuV (sample 231 20-2 (25–28 cm)) also match the Arabian Sea oxide data (SC = 0.94; Fig. 3.9, Table 3.5). The orbitally tuned age for the Lokochot TuV in the Arabian Sea sequence is 3.57 ± 0.01 Ma (Fig. 3.5, Table 3.6). Sample 722A 11-1 (30–60 cm) – -Tulu Bor Tuff (SC = 0.95) The oxide compositions for Arabian Sea sample 722A 11-1 (30–60 cm) are shown compared to the East Africa tuV compositions in Fig. 3.10. Although

45

Palaeogeography of the circum-Mediterranean region

46

[Figure 3.11] Comparison of radiometric and stratigraphic (interpolated) ages for East African tuffs between 4.0–3.4 Ma and their orbitally tuned ages derived from the marine sediment chronostratigraphy at Sites 721 and 722 in the Arabian Sea. Tephra correlations between East Africa, the Gulf of Aden, and the Arabian Sea sediments were accomplished using major element oxide compositions (Figs. 3.6–3.10; Tables 3.2–3.5). Note that the radiometric and orbitally tuned ages agree to within their joint error, or generally within 0.01 Ma. New, orbitally tuned ages can be applied to the Lokochot and Lomogol Tuffs for which direct radiometric age determinations have not yet been possible. Placement of the fossil hominid cranium A.L. 417 (Australopithecus afarensis) within the context of the eolian dust record at Site 721/722 was constrained by its stratigraphic position relative to the Sidi Hakoma Tuff (SHT, = -Tulu Bor).

Pliocene tephra correlations

we were only able to analyse a single shard from this level (Table 3.2), its composition was suYciently diagnostic (very low Fe and Ti, high Al and Ca oxide values) to make a tentative correlation to the -Tulu Bor TuV (SC = 0.95; Table 3.5). This is the only East African tuV within this time interval that has such a composition. The -Tulu Bor TuV was also detected in the Gulf of Aden sediments (Sample 231 19-2 (33–34 cm)) and those oxide data are shown in Fig. 3.10 (Sarna-Wojcicki et al., 1985). The Gulf of Aden data from sample 231 19-2 (33–34 cm) also match the Arabian Sea oxide data (SC = 0.96; Table 3.5).The orbitally tuned age for the -Tulu Bor TuV in the Arabian Sea sequence is 3.41 ± 0.01 Ma (Fig. 3.5, Table 3.6).

Discussion Tephrostratigraphic links between East Africa, the Gulf of Aden, and the Arabian Sea Despite very low volcanic shard abundances in these Arabian Sea samples we have been able to extract, concentrate, and analyse them by electron microprobe and establish oxide correlations to known East African tuVs. For each of the Wve tephra presented in Figs. 3.6–3.10 the geochemical data were suYciently diagnostic to establish Wrm correlations to speciWc East African tuV compositions with Similarity CoeYcient values ranging between 0.92 and 0.98. Additionally, we used the Squared Chord Distance parameter (D2) as deWned by Perkins et al. (1995) as a more quantitative means to evaluate the oxide similarity between an analysed tephra and other candidate match tephras. The advantage of this D2 parameter is that it employs the mean oxide data as well as their associated analytical standard deviations to assign statistical signiWcance levels to a proposed tephra correlation, since D2 has a Chi-square distribution among compositionally identical samples (Perkins et al., 1995, 1998). Calculating this chord distance parameter for our analyses we found that the -Tulu Bor TuV correlation was signiWcant at the 99% level, the Moiti Lokochot TuV correlations were signiWcant at the 95% level, and the Wargolo and Lomogol TuV correlations were signiWcant between the 90% and 95% conWdence levels. Using these tephrostratigraphic correlations we can now compare directly the radiometric ages obtained from land tephra to the orbitally tuned ages of correlative tephra preserved in the marine sediments. Additionally, we can use the correlations to assign new, orbitally tuned ages to those tuVs which have not been radiometrically dated and are only presently constrained by stratigraphic age interpolations.

47

Palaeogeography of the circum-Mediterranean region

48

Comparison of radioisotopic and orbitally tuned tephra ages The eolian dust record at Sites 721/722 is constrained by an orbitally tuned chronostratigraphy with an approximate tuning precision of ± 5 kyr (see deMenocal, 1995) The absolute error of the chronology may deviate from sideral time by up to 5–15 kyr due to remaining uncertainties of the phase of this record with respect to orbital forcing (Clemens et al., 1996). The nominal temporal resolution of the tephra extraction samples from Sites 721/722 was 10 kyr. Thus, we adopt a nominal joint tuning and sampling error of ± 0.01 Ma for the orbitally tuned ages presented in the following discussion. The Moiti Tuff – 722A 12-4 (90–120 cm) We obtain an orbitally tuned age of 3.96 ± 0.01 Ma for the Moiti tephra identiWed within sample 722A 12-4 (90–120 cm) (Figs. 3.3–3.5 and Fig. 3.11). Tephra shards were found at high concentrations over a 60 cm interval (119.00–119.60 mcd), spanning 3.95–3.98 Ma. Feldspar grains within the Moiti tuV from the Turkana Basin, northern Kenya were initially K–Ar dated with a maximum age of O 4.10 ± 0.07 Ma (McDougall, 1985). Subsequent 40Ar/39Ar laser-fusion dating of single crystals within the VT-1 TuV from Aramis, Ethiopia (= Moiti) has reWned this estimate to 3.89 ± 0.02 Ma (White, 1993) and 3.94 ± 0.04 Ma (White et al., 1994). Eight published dates for the VT-1 TuV in White (1993) ranged between 3.786 and 4.065 Ma with a single (unweighted) standard deviation of 0.08 Ma about the 3.89 Ma mean value (see Table 1 in White et al., 1993). The Wargolo Tuff – 721C 13-3 (90–120 cm) We obtain an orbitally tuned age of 3.80 ± 0.01 Ma for the Wargolo tephra identiWed within sample 721C 13-3 (90–120 cm) (Figs. 3.3–3.5, Fig. 3.11). Shards were found at high concentrations between 114.45–115.05 mcd, spanning 3.79–3.81 Ma. This age also agrees extremely well with the cooccurrence of the extinction of the coccolithophorid Reticulofenestra pseudoumbilica at this same level within these sediments, also dated at 3.80 ± 0.04 Ma (Tiedemann et al., 1994). Haileab & Brown (1992) estimated the age of the Wargolo TuV at 3.80 ± 0.05 Ma based on stratigraphic interpolation between adjacent well-dated tuVs. Single crystal sanidine 40Ar/39Ar dates on the VT-3 TuV (= Wargolo) from Aramis, Ethiopia yielded an age of 3.75 ± 0.02 Ma for this horizon (White, 1993). Published dates (13 analyses) on VT-3 TuV (= Wargolo) ranged between 3.65 and 3.89 Ma, with a single (unweighted) standard deviation of 0.09 Ma about the 3.75 Ma mean value (see Table 1 in White, 1993).

Pliocene tephra correlations

-Tulu Bor Tuff – 722A 11-1 (30–60 cm) We obtain an orbitally tuned age of 3.41 ± 0.01 Ma for the -Tulu Bor tephra identiWed within sample 722A 11-1 (30–60 cm) (Figs. 3.3–3.5, Fig. 3.11). Shards were found in high concentrations over a broader 90 cm interval (102.15–102.45 mcd), spanning 3.40–3.43 Ma, with a well deWned peak at 102.30 mcd (3.41 Ma). The -Tulu Bor TuV correlates between many localities on land and the deep sea (Brown, 1982; Brown & Cerling, 1982; Brown et al., 1992; Sarna-Wojcicki et al., 1985), and has been the focus of many dating eVorts (Haileab & Brown, 1992, 1994; McDougall, 1985; McDougall et al., 1992; Walter & Aronson, 1993; White, 1993). The Sidi Hakoma TuV of the Hadar Formation (Middle Awash) is compositionally similar and temporally equivalent to the -Tulu Bor TuV of the Koobi Fora Formation (Turkana Basin), TuV B of the Shungura Formation and TuV U-10 of the Usno Formation of southwestern Ethiopia, and a tuV in the Kipcherere section of the Chemeron Formation in the Baringo Basin of Kenya (Brown, 1982; Walter & Aronson, 1993; Namwamba, 1993), suggesting that it is the product of a very broadly dispersed airfall tephra from a large volcanic eruption, most likely near Munesa at the eastern rim of the Main Ethiopian Rift (Walter & Aronson, 1993). Single crystal laser-fusion 40Ar/39Ar dates on ten (rare) feldspars from the Sidi Hakoma TuV document an age of 3.406 ± 0.007 Ma for this eruption (Walter & Aronson, 1993). Also using single crystal, laser-fusion analysis, White (1993) dated the Sidi Hakoma TuV at 3.39 ± 0.04 Ma. Both of these radioisotopic dates for the -Tulu Bor TuV agree well with our orbitally tuned age of 3.41 ± 0.01 Ma.

Assignment of orbitally tuned ages to previously undated tephra We can assign orbitally tuned ages to the Lokochot and Lomogol TuVs for which direct radioisotopic dates have not been obtainable. Ages for these tuVs have been estimated based on stratigraphic interpolation between adjacent, radioisotopically dated horizons and paleomagnetic polarity stratigraphy.

Lokochot Tuff – 722A 11-4 (60–90 cm) The orbitally tuned age of the Lokochot TuV for sample 722A 11-4 (60– 90 cm) is 3.58 ± 0.01 Ma (Fig. 3.11, Table 3.7). Vitric shards occur in elevated abundances over a 60 cm interval associated with this horizon (107.25– 107.85 mcd) which spans 3.56–3.58 Ma; peak abundances occur at the analysed level (107.70 mcd; 3.58 Ma). The age of the Lokochot TuV was

49

Palaeogeography of the circum-Mediterranean region

50

Table 3.7. Comparison of orbitally tuned ages and stratigraphically interpolated ages of East African tuVs between 3.4 Ma and 4.0 Ma

Tuff

721/722 Orbitally tuned Age (Ma)

Kenyan/Ethiopian/ marine sediment Stratigraphic Age (Ma)

Reference

Lokochot Tuff

3.58 ± 0.01 Ma

Lomogol Tuff

3.62 ± 0.01 Ma

~ 3.50–3.60 Ma ~ 3.50 Ma 3.58 Ma 3.57 Ma 3.59 ± 0.03 Ma 3.59 Ma ~ 3.6 Ma 3.6 ± 0.1 Ma

(Sarna-Wojcicki et al., 1985) (Haileab and Brown, 1992) (Hilgen, 1991) (McDougall et al., 1992) (Tiedemann et al., 1994) (Shackleton et al., 1990) (Haileab and Brown, 1992) (Pickford et al., 1991)

estimated at ~ 3.5–3.6 Ma based on its stratigraphic placement within Site 231 from the Gulf of Aden (Sarna-Wojcicki et al., 1985). Within the Kenyan and Ethiopian sequences the Lokochot TuV occurs almost precisely at the Gauss/Gilbert paleomagnetic reversal (Feibel et al., 1989; Hillhouse et al., 1986) which is presently dated by orbital tuning of various independent stratigraphic sequences at 3.59 ± 0.03 Ma (Tiedemann et al., 1994), 3.59 Ma (Shackleton et al., 1990, 1994), and 3.58 Ma (Hilgen, 1991), and 3.57 Ma by stratigraphic interpolation between adjacent radioisotopically dated tuVs from the Turkana Basin (McDougall et al., 1992).

Lomogol Tuff – 722A 11-5 (60–90 cm) The orbitally tuned age of the Lomogol TuV within sample 722A 11-5 (60–90 cm) is 3.62 ± 0.10 Ma (Fig. 3.11, Table 3.7). Vitric shards that occur are abundant over a 60 cm interval associated with this horizon (108.75– 109.35 mcd) which spans 3.61–3.63 Ma. The Lomogol TuV is known from the north end of the Labur Range in the Turkana Basin of north Kenya (K82742). It has also been detected in the Western Rift Valley of Uganda (Pickford et al., 1991) and in the Gulf of Aden marine sequences (Brown et al., 1992). In the Turkana Basin the age of Lomogol TuV was estimated to be ~ 3.6 Ma because it lies stratigraphically between the Lokochot and the Moiti TuVs (Feibel et al., 1989; Haileab & Brown, 1992).

Conclusions Ages of hominid and other vertebrate fossils from East African Pliocene– Pleistocene sedimentary sequences are largely constrained by direct

Pliocene tephra correlations

radiometric dating and regional intercorrelation of numerous volcanic tuV beds. Several of these tuV beds have been directly dated using K/Ar and 40Ar/39Ar methods but many tuVs have not yet been dated or have insuYcient juvenile feldspar grain concentrations for analysis. The detection and analysis of tephra within adjacent marine sediments (Brown et al., 1992; Sarna-Wojcicki et al., 1985) presents a new opportunity to date a majority of tephra within the context of the very detailed and precise orbitally tuned marine sediment chronostratigraphy (Hilgen, 1991; Shackleton et al., 1990, 1994; Tiedemann et al., 1994). Here we have focused on the mid-Pliocene (3.4–4.0 Ma) record of tephra deposition and associated oxide chemistry within marine sediments from the Arabian Sea, some 2000 km distant from East African source volcanoes. Despite very low vitric shard abundances within the Arabian Sea sediments, suYcient shards were separated from the layers with the most abundant shards for electron microprobe analysis. Major element oxide compositions were used to establish Wrm geochemical correlations to Wve known Turkana Basin tuVs between 3.4–4.0 Ma and their correlative tephra in the Gulf of Aden (-Tulu Bor, Lokochot, Lomogol, Wargolo, and Moiti TuVs). As the Arabian Sea sequence is temporally constrained by an orbitally tuned chronostratigraphy we can use these tephrostratigraphic correlations to directly compare the East African K/Ar and 40Ar/39Ar radiometric ages with the marine orbitally tuned ages. Radioisotopic and orbitally tuned ages generally agree within ± 0.01 Ma for each of the three mid-Pliocene tuVs where both age determinations were available (-Tulu Bor, Wargolo, Moiti TuVs). We have assigned new (orbitally tuned) ages to the Lokochot and Lomogol TuVs for which direct radiometric dates were not available. We present this as a proof-of-concept study. We estimate that a total of 25–30 separate tephra horizons can be similarly analysed, intercorrelated, and dated for the Pliocene–Pleistocene (0.5–4.5 Ma) interval at Sites 721 and 722 in the Arabian Sea. Once complete, these terrestrial–marine tephra correlations will provide a framework for direct testing of hypotheses linking known changes in African hominid and other vertebrate evolution to known changes in regional (African) and global climate.

References Behrensmeyer, A. K., Todd, N. E., Potts, R. & McBrinn, G. E. 1997. Late Pliocene faunal turnover in the Turkana Basin, Kenya and Ethiopia. Science, 278, 1589–94. Berger, A. & Loutre, M. F. 1991. Insolation values for the climate of the last 10 million years. Quat. Sci. Reviews, 10, 297–317.

51

Palaeogeography of the circum-Mediterranean region

52

Bloemendal, J. & deMenocal, P. B. 1989. Evidence for a change in the periodicity of tropical climate cycles at 2.4 Myr from whole-core magnetic susceptibility measurements. Nature, 342, 897–9. Borchardt, G. A., Aruscavage, P. J. & Millard, H. T. 1972. Correlation of the Bishop Ash, a Pleistocene marker bed, using instrumental neutron activation analysis. J. Sed. Petrol., 42 (2), 301–6. Brown, F. H. 1982. Tulu Bor tuV at Koobi Fora correlated to the Sidi Hakoma TuV at Hadar. Nature, 300, 631–5. Brown, F. H. & Cerling, T. E. 1982. Stratigraphic signiWcance of the Tulu Bor TuV. Nature, 299, 212–15. Brown, F. H., Sarna-Wojcicki, A. M., Meyer, C. E. & Haileab, B. 1992. Correlation of Pliocene and Pleistocene tephra layers between the Turkana Basin of East Africa and the Gulf of Aden. Quaternary International, 13/14, 55–67. Clemens, S. C., Murray, D. W. & Prell, W. L. 1996. Nonstationary phase of the Plio-Pleistocene Asian Monsoon. Science, 274, 943–8. Clemens, S. C. & Prell, W. L. 1990. Late Pleistocene variability of Arabian Sea summer monsoon winds and continental aridity: Eolian records from the lithogenic component of deep-sea sediments. Paleoceanography, 5, l09–45. Clemens, S. & Prell, W. L. 1991. One million year record of summer monsoon winds and continental aridity from the Owen Ridge (Site 722), Northwest Arabian Sea. In Preceedings of the Ocean Drill. Prog., Prell, W. J. & Niitsuma, N. (eds.), pp. 365–88. Ocean Drill. Proj., College Station, TX. Clemens, S., Prell, W., Murray, D., Shimmield, G. & Weedon, G. 1991. Forcing mechanisms of the Indian Ocean monsoon. Nature, 353, 720–5. deMenocal, P. B. 1995. Plio-Pleistocene African Climate. Science, 270, 53–9. deMenocal, P. B. & Bloemendal, J. 1995. Plio-Pleistocene subtropical African climate variability and the paleoenvironment of hominid evolution: A combined data-model approach. In Paleoclimate and Evolution With Emphasis on Human Origins, Vrba, E., Denton, G., Burckle, L. & Partridge, T. (eds.), pp. 262–88. Yale University Press, New Haven. deMenocal, P. B., Bloemendal, J. & King, J. W. 1991. A rock-magnetic record of monsoonal dust deposition to the Arabian Sea: Evidence for a shift in the mode of deposition at 2.4 Ma. In Proc. Ocean Drill. Prog., ScientiWc Results, Prell, W. L., Niitsuma, N. et al. (eds.), pp. 389–407. Ocean Drilling Program, College Station, TX. deMenocal, P. B. & Rind, D. 1993. Sensitivity of Asian and African climate to variations in seasonal insolation, glacial ice cover, sea-surface temperature, and Asian orography. J. Geophys. Res., 98 (4), 7265–87. deMenocal, P. B., Ruddiman, W. F. & Pokras, E. M. 1993. InXuences of high- and low-latitude processes on African climate: Pleistocene eolian records from equatorial Atlantic Ocean Drilling Program Site 663. Paleoceano., 8 (2), 209–42. Feibel, C. S., Brown, F. H. & McDougall, I. 1989. Stratigraphic context of fossil hominids from the Omo Group Deposits: Northern Turkana Basin, Kenya. Amer. J. Phys. Anthro., 78, 595–622. Grine, F.E. 1986. Ecological causality and the pattern of Plio-Pleistocene hominid evolution in Africa. S. Afr. J. Sci., 82, 87–9. Haileab, B. & Brown, F. H. 1992. Turkana Basin-Middle Awash Valley correlations and the age of the Sagantole and Hadar Formations. J. Hum. Evol., 22, 453–68. Haileab, B. & Brown, F. H. 1994. Tephra correlation between the Gadeb prehistoric site

Pliocene tephra correlations

and the Turkana Basin. J. Hum. Evol., 26, 167–73. Hastenrath, S. 1985. Climate and Circulation of the Tropics, D. Reidel, Boston. Hilgen, F. J. 1991. Astronomical calibration of Gauss to Matuyama sapropels in the Mediterranean and implication for the geomagnetic polarity timescale. Earth and Planet. Sci. Lett., 104, 226–44. Hillhouse, J. W., Cerling, T. E. & Brown, F. H. 1986. Magnetostratigraphy of the Koobi Fora Formation, Lake Turkana, Kenya. J. Geophys. Res., 91 (B11), 11581–95. Kutzbach, J.E. 1981. Monsoon climate of the early Holocene: Climatic experiment with the earth’s orbital parameters for 9000 years ago. Science, 214, 59–61. Kutzbach, J.E. & Guetter, P. J. 1986. The inXuence of changing orbital parameters and surface boundary conditions on climate simulations for the past 18,000 years. J. Atmos. Sci., 43, 1726–59. McDougall, I. 1985. K-Ar and 40Ar-39Ar dating of the hominid-bearing Pliocene-Pleistocene sequence at Koobi Fora, Lake Turkana, northern Kenya. Geol. Soc., Amer. Bull., 96, 159–75. McDougall, I., Brown, F. H., Cerling, T. E. & Hillhouse, J. W. 1992. A reappraisal of the geomagnetic polarity timescale to 4 Ma using data from the Turkana Basin, East Africa. Geophys. Res. Lett., 19 (23), 2349–52. Murray, D. W. & Prell, W. L. 1991. Pliocene to Pleistocene variations in calcium carbonate, organic carbon, and opal on the Owen Ridge, Northern Arabian Sea. In Proc. O.D.P., Sci. Res., Prell, W. L. & Niitsuma, N. (eds.), pp. 343–55. Ocean Drilling Program, College Station, TX. Nair, R. R., Ittekot, V., Manganini, S., Ramaswamy, V., Haake, B., Degens, E., Desai, B. & Honjo, S. 1989. Increased particle Xux to the deep ocean related to monsoons. Nature, 338, 749–51. Namwamba, F. 1993. Tephrostratigraphy of the Chemeron Formation, Baringo Basin, Kenya, M.S. thesis, University of Utah, Salt Lake City. Nash, W. P. 1992. Analysis of oxygen with the electron microprobe: applications to hydrated glass and minerals, Amer. Min., 77, 453–7. Perkins, M. E., Brown, F. H., Nash, W. P., McIntosh, W. & Williams, S. K. 1998. Sequence, age, and source of silicic fallout tuVs in middle to late Miocene basins of the northern Basin and Range province. Geol. Soc. Amer. Bull., 110 (3), 344–60. Perkins, M. E., Nash, W. P., Brown, F. H. & Fleck, R. J. 1995. Fallout tuVs of Trapper Creek, Idaho – A record of Miocene explosive volcanism in the Snake River Plain volcanic province. Geol. Soc. Amer. Bull., 107, 1484–506. Pickford, M., Senut, B., Poupeau, G., Brown, F. H. and Haileab, B. 1991. Correlation of tephra layers from the Western Rift Valley (Uganda) to the Turkana Basin (Ethiopean/Kenya) and the Gulf of Aden. C. R. Acad. Sci. Paris, 313 (II), 223–9. Pouchou, J.-L. & Pichoir, F. 1991. Quantitative analysis of homogeneous or stratiWed microvolumes applying the model ‘PAP’. In Electron Probe Quantitation, Heinrich, K. F. J. & Newbury, D. E. (eds.), pp. 31–75. Plenum Press, New York. Prell, W. L. & Kutzbach, J. E. 1987. Monsoon variability over the past 150,000 years. J. Geophys. Res., 92, 8411–25. Prell, W. L. & Kutzbach, J. E. 1992. Sensitivity of the Indian monsoon to forcing parameters and implications for its evolution. Nature, 360, 647–52. Prell, W. L. & Niitsuma, N. 1988. Milankovitch and monsoons. Nature, 331, 663–4. Prell, W. L., Niitsuma, N. & Emeis, K. 1988. Leg 117 Wnds mountains, monsoons. Geotimes, 33 (3), 13–16. Pye, 1987. Eolian Dust and Dust Deposits, Academic Press, New York.

53

Palaeogeography of the circum-Mediterranean region

54

Renne, P., Walter, R., Verosub, K., Sweitzer, M. & Aronson, J. 1993. New data from Hadar (Ethiopia) support orbitally-tuned time scale to 3.3 Ma. Geophys. Res. Lett., 20 (11), 1067–70. Sarna-Wojcicki, A. M. & Davis, J. O. 1991. Quaternary tephrochronology. In Quaternary Nonglacial Geology, Conterminous U.S., R. B. Morrison (ed.), pp. 93–116. Geol. Soc. Amer., Geol. N. Amer., Boulder, CO. Sarna-Wojcicki, A. M., Meyer, C. E., Roth, P. H. & Brown, F. H. 1985. Ages of tuV beds at East African early hominid sites and sediments in the Gulf of Aden. Nature, 313, 306–8. Shackleton, N. J., Berger, A. & Peltier, W. R. 1990. An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Transactions of the Royal Society of Edinburgh: Earth Sciences, 81, 251–61. Shackleton, N. J., Crowhurst, S., Hagelberg, R., Pisias, N. J. & Schneider, D, A. 1994. A new late Neogene time scale: Application to leg 138 sites. In ScientiWc Results, ODP Leg 138, Mayer, L., Pisias, N. J. & Janecek, T. (eds.). Ocean Drilling Program, College Station, TX. Sirocko, F. 1989. Accumulation of eolian sediments in the northern Indian Ocean; record of the climatic history of Arabia and India. Berichte-Reports, Geol.-Pala¨ont. Inst. Univ. Kiel, 27, 1–114. Sirocko, F., Garbe-Schonberg, D., McIntyre, A. & MolWno, B. 1996. Teleconnections between the subtropical monsoons and high-latitude climates during the last deglaciation. Science, 272, 526–9. Sirocko, F. & Sarnthein, M. 1989. Wind-borne deposits in the northwestern Indian Ocean: record of Holocene sediments versus modern satellite data. In Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport, Leinen, M. & Sarnthein, M. (eds.), pp. 401–33. Kluwer Academic Publishers. Tiedemann, R., Sarnthein, M. & Shackleton, N. J. 1994. Astronomic timescale for the Pliocene Atlantic d18O and dust Xux records of ODP Site 659. Paleoceanography, 9 (4), 619–38. Vrba, E. 1995. The fossil record of African antelopes (Mammalia, Bovidae) in relation to human evolution and paleoclimate. In Paleoclimate and Evolution With Emphasis on Human Origins, Vrba, E., Denton, G., Burckle, L. & Partridge, T. (eds.), pp. 385–424. Yale University Press, New Haven. Vrba, E. S. 1985. Ecological and adaptive changes associated with early hominid evolution, in Ancestors: The Hard Evidence, E. Delson (ed.), pp. 63–71. A. R. Liss, New York, NY. Walter, R. C. 1994. Age of Lucy and the First Family: Single-crystal 40/39 Ar dating of the Denen Dora and lower Kada Hadar members of the Hadar Formation, Ethiopia. Geology, 22, 6–10. Walter, R. C. & Aronson, J. L. 1993. Age and source of the Sidi Hakoma TuV, Hadar Formation, Ethiopia. J. Hum. Evol., 25, 229–40. White, T. D. 1993. New discoveries of Australopithecus at Maka in Ethiopia. Nature, 366, 261–5. White, T. D., Suwa, G. & Asfaw, B. 1994. Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature, 371, 306–12. WoldeGabriel, G., White, T, D., Suwa, G., Renne, P., deHeinzelin, J., Hart, W. K. & Heiken, G. 1994. Ecological and temporal placement of early Pliocene hominids at Aramis, Ethiopia. Nature, 371, 330–3.

4 Climatic perspectives for Neogene environmental reconstructions Eileen M. O’Brien and Charles R. Peters

Ex A fr ica se mp er aliqu id n o vi ( P l iny th e El der )

Introduction Paleoenvironmental reconstructions are based on theoretical implications (Milankovitch cycles, etc.) and empirical evidence (e.g. paleontology). The general tendency is to attempt to articulate the latter with the former to try to make scanty empirical evidence Wt theoretical expectations. Under the current paradigm, the major tendency is to assume that climatic changes in the mid-to-high latitudes (especially those involving temperature) drive those in the tropics. Another major tendency is to assume that temperature changes in the oceans are reasonable analogs for temperature changes on land. There are reasons to doubt these practices. Some ideas and evidence for why they can be inappropriate are presented below, based on how climate operates today, on how it theoretically relates to spatio-temporal variations in biological activity, how it relates to woody-plant species richness (a partial proxy for vegetation structure/complexity), and on the little that is known about the physiography, climate and vegetation of Pliocene Africa. Four key topics considered below are: 1) how both precipitation and temperature need to be considered when evaluating changes in climate/ vegetation; 2) how changes in climate/vegetation can be a function of regional-to-local changes in physiography rather than global climatic change (e.g. Milankovitch cycles); 3) how wetlands, especially grasslands resulting from impeded drainage, make grass pollen an ambiguous indicator of paleoclimate change in Africa; and 4) how climate’s relation to woody-plant species richness, and thus vegetation, can be used to elucidate paleoecological anomalies. Although we focus primarily on Africa and on understanding Pliocene environmental changes, we expect that some of the ideas, implications, and questions raised herein apply to other time periods, as well as elsewhere in the world, especially Eurasia.

Both the water and energy regimes need to be considered when describing climate, as well as climate’s relationship to vegetation It is a common practice to use global changes in annual temperature as an indicator of changes in terrestrial climate and vegetation (as well as asso-

Palaeogeography of the circum-Mediterranean region

56

ciated biota, etc.). However, as described below, neither temperature nor precipitation alone is suYcient for describing or evaluating changes in terrestrial climate, let alone changes in vegetation. Moreover, except perhaps at some gross mega-scale level of analysis, annual measures of the ambient energy regime (temperature or potential evapotranspiration) are not signiWcantly related to variability in most aspects of terrestrial vegetation (e.g. productivity, biological activity, species richness) to predict patterns at the continental or regional scales of analysis (O’Brien, 1998).

Climate today Summarized here are some of the key factors related to geographic and temporal variations in climate, gleaned from a variety of standard texts (e.g. Lutkens & Tarbuck, 1992; Grotjahn, 1993). The intent is to provide a concise synthesis of how these factors operate and relate to each other and, in so doing, provide the foundation for our alternate interpretations, queries and ideas. Climate is a complex negative–positive feedback system that is fueled by solar radiation (insolation) and driven by planetary surWcial thermal dynamics, especially atmospheric/oceanic thermal exchange and the hydrologic cycle (e.g. Grotjahn, 1993). Climatic variability is minimally described by intra-annual variations in precipitation, temperature and daylength, i.e. by variations in the ambient water and energy (heat/light) regimes, respectively. In accord with the physical principles governing atmospheric/oceanic thermal exchange and water energy interactions, climate varies dynamically over space and time due to: 1) mega-scale earth– sun relationships limiting the interception of insolation; 2) mega-to-macroscale planetary thermal processes and physical factors controlling the hydrologic cycle, atmospheric/oceanic circulation/composition, landmass– waterbody distribution, etc.; and, 3) continental, regional and sub-regional factors and processes that further modify climate (topography, elevation, coastal oceanic currents, etc.). Over waterbodies, energy is the active factor with regard to climatic variability (availability of water being a constant). Over landmasses, both energy and available water are active factors.

First-order global variations in climate Spatio-temporal variability in the interception of insolation results in a Wrst-order latitudinal gradient in climate. As can be seen in Fig. 4.1a, the net earth–atmosphere radiation curve describes a latitudinal gradient of decreasing thermal energy (terrestrial radiation in response to incident insola-

Climatic perspectives for Neogene environment

57

[Figure 4.1] Generalized global latitudinal variation in critical climatic parameters: (a) annual net earth-atmosphere radiation (Q*e−a) (after Suckling & Doyan, 1981; see Grotjahn for absolute values) – indicating latitudes associated with annual energy surpluses or deficits, as well as associated trends in thermal seasonality and average minimum and maximum monthly potential evapotranspiration (PET) (based on Thornthwaite & Mather 1962, 1964a,b, 1965); (b) average annual precipitation (based on Thornthwaite & Mather, 1962, 1964a,b, 1965); and (c) atmospheric circulation pattern (at equinox) (based on Lutkens & Tarbuck, 1992) – indicating perpetual zones of high and low pressure, the prevailing winds between them, and the Intertropical Convergence Zone (ITCZ) of low pressure associated with easterly winds. Note: the decreasing distance between latitudes along the abscissae reflects their positions relative to the plane of solar radiation interception. From O’Brien & Peters, n.d.

tion) polewards of the equator; an annual surplus occurring between c. 30 + ° south and north of the equator, an annual deWcit elsewhere. This pattern arises for two reasons: 1) due to the planet’s sphericity the angle of incident incoming solar radiation, or insolation, decreases exponentially polewards of the equator, causing a perpetual concordant gradient of decrease in the

Palaeogeography of the circum-Mediterranean region

58

intensity (amount per unit area) of insolation; and 2) due to the planet’s axial tilt, there is a concordant gradient of exponentially increasing intra-annual variability in daylength, and thus thermal seasonality, that becomes signiWcant polewards of the Tropics of Cancer (23.5° N) and Capricorn (23.5° S). During summer, when the zone of maximum insolation intensity is centered over the Tropic of Cancer, Eurasia (up to c. 50° N) experiences temperatures on a par with that of Africa. Theoretically, over time, the gradient of this global curve can change, steepening (e.g. beyond present?) or Xattening (e.g. Cretaceous), depending upon cloud cover (see below). Fluctuations in the intensity of insolation (e.g. during Milankovitch cycles) can alter the absolute amount or seasonal duration of insolation, but not the fundamental latitudinal gradient.

Precipitation curve There is a marked tendency for annual precipitation over landmasses to increase from a minimum at the poles to a small maximum in the midlatitudes, to decrease to another minimum at c. 30° S/N, and then to increase continuously thereafter to a massive maximum within the Tropics that peaks in the equatorial regions (Fig. 4.1b). This undulating latitudinal gradient in precipitation is a function of two main factors: Wrst, energy’s direct relationship to the hydrological cycle (more energy, more water moving through the cycle); and second, the global distribution of high and low atmospheric pressure cells which modify the velocity, moisture content and direction of prevailing winds (Fig. 4.1c). The hydrologic cycle contributes to atmospheric heating and planetary thermal equilibrium at all scales – moving energy from the earth’s surface (via evaporation) into the atmosphere, where it (and water) can be redistributed to elsewhere in the world. The main processes involved are: evaporation (oV oceans, lakes, land surface) ; condensation and cloud formation ; advection (horizontal Xow of air over water/land) ; precipitation ; runoV back to sea/lake, and/or in situ evaporation. Since many factors come to bear on the ultimate distribution of precipitation, especially atmospheric circulation, it can be much more variable geographically than the energy regime. The potential for precipitation is both a positive and negative function of the ambient energy regime. As ambient air temperature increases there is an increase in the potential for evaporation (water changing from a liquid to gas state), in the water-holding capacity of the air, and in the saturation point at which precipitation occurs. However, increases in temperature decrease the potential for condensation and cloud formation and thus the

Climatic perspectives for Neogene environment

potential for precipitation. In eVect, condensation, cloud formation and precipitation depend on moisture-laden air being cooled, thereby decreasing the water-holding capacity and saturation point of air and promoting precipitation. Heated air cools because it rises and/or is advected into colder regions of the atmosphere. A primary cooling mechanism here is the negative relationship between air temperature and elevation above sea level (environmental lapse rate: − 6.5 °C per km). Changes in air temperature with elevation also occur adiabatically when air is forced to rise/descend, and thus expand (cool) or compress (warm up). The rate of adiabatic cooling or heating is more variable than the environmental lapse rate; being slower when the air is wet ( − / + 5 °C to − / + 9 °C per km, depending upon moisture content), faster when it is dry ( − / + 10 °C per km). Adiabatic cooling/heating promotes orographic rainfall and rainshadow eVects. The amount of water moving through the hydrologic cycle is a function of the energy regime and the amount of surface water available for evaporation. Over waterbodies, surface water is a constant and the potential amount of water moving through the hydrologic cycle can increase in a positive linear fashion with regard to increases in the ambient energy conditions. Over landmasses, however, surface water is not a constant, varying geographically and seasonally. As a consequence, the realized amount of precipitation over land can increase in accord with increases in temperature, but only as long as the environmental demand for water (potential evapotranspiration) is being met. The amount of precipitation decreases once the water-holding capacity of air (and its saturation point) becomes greater than the amount of water in it; at which point, the potential for condensation, cloud formation and precipitation decreases. For Africa, for example, the relationship between realized precipitation and the ambient energy regime describes a parabolic curve – precipitation increasing as the energy demand increases up to a point beyond which precipitation decreases as the energy demand continues to increase (Fig. 4 in O’Brien, 1998).

Atmospheric circulation Atmospheric circulation governs the advection phase of the hydrologic cycle, and thus the geographic distribution of water on landmasses (Fig. 4.1c). It is driven by the earth’s rotation and by thermal and pressure diVerentials in the atmosphere. It contributes to planetary thermal equilibrium by moving surplus heat from the Tropics polewards, while drawing cooler air equatorwards. In general, air Xows from high to low pressure at speeds determined by the temperature gradient – the steeper the temperature gradient, the greater the velocity of wind Xow.

59

Palaeogeography of the circum-Mediterranean region

60

The ITCZ (Intertropical Convergence Zone) (Fig. 4.1c) is a continuous and perpetual low pressure belt that girdles the equatorial regions. If the surplus of energy in Tropical regions is the engine driving the atmospheric (and oceanic) circulation system, then the ITCZ is the drivebelt – drawing warm and usually wet air (trades) from both hemispheres equatorwards and then upwards into the troposphere where it is then injected (via rotational deXection of the atmosphere) into the higher latitudes of both hemispheres (Fig. 4.1c). The convergence and convectional uplift (and thus cooling) of air by the ITCZ drives monsoonal rainfall patterns and results in the massive precipitation maximum in the Tropics. How much, where and when rainfall and cloud cover occur within the Tropics, however, depends on the moisture content of the air advected into the Tropics, the location of the ITCZ, and ultimately on regional-to-local physiography and surface–air interactions aVecting the heating–cooling and moisture content of the air (e.g. inland lakes, topography). Seasonal spatial oscillations in the location of the ITCZ usually range within 10° N/S of the equator over the oceans, and between as much as  20° N/S over the continents. The resulting seasonal changes in the moisture regime are the primary reason for seasonality in vegetative growth in tropical regions of the world where energy (heat/light) for vegetative growth is non-limiting. The Subtropical High Pressure Cells (STHP), located over oceans and seasonally over land (e.g. southern Africa), are sources of descending (warming) dry air that absorbs moisture from the planet’s surface. The STHP over oceans generate the warm and usually wet northeasterly and southeasterly tradewinds, as well as the wet westerly winds of the mid-tohigh latitudes. High pressure cells associated with landmasses usually generate dry winds (e.g. northeasterlies into Africa out of Eurasia, see below). During the Eocene, the STHP may have been located further polewards, at about 65° N/S (Crowley & North, 1991).

Factors that modify first-order patterns in climate There are several ways in which the Wrst-order patterns and relations imposed by insolation can be mitigated or exaggerated. First, climate is strongly inXuenced by the diVerential heating–cooling properties of land vs. water. Land heats up and cools oV faster and to a greater degree than water. Unlike the continents, the oceans are giant thermal sponges, storing heat from insolation longer than land. As a consequence, they heat air advected over them long after the adjacent land is cooling that air; or cool over-heated air advected on to them oV continents (maritime vs. continental climates). This heating–cooling diVerential also contributes to the seasonal development of

Climatic perspectives for Neogene environment

low pressure (rising air) over continents during periods of high insolation, and high pressure (descending air) over continents during periods of low insolation. The former draws air onto the continents (ideally wet maritime air, e.g. monsoons), the latter causes airXow onto the oceans. As a consequence, terrestrial environments can over-heat or over-cool relative to the atmosphere overlying large waterbodies at the same latitude. Atmospheric temperature diVerentials between landmasses and waterbodies are especially marked if: a) surface water on land is insuYcient to meet the evaporative demands of the thermal environment; and/or b) cloud cover is not available to inhibit further insolation when insolation is excessive (e.g. in summer at mid-to-high latitudes, year-round at low latitudes), or to inhibit heat loss when insolation is low (e.g. in winter at mid-to-high latitudes, including northern- and southern-most Africa). Thus, over landmasses, the latitudinal gradient in climate can be extremely modiWed, especially where water for the hydrologic cycle is a limiting factor (e.g. the longitudinal gradient across the USA and southern Africa). Second, the geographic location and conWguration of landmasses with respect to other landmasses and waterbodies inXuences the temperature and moisture content of air advected over them, the prevailing wind patterns, the distribution, type and strength of atmospheric pressure gradients, and thus the direction and velocity of air Xow. Given the geographic location and conWguration of Africa and Eurasia, for example, the prevailing northeasterly tradewinds and northern hemisphere easterlies out of Eurasia into Africa are warm and dry. By comparison, the northeasterly trades and easterlies oV the Atlantic Ocean into South America are warm and wet. The former promotes the Saharan desert; the latter, the Amazon jungle. Third, oceanic currents are driven by atmospheric currents and contribute to planetary thermal equilibrium by transferring warm equatorial waters polewards while drawing cold polar waters equatorwards. These currents can cause spatially anomalous sea surface temperatures, which in turn aVect the temperature and moisture content of air advected over them and onto land (e.g. the Gulf Stream’s warming eVect on Europe). At the local scale, diurnal onshore/oVshore air exchange (roughly a 10 km wide swath) ameliorates temperature and precipitation regimes on land (i.e. maritime climates being warmer in winter, cooler in summer than continental climates). Fourth, the eVects of topographic relief on climate are a function of elevation’s negative relationship to ambient air temperature (see above). As a consequence, uplifted plateaus, for example, have lower maximum and minimum monthly temperatures (and PET values) than do low elevation Xatlands, at the same latitude with the same precipitation regime. Mountainous regions promote orographic rainfall and rainshadow eVects as air is

61

Palaeogeography of the circum-Mediterranean region

62

forced to rise up and over them. Moreover, the thermal regime on the windward sides of mountains can be ameliorated by cloud cover; while the absence of cloud cover on the leeward sides makes them subject to overheating when insolation intensity is high, over-cooling when it is low.

Africa Placing Africa within this global climate perspective (37° N to 35° S), we can say that it falls almost entirely within the surplus energy zone, making the water regime the major factor governing Africa’s climatic variability. The amount, duration, and timing of rainfall deWne Africa’s four basic climate zones (Fig. 4.2a). For detailed reviews of Africa’s climate see GriYths (1976), and Tyson (1986); for maps and climate-diagrams see Jackson (1961) and Walter et al. (1960–67), respectively. Africa suVers from too much rather than too little energy, seasonally or year-round, except at high elevations. Daylength varies minimally. Only at its highest latitudes and highest elevations is there the likelihood of very low ambient temperatures and frost (or snow), seasonally or year-round. Extremely high ambient temperatures (over-heating) are typical where and when rainfall is insuYcient to meet the environmental energy demand for water (potential evapotranspiration, or PET). This can be year-round, as in the Horn of Africa, with its arid extension into Kenya; or seasonally, as in the Sahara or Kalahari (or as in the Arabian and Iberian peninsulas or interior of Asia). The availability of surface water to meet the environmental demand for water (PET) depends today on monsoonal rainfall (Fig. 4.2b). Within Africa, sources of surface water to meet this demand are limited to soil moisture, relatively small bodies of water (Lake Victoria), and convectional recycling of rainfall. Continental waterbodies of suYcient size to contribute signiWcantly to the hydrologic cycle, and thus climate (at other than the local scale), do not exist in Africa today. Only the prevailing equatorial easterlies, southeasterly trades, and equatorial westerly winds, originating over oceans, advect warm wet air into Africa (Fig. 4.2c–d). Winds originating over landmasses in the northern hemisphere (northeasterlies, ITCZ jet stream) seasonally advect warm/cool and dry (or low moisture) air into, across and out of northern subtropical Africa. In the absence of surface water to meet the evaporative demand, these winds heat-up, further promoting aridity and simultaneously steepening the atmospheric temperature gradient, thereby increasing wind velocities. These winds exit primarily to the west (Atlantic Ocean), but also into the Bay of Guinea via the Dahomey Gap. In conjunction with the Himalayan low pressure cell during boreal summer,

Climatic perspectives for Neogene environment

[Figure 4.2] Africa’s prevailing winds, rainfall and climate regions. (a) Climate regions: Y, humid-equatorial diurnal climate with rain more-or-less year-round (only a short dry season invariably occurs in most years); B, equatorial bimodal-rain climate with two marked dry seasons, particularly pronounced in semi-arid East Africa; S, tropical summer-rain climate (subhumid to desert); W, winter-rain climate of the Cape and circum-Mediterranean. Coastal black dots denote littoral climates with seasonal mists (fog) chiefly in winter; coastal spikes denote littoral climates with seasonal mists chiefly in summer. From Peters (1990). Frost designations, based on Werger & Coetzee (1978): the hatched area designates three or more months of frost annually; the dashed outlines indicate the areas that have frost annually; the dotted outline indicates the frost-susceptible subregion, where frost is known to occur, but not annually. (b) Average annual preciptation (mm). After Jackson (1961). Dashed lines in eastern Africa and the Limpopo Valley of southern Africa mark the 400 mm isoline. Blackened areas have average annual precipitation greater than 1800 mm. (c–d) Prevailing winds and associated migration of the Intertropical Convergence Zone (ITCZ) during (c) austral summer (January) and (d) boreal summer (July). Dotted lines indicate the location of the ITCZ. Dashed arrows indicate sporadic extension of prevailing winds beyond their normal range. Based on Martin & O’Meara (1986), modified somewhat by reference to Fullard (1980). After O’Brien & Peters, n.d.

Palaeogeography of the circum-Mediterranean region

64

these winds foster the warm dry northwesterly winds along the Red Sea and the dry southwesterly winds drawn out of the Horn of Africa and into the Bay of Arabi in the Indian Ocean (Fig. 4.2d). Winds originating over oceans but traversing cold currents may have moisture, but are heated when they hit land, resulting in aridity (e.g. coastal Namibia). Orographic rainfall and rainshadow eVects are common throughout the mountainous regions of Africa, providing, for example, islands of high rainfall on the windward sides of the volcanic uplands of eastern Africa. Thus, the geographic variation in rainfall today (Fig. 4.2b) is primarily a function of inter-continental/oceanic relationships and Africa’s physiography (especially elevation and topographic relief), which determine the distribution of cold currents and the moisture content of prevailing winds. The resulting distribution and amount of annual precipitation is depicted in Fig. 4.2b. It is consistent with the intra-annual variation in the ITCZ, and emphasizes the much greater aridity in Africa north of the equator (Sahara and the Horn) than south of it (Namib and Kalahari). It is also consistent with the existence of two cold currents oV the west coast of Africa: the Benguela current oV Namibia (and Angola) and the Canary current oV the western Sahara. Their associated cool and dry westerly winds provide little, if any, precipitation to adjacent onshore regions, except from fog condensation (e.g. the Namib Desert during austral winter, Fig. 4.2c). We should also mention that precipitation, following uplift of onshore airXow, can occur further inland during austral summer (e.g. the Namibian Highlands). The arid westerlies inXuencing southern Africa are driven by the South Atlantic subtropical high pressure cell and cooled by the Benguela current. Their consequent aridity, when combined with the eVects of the continental high pressure cell over Botswana/South Africa (austral winter), account for southern Africa’s anomalous longitudinal rather than latitudinal gradient in precipitation (see Fig. 4.2b) (Tyson, 1986).

Climate’s relationship to vegetation The same factors that deWne geographic variability in climate, deWne geographic variability in terrestrial vegetation productivity and woody-plant species richness, and thus vegetation physiognomy (shrubland, woodland, forest, etc.). In eVect, as described in Fig. 4.3, biological activity, and woodyplant species richness (number of species) tend to increase as a linear function of the water regime (right ordinate), and as a parabolic function of the energy regime (abscissa) (O’Brien, 1998). The best climate variable for describing energy’s relation to woody-plant species richness/vegetation physiognomy is minimum monthly PET. As noted earlier, annual values for

Climatic perspectives for Neogene environment

[Figure 4.3] The relations of climate-controlled water–energy dynamics to first-order variations in woody plant species richness. Modified from O’Brien et al. (1998).

the energy regime, whether based on temperature or PET, are not signiWcantly related to variations in species richness/vegetation. The best water variable is annual rainfall. Increasing annual rainfall is also very strongly correlated with the increasing duration of the growing season. When annual rainfall and minimum monthly PET are combined, climatebased water–energy dynamics account for 79% of the macro-scale variation in woody-plant species richness in southern Africa (richness per 25 000 km2 gridcell; O’Brien, 1993). This is emphasized by a marked concordance in the geographic distribution between woody-plant species richness and vegetation physiognomy, i.e. species richness increasing as vegetation shifts from desert to shrubland, woodland, forest, etc. The importance of topographic relief as a modiWer of climate and thus species richness/vegetation is supported by the fact that when topographic relief is added as a third variable, 85% of the variation in woody-plant species richness is accounted for (O’Brien et al., in prep. a,b). That these relations apply globally is suggested by their usefulness in describing and predicting Wrst-order geographic variations in species richness/vegetation elsewhere in Africa, as well as elsewhere in the world (O’Brien, 1998). That these relations persist over time is suggested by the Wnding that similar relations obtain between climate and woody-plant genus/family richness (O’Brien et al., 1998, for southern Africa; O’Brien et al., in prep. a,b, for Africa and elsewhere in the world). The foregoing emphasizes the underlying or fundamental nature of climate’s relationship to variations in woody vegetation, the dominant terrestrial vegetation of the earth. In eVect, changes in the energy regime can be positively or negatively related to woody-plant species richness/vegetation (see Fig. 4.3), depending on whether they move the energy regime towards or away from optimal conditions, on how they interact with the water regime, and on whether or not they alter the eVective moisture regime – especially during the growing season. Changes in the water regime, being

65

Palaeogeography of the circum-Mediterranean region

66

linearly related to woody-plant species richness/vegetation, appear more straightforward, but their eVect on richness/vegetation also depends on the ambient energy regime. Seen from this perspective, the impact of climatic changes on vegetation is not as simple to predict or interpret as we might like. Especially since changes in climate can occur without causing changes in vegetation if the eVective moisture regime remains the same. EVective moisture is a critical concept when evaluating the eVects of climatic changes on vegetation. It is the amount of water actually used to meet the environmental demand for evaporation and biological processes (one measure being actual evapotranspiration or AET). In eVect, the same energy demand for water (PET) can result in diVerent levels of eVective moisture (AET), depending upon the amount of water available to meet the environmental demand. Roughly speaking, the eVective moisture regime is equal to PET when precipitation is equal to or greater than PET. The eVective moisture regime is equal to precipitation when precipitation is less than PET, and varies accordingly.

The physiographic evolution of the African continent is more critical to understanding late Neogene paleoenvironmental changes than are Milankovitch cycles How much of the environmental change that occurred during the Neogene, especially during the Pliocene, is due to local or regional changes in climate (e.g. uplift/subsidence), and how much to global changes in climate (e.g. Milankovitch cycles)? How do we discriminate between them? The same factors and processes contributing to variations in climate/ vegetation over space today, cause or contribute to variations in climate/ vegetation over time (Burckle, 1995; Ruddiman & Kutzbach, 1989, 1991; Molnar & England, 1990; Raymo & Ruddiman, 1992; Kukla & Cilek, 1996; Ruddiman, 1997). They include changes in: (a) the intensity/duration of insolation (Milankovitch cycles); (b) the conWguration and physiography of landmasses and waterbodies (e.g. orogeny, tectonic uplift/subsidence, sea Xoor spreading, continental surface hydrology); and (c) atmospheric– oceanic circulation/composition. Of these possibilities, given Africa’s location in the zone of surplus energy (Fig. 4.1), the Milankovitch cycles are the least likely to cause major changes in Africa’s climate and vegetation. The one most likely to cause major changes in climate/vegetation is the physiographic evolution of the continent. Cyclical changes in climate due to variations in earth–sun relationships (insolation) have been operating for eons, and those due to Milankovitch

Climatic perspectives for Neogene environment

cycles are now documented to be continuous as far back as the Triassic (Olsen & Kent, 1996). (See deMenocal & Brown in this volume for details on Milankovitch cycles.) In eVect, they operated in a similar fashion, exerted similar variability in the intensity of incoming solar radiation, throughout the Cenozoic, including the Neogene. So, Milankovitch cycles alone cannot explain why there are ice ages during some time periods in the earth’s history, or why, for example, north Africa’s vegetation shifted from one dominated by wetlands and forest in the early Neogene, to one dominated by a great desert today. Something else is going on that is either driven independently by planetary thermal dynamics and exaggerated by the Milankovitch cycles, or that is somehow directly related to global–regional– local climate/vegetation, regardless of Milankovitch cycles (e.g. see various reports in Eddy & Oeschger, 1993). The point here is that, holding all else equal, the amount of water available for terrestrial plant biological processes would remain the same despite Milankovitch cycles. Changes in the intensity of incoming solar radiation are equal over land and sea at any given latitude. In other words, a decrease/increase in the demand for water for evaporation over land, would be matched by a decrease/increase in the amount of evaporation over waterbodies; and thus a decrease/increase in rainfall over landmasses that is on a par with the decrease/increase in the water needed for evaporation on land. In eVect, although precipitation and temperature may vary as a function of Milankovitch cycles, theoretically the eVective moisture regime, and thus vegetation, could remain the same. One thing is certain, given Africa’s location within the zone of surplus energy (Fig. 4.1) and given climate’s relationship to vegetation (Fig. 4.3); decreases in the intensity of insolation are more likely to positively, rather than negatively, aVect vegetation everywhere in Africa except at high elevations ( 2000 m), and perhaps the winter rainfall zone (Fig. 4.2a). By decreasing the demand for water for evaporation during the growing season, there would be an increase in the water available for productivity etc., if rainfall (and thus soil moisture) remained the same or increased. And, as noted above, even if rainfall decreased, productivity etc. could have remained the same. In other words, for most of Africa during Milankovitch cycles, productivity, and thus vegetation, is more likely to remain the same or to increase during periods of reduced insolation and to decrease during periods of increased insolation (e.g. present). The negative eVects of decreases in insolation on Africa’s vegetation would be most likely to occur at high elevations, especially at high latitudes, where a seasonally cold (frost) or low thermal energy regime could be a limiting factor to plants (see Fig. 4.2a). Although the vegetation at high

67

Palaeogeography of the circum-Mediterranean region

68

elevations and latitudes is adapted to such seasonally cool or cold conditions (e.g. winter rainfall, afro-alpine), some downslope or equatorward shifts in its distribution might occur. Elsewhere in Africa, plants are dormant during the cool season (summer rainfall region) and/or thermal energy is in surplus supply during the growing season (the bimodal rainfall, equatorial humid, and summer rainfall regions), and/or water is the only limiting factor for vegetative growth. Based on a recent overview of Africa’s landforms, climate and vegetation during the Pliocene (O’Brien & Peters, n.d.), with the exception of those areas susceptible a priori to global changes in climate due to changes in insolation (e.g. northern and southern fringes of the continent, high elevations), there is little, if any, evidence of vegetation change during the Pliocene (including the Late Pliocene) that cannot be more simply attributed to the physiographic evolution of the continent rather than to global climate changes. During the Pliocene, the geographic location of Africa and its continental conWguration relative to other continents was similar to the present. The prevailing wind patterns into Africa were essentially the same as today; as was the distribution of warm and cold currents. There was apparently little, if any, signiWcant decrease in sea surface temperatures (Crowley & North, 1991), except oV southern Africa. In association with increases in Antarctic ice volume (c. 5 Ma) and the strengthening of the south Atlantic Subtropical High Pressure Cell (c. 3.2–2.5 Ma), sea surface temperatures in the Southern Ocean decreased, while along southwest Africa there was increased cold water upwelling in the Benguela current (Namibia) (Deacon et al., 1992). The latter changed the prevailing westerlies in this region from warm and wet to cool and dry by the Late Pliocene. In northern Africa, aeolian dust deposits in the Sea of Arabia and in the Atlantic Ocean oV western Africa increased in the Late Pliocene at a time when ice volume increased in the northern hemisphere (3.2 and 2.4 Ma) (Crowley & North, 1991; deMenocal & Rind, 1993). Increased aridity in Africa is postulated. However, an alternate explanation is simply increased wind velocity, and thus an increased potential to transport dust (i.e. more dust be moved from the same source areas). This is consistent with a steepened temperature gradient between the poles and the equator, and with the hypothesized eVects of the Himalayan uplift (c. 2.8 Ma) on atmospheric circulation patterns (see Prell & Kutzbach, 1992). What other evidence exists suggests Africa’s climate was wetter during the Pliocene than during the Middle and Late Pleistocene. Before 3.5 Ma and up to 2.5 Ma, tropical forest apparently extended 5° further northwards than present along the coast of western Africa (into what is now a climatically

Climatic perspectives for Neogene environment

sensitive area of Africa, the Sahel-Sudan) (Leroy & Dupont, 1994; Dupont & Leroy, 1995). Despite two short spikes of ‘aridity’ in the pollen record (c. 3.26 and 2.7 Ma), associated with global ice volume Xuxes and increased tradewind velocity (Leroy & Dupont, 1994; Tiedemann et al., 1994), tropical forest and mangrove swamps persist up to 1.9 Ma (Leroy & Dupont, 1994). Malay (1980) suggests that deserts Wrst appear in parts of the northern Sahara during the Early Pliocene, while humid conditions prevail (at least seasonally) in the southern Sahara (Sahel) and the equatorial regions of west and central Africa, with Paleolake Chad being an extant waterbody during the Pliocene. Mesic conditions exist contemporaneously in parts of eastern Africa that are also arid today: e.g. in the Turkana Basin c. 3.4–3.3 Ma (Williamson, 1985) and in the Omo Valley  4.0–2.5 Ma (BonneWlle & Hamilton, 1986; see also Yemane et al., 1985, 1987, for post-Miocene Lake Chilga environs, Ethiopia). Between 2.5–2.0 Ma the pollen record is poor for Africa. For the record between 2.0–1.14 Ma, less mesic conditions are interpreted to exist in the Turkana Basin (Vincens 1987), based on an increase in grass pollen and semi-arid vegetation. Malay (1980) also notes the appearance of aeolian dust in the lacustrine deposits of the northern portions of Lake Chad beginning 2.2–1.8 Ma. Do the causes of these changes necessarily lie in the physical dynamics of earth–sun relations? Apparently not. Physiographically, Africa changed dramatically during the Pliocene, especially the Late Pliocene. High Africa, which includes eastern, south central and southern Africa, was uplifted by as little as 200 m in some places, by at least 1000 m in others (Partridge, 1995a,b). For the southern platform (south central and southern Africa), the uplift was broadscale and probably relatively synchronous since it resulted in minimal fragmentation of environments, and it was not associated with volcanic activity. In eastern Africa, however, uplift events were asynchronous and usually accompanied by subsidence events, extensive fragmentation of the terrain, and volcanics. In the Ethiopian Highlands, for example, evidence from the pollen record indicates non-Afromontane evergreen forest (‘wet miombo’) pollen in post-Miocene lacustrine deposits now at 2000 m elevation (Yemane et al., 1985, 1987). While at Hadar in the Afar graben, there is evergreen shrubland and montane forest pollen (c. 3.3–2.9 Ma) in deposits now at 500 m elevation (BonneWlle et al., 1987). In both cases, the estimated shift in elevation (up or down) was at least 1000 m. Volcanics can inXuence climate in a fashion similar to tectonic uplift/ subsidence events. The resulting topography, however, can be more complex and can result in local wetlands or mock aridity. Topographic relief associated with mountains or volcanoes contributes to orographic rainfall and rainshadow eVects, as well as increased variability in climate at the local

69

Palaeogeography of the circum-Mediterranean region

70

scale: e.g. morning vs. evening shade diVerences, valley side vs. valley bottom heat diVerentials (cooler air subsiding into valleys at night, while warmer valley air rises upwards during the day); in situ origins of streams and springs; and cloud formation, especially during the rainy season, which inhibits insolation and reduces PET. Mock aridity, on the other hand, is the result of ongoing volcanic eruptions and debris Xows inducing barrenness, xeric conditions, and extreme geochemical alkalinity or salinity in the context of a regionally more humid climate (Harris & Van Couvering, 1995). See Yuretich (1982) for a related discussion. Changes in hydrology also aVect regional/local climate and vegetation independently of global climate. Indeed, the presence of major interior paleolakes (Fig. 4.4) may be the most signiWcant factor aVecting Africa’s general-to-regional climate during the Pliocene. These interior lakes were internal buVers, dampening or eliminating the potential eVects of anomolous or cyclical variability in the amount and distribution of ITCZdriven rainfall, as well as oVsetting or minimizing the eVects of the arid northeasterlies and cyclical changes due to earth–sun relationships or atmospheric circulation. Being sources of surface water for the hydrologic cycle, these lakes could generate rainfall/clouds, contribute regionally via rainfall to water table and soil moisture stability, and thereby reduce seasonal stress on vegetation during the dry season and droughts, while increasing the potential for productivity, etc. How much of the change in climate and vegetation (and thus fauna) during the Pliocene in Africa can be attributed simply to the demise of these interior waterbodies? A similar question can be asked about the interior wetlands and waterbodies of Eurasia and how changes in its climate, vegetation and fauna during the Neogene relate to the demise of the epicontinental seas. Moreover, the demise of the Tethys and Eurasia’s epicontental seas probably played a major role in altering Africa’s climate during the Neogene by changing the moisture content of prevailing easterly and northeasterly winds into Africa from wet to dry. For Africa, the most important lake, and the one thought to have been perennial throughout most of the Pliocene, was Paleolake Congo (Cahen, 1954; Beadle, 1981; Pritchard, 1979). Given its location and size, it could have contributed signiWcantly to atmospheric moisture, cloud cover and precipitation in all of its surrounding regions, not only in the equatorial zone, but also north and south of it as the ITCZ migrated. Due to river capture, faulting and/or uplift, most of these interior basins have been breached; the Kalahari basin sometime during the Pliocene or Pleistocene; the Congo basin, in the very Late Pliocene or Early Pleistocene; the Sudd basin, in the Middle Pleistocene; and the Arouane basin, in the Late Pleistocene (Pettars, 1991; Pritchard, 1979; Cahen, 1954; Cooke, 1980;

Climatic perspectives for Neogene environment

71

[Figure 4.4] Pre-Rift Mio-Pliocene Africa: based on de Heinzelin (1963); sketchy and highly conjectural in part. Interior, unbreached drainage basins expected to have contained seasonal, if not perennial water; these waterbodies are boldly outlined and larger basins named (cf. Pritchard, 1979). Blackened areas designate some ancient massifs. Dotted outline of continent denotes present-day coastline of Africa. Question marks indicate where data was unavailable or unobtainable (buried, e.g. Ethiopia). The Sahabi River in the northern coastal region has an indeterminate source, based on Barr & Walker (1973) and de Heinzelin & Arnauti (1987). Presently submerged west coast river canyons are indicated by dashed lines. Modified from O’Brien & Peters, n.d.

Dingle et al., 1983). The consequences with regard to the associated regional climates and to the biogeography of Africa’s vegetation and biota would have been dramatic (for an overview see O’Brien & Peters, n.d.). Large parts of Africa are certainly drier as a result of their cumulative loss.

The wetland grasslands of Africa – why grass pollen is not a definitive indicator of aridity Although grasses play a signiWcant role in much of the vegetation that covers the Sahel, Sudano-Zambezian and Kalahari portions of the continent, today, they are probably not the natural dominants in these areas (Fig. 4.5). Walter

Palaeogeography of the circum-Mediterranean region

72

(1971; White, 1983) concluded that wooded grassland is the zonal vegetation on sandy soils covering Xat terrain receiving 250–500 mm annual rainfall in the summer-rain regions (compare Figs. 4.2 and 4.5). These requirements are found extensively only in the wetter half of the Kalahari portion of the Kalahari–Highveld transition zone, and the wetter half of the Sahel transition zone (White, 1983). These zones stand in contrast to East Africa. For East Africa, in the bimodal-rain region, where the annual rainfall is 250– 500 mm, closed-canopy bushland is the dominant vegetation (perhaps in large part because of the near absence of extensive Xat areas mantled with deep sands). Grasses are present but they are physiognomically subordinate. Generally, where the annual rainfall exceeds 500 mm, woodland replaces wooded grassland and bushland as the natural vegetation. Where the annual rainfall is between 250 and 100 mm, pure grassland occurs on sandy soils. Most extensively, this would be the Saharo-Sahelian grasslands (Le Hourou, 1993), the Namib desert grasslands (Tainton & Walker, 1993), and portions of the Somalia coastal plain grasslands (Michelmore, 1939; Herlocker et al., 1993). Elsewhere in Africa, the extensive grasslands we commonly see, lacking any signiWcant growth of woody-plants, are the consequences of local edaphic and/or pyric conditions (Herlocker et al., 1993). The Highveld Grassland subregion of South Africa is apparently a product of frost, Wre and soil conditions (Tainton & Walker, 1993). The dambos, which permeate much of the woodland and forest of Africa, are a product of poorly drained wet soils (Fig. 4.6). Dambos are treeless seasonally waterlogged shallow channel-less drainage lines (a few tens to hundreds of meters in width) at the upper end of plateau drainage systems (Thomas & Goudie, 1985; Acres et al., 1985). Typically, they are surrounded by woodland/forest, except in the vleis of the South African High Veldt. In all cases, the herbaceous vegetation (primarily hydromorphic grasses) of the dambos provides important dry season pasture for grazers. Today, these natural glades of grass permeate the entire Sudanian and Zambezian woodland regions, especially the Zambezian (compare Figs. 4.5 and 4.6). During the Neogene, they may have been more extensive. Widely distributed edaphic grasslands of Africa, especially those associated with moderate to high rainfall (Xoodplains, and especially dambos) were probably an important feature of the African landscape during the Miocene and Pliocene. How do we discriminate these grasslands from grasslands associated with semi-arid to arid conditions? The existence of dambos emphasizes two points. First, increased aridity to produce grasslands in what are now woodlands or forest is not necessary to accommodate the movement of grazers between northern/eastern Africa and southern Africa during the Neogene. Second, changes in grass pollen in

Climatic perspectives for Neogene environment

[Figure 4.5] (a) Present-day African ecophysiognomic vegetation formations of regional extent: 1, tropical lowland rain forest; 2, broad-leaved woodland; 3, thorn bushland, wooded grassland and semi-desert vegetation; 4, desert; 5, Mediterranean sclerophyllous forest and Cape sclerophyllous shrubland; 6, Karoo-Namib semi-desert vegetation; 7, highveld grassland; 8, eastern coastal forest; 9, afromontane (including Afro-alpine) vegetation. Based on White (1984), modified somewhat by reference to CIA map 800630(547147)6-86. From O’Brien & Peters, n.d. (b) Inferred vegetation ecophysiognomy for Africa during the Late Miocene–Early Pliocene. Based on Axelrod & Raven (1978), modified somewhat in terms of montane rain forest. Numbers identify ecophysiognomic vegetation types of regional extent: 1, tropical lowland (and mid-elevation montane) rain forest; 2, broad-leaved woodland; 3, (high-elevation) montane rain forest to afroalpine; 4, subtropical laurel forest – a, Canarian; b, Cape; 5, sclerophyll vegetation – a, Tethyan; b, Cape; 6, thornscrub-succulent woodland – a, Sahelian; b, Kalaharian. Note numbering does not match that of Axelrod & Raven (1978). The hatched circles outline areas subjected to tectonic doming and rifting prior to the Pliocene that were not uplifted to their present-day elevations until the Late Pliocene–Early Pleistocene. The dashed lines suggest a hypothetical division of broad-leaved woodland by the Late Pliocene into subhumid woodland/forest vs. more seasonal or semi-arid summer and bimodal rainfall associated woodland/bushland. From O’Brien & Peters, n.d.

the paleobotanic record can be ambiguous with regard to prevailing moisture conditions. In Africa, grass pollen could increase as a function of increasing wetlands (due to changes in rainfall or terrain conditions), or as a function of increasing aridity (including mock aridity).

Elucidating paleoecological anomalies in species richness/vegetation Although climate and other environmental parameters can return to ‘normal’ following a disturbance (e.g. ice age cycle), the biota that originally

Palaeogeography of the circum-Mediterranean region

[Figure 4.6] (a) Distribution of dambos in Africa. Dark stippling identifies main areas of occurrence; light stippling, areas of sporadic occurrence. After Acres et al. (1985). From O’Brien & Peters, n.d. (b) Dambo grasslands associated with the Upper Kafue River system north of the Lukanga Marshlands (Zambia). After Michelmore (1939). From O’Brien & Peters, n.d.

occupied an area may not be able to return, even under ‘normalizing’ conditions. For that matter, some members of the original biota may not have successfully emigrated in the Wrst place. In other words, the disappearance of a particular species from an area where it is expected to be may indicate that local environmental parameters no longer permit it to be there, or that for some other reasons it has not been able to return. Based on climate’s relationship to woody-plant species richness in Africa, it is possible to predict/estimate Wrst-order variations in woody-plant species richness elsewhere in the world (O’Brien, 1998). This is exempliWed herein for the United States of America (hereinafter USA) (Fig. 4.7). The predicted values when compared with actual values do more than describe how well the model works; they also provide a tool for locating anomalies – where richness is grossly over- or under-predicted. In the USA, there is a surprisingly close Wt between predicted and actual values for woody-plant species richness, except in the northwest and southeast (see Fig. 4.7). In the northwest, gross over-prediction can be attributed

[Figure 4.7] Predicted woody-plant species richness across the United States of America, based on O’Brien’s (1998) interim general model of the Climatic Potential for Species Richness (CPSR). Demarcated areas are those of gross over-prediction. After O’Brien, 1998.

Palaeogeography of the circum-Mediterranean region

76

to the extremely high annual precipitation (due to the Rockies inhibiting the Xow of maritime air oV the PaciWc into the interior USA). In the southeast, however, gross over-prediction is not a function of climate, since all climate parameter values fall well within the range of values empirically analyzed when developing the predictive model of climate’s relationship to species richness. So, why should there be fewer than otherwise expected species, when elsewhere richness is in relative accord with its climatic potential? Over-prediction in the southeast suggests that the Xora in this region is impoverished. What seems to be missing is the subtropical-adapted neotropical species that could be here if they had not been eliminated from mainland USA with the onset of the ice ages (e.g. Daubenmire, 1978). This is consistent with the existence of major barriers to southward emigration and subsequent northward recolonization. The only sanctuaries/source areas for subtropical species would have been southern Mexico, Central America, the Caribbean Islands, and South America, but the Gulf of Mexico, deserts of northern Mexico and the Caribbean Sea are major barriers. By way of contrast, there are no barriers to equatorward emigration and poleward re-colonization by plants in Africa.

Summary and conclusions Water–energy dynamics are fundamental to how climate operates, how climate relates to vegetation (sensu its physiognomy, woody-plant species richness, biological activity, and productivity), and how climate/vegetation vary over space and time. In eVect, both the energy and water regimes need to be considered when evaluating paleoenvironmental changes, neither factor alone being necessarily a suYcient indicator. Moreover, given the parabolic function of energy’s relation to rainfall (best described by minimum monthly PET), increases/decreases in the energy regime (e.g. as a function of Milankovitch cycles) are both negatively and positively related to changes in climate/vegetation. What this highlights is that both decreases and increases in the energy regime can be associated with wetter or dryer climates/vegetation. It depends on whether the eVective moisture regime increases, decreases or remains the same. For Africa, the eVective moisture regime can remain the same or increase with decreases in the energy regime, promoting the same or lusher vegetation. At Africa’s high elevations, especially at high latitudes, cooler/colder conditions are likely to result in dryer climates/vegetation. Annual measures of the energy regime (temperature or potential evapotranspiration) appear to be unreasonable indicators of climate’s eVect on

Climatic perspectives for Neogene environment

continental-to-regional scale variations in woody-plant species richness, productivity, biological activity, or vegetation physiognomy in Africa. Annual measures are not signiWcantly correlated with geographic variations in these parameters today. It is unlikely to be otherwise for the past. In terms of describing changes in climate/vegetation based on changes in the paleobotanic record, a common assumption is that changes in grass pollen can be used as indicators of increases/decreases in aridity. In Africa, this assumption is dubious since changes in grass pollen can also be associated with increases/decreases in wetlands and moderate-to-high rainfall conditions. Lastly, there appears to be little, if any, evidence of change in Africa’s vegetation during the Pliocene that cannot be more simply attributed to the physiographic evolution of the continent than to global climate changes.

Acknowledgements The Wrst author wants to thank the organizers of the conference, especially Peter Andrews and Lorenzo Rook, for inviting her to attend, and the European Science Foundation for providing the funds.

References Acres, B. D., Rains, A. B., King, R. B., Lawton, R. M., Mitchell, A. J. B. & Rackham, L. J. 1985. African dambos: Their distribution, characteristics and use. In Dambos: Small Channelless Valleys in the Tropics – Characteristics, Formation, Utilisation. Thomas, M. F. & Goudie, A. S. (eds.). Zeitschrift fur Geomorphologie, Supplementband 52. Berlin, Gebruder Borntraeger. Axelrod, D. I. & Raven, P. H. 1978. Late Cretaceous and Tertiary vegetation history of Africa. In Biogeography and Ecology of Southern Africa, Vol. 31, M. J. A. Werger, (ed.), pp. 77–131. The Hague, Dr W. Junk. Barr, F.T. & Walker, B. R. 1973. Late Tertiary channel system in northern Libya and its Implications for Mediterranean sea-level changes. In Initial Reports, Deep Sea Drilling Project, W. D. K. Ryan et al. (eds.), 12 1244–50. Beadle, L. C. 1981. The Inland Waters of Tropical Africa: An Introduction to Tropical Limnology, 2nd Edition. New York, Longman. BonneWlle, R. & Hamilton, A. 1986. Quaternary and Late Tertiary History of Ethiopian Vegetation. Symb. Bot. Ups., 26(2), 48–63. BonneWlle, R., Vincens, A. & Buchet, G. 1987. Palynology, Stratigraphy and Palaeoenvironment of a Pliocene Hominid Site (2.9-3.3 m.y.) at Hadar, Ethiopia. Palaeogeography, Paleoclimatology, Palaeoecology, 60, 249–81. Burckle, L. H. 1995. Current Issues in Pliocene Paleoclimatology. In Paleoclimate and Evolution, with Emphasis on Human Origins, Vrba, E. S., Denton, G. H.,

77

Palaeogeography of the circum-Mediterranean region

78

Partridge, T. C. & Burckle, L. H. (eds.), pp. 319–31. Yale University Press, New Haven. Cahen, L. 1954. Geologie du Congo Belge. Vaillant-Carmanne, Lie`ge. Cooke, H. J. 1980. Landform evolution in the context of climatic change and neo-tectonism in the Middle Kalahari of north-central Botswana. Transactions, Institute of British Geographers, 5(1), 80–99. Cowling, R. (Ed.) 1992. The Ecology of Fynbos: Nutrients, Fire and Diversity. Oxford University Press, Cape Town. Crowley, T. J. & North, G. R. 1991. Paleoclimatology. Oxford University Press, Oxford. Daubenmire, R. 1978. Plant Geography. Academic Press, New York. de Heinzelin, J. 1963. Addendum to Transcript of Discussions. In African Ecology and Human Evolution, Howell, F. C. & Bourliere, F. (eds.), pp. 648–54. Aldine Publishing Company, Chicago. de Heinzelin, J. & Arnauti, A. 1987. The Sahabi Formation and Related Deposits. In Neogene Paleontology and Geology of Sahabi, Boaz, N. & de Heinzelin, J. (eds.), pp. 1–21. Alan R. Liss, Inc., New York. Deacon, H. J., Jury, M. R. & Ellis, F. 1992. Selective regime and time. In The Ecology of Fynbos: Nutrients, Fire and Diversity, R. M. Cowling, (ed.), pp. 6–22. Oxford University Press, Cape Town. deMenocal, P. B. & Rind, D. 1993. Sensitivity of Asian and African Climate to Variations in Seasonal Insolation, Glacial Ice Cover, Sea Surface Temperature, and Asian Orography. Journal of Geophysical Research, 98(D4), 7265–87. Dingle, R. V., Siesser, W. G. & Newton, A. R. 1983. Mesozoic and Tertiary Geology of Southern Africa. A. A. Balkema, Rotterdam. Dupont, L. & Leroy, S. 1995. Steps toward Drier Climatic Conditions in Northwestern Africa during the Upper Pliocene. In Paleoclimate and Evolution, with Emphasis on Human Origins, Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, L. H. (eds.), pp. 289–98. Yale University Press, New Haven. Eddy, J. A. & Oeschger, H. 1993. Global Changes in the Perspective of the Past. John Wiley & Sons, New York. Fullard, H. (ed.) 1980. Philips’ Large Print Atlas for Southern Africa, 42nd Edition. George Philip & Son Ltd., London. GriYths, J. F. 1976. Climate and the Environment. Westview Press, Boulder. Grotjahn, R. 1993. Global Atmospheric Circulations: Observations and Theories. Oxford University Press, Oxford. Harris, J. & Van Couvering, J. 1995. Mock aridity and the paleoecology of volcanically inXuenced ecosystems. Geology, 23, 593–6. Herlocker, D. J., Dirschl, H. J. & Frame, G. 1993. Grasslands of East Africa. In Natural Grasslands: Eastern Hemisphere & Resume, R. T. Coupland (ed.), pp. 221–64. London, Elsevier Science Publishers. Jackson, S. P. 1961. Climatological Atlas of Africa. CCTA/CSA, Government Printer, Pretoria. King, L. 1978. The geomorphology of central and southern Africa. In Biogeography and Ecology of Southern Africa, M. J. A. Werger (ed.), 31, 1–17. Dr. W. Junk, The Hague. Kukla, G. & Cilek, V. 1996. Plio-Pleistocene megacycles: Record of climate and tectonics. Palaeogeography, Palaeoclimatology, Palaeoecology, 120(1–2), 171–94.

Climatic perspectives for Neogene environment

Le Houerou, H. N. 1993. Grasslands of the Sahel. In Natural Grasslands: Eastern Hemisphere & Resume, R. T. Coupland, (ed.), pp. 197–220. London, Elsevier Science Publishers. Leroy, S. & Dupont, L. 1994. Development of vegetation and continental aridity in northwestern Africa during the Late Pliocene: the pollen record of ODP Site 658. Palaeogeography, Palaeoclimatology, Palaeoecology, 109, 295–316. Lutkens, F. K. and Tarbuck, E. J. 1992. The Atmosphere, 5th Edition. Prentice-Hall, Englewood CliVs, NJ. Malay, J. 1980. Les changements climatiques de la Wn du Tertiare en Afrique: leur consequence sur l’apparition du Sahara et de sa vegetation. In The Sahara and the Nile, Williams, M. A. J. & Faure, H. (eds.), pp. 63–86. A. A. Balkema, Rotterdam. Martin, P. M. & O’Meara, P. 1986. Africa, 2nd Edition. Indiana University Press, Bloomington. Michelmore, A. P. G. 1939. Observations on Tropical African Grasslands. Journal of Ecology (British), 27, 282–330. Molnar, P. & England P. 1990. Late Cenozoic uplift of mountain ranges and global climate change: Chicken or egg? Nature, 346, 29–34. O’Brien, E. M. 1993. Climatic Gradients in Plant Species Richness: Towards an Explanation based on an Analysis of Southern Africa’s Woody Flora. J. Biogeography, 20, 181–98. O’Brien, E. M. 1998. Water-energy dynamics, climate, and prediction of woody plant species richness: An Interim General Model. Journal of Biogeography, 25, 379–98. O’Brien, E. M. & Peters, C. R. (n.d.). Landforms, Climate, Ecogeographic Mosaics, and the Potential for Hominid Diversity in Pliocene Africa. In African Biogeography, Climate Change, and Early Hominid Evolution, Bromage, T. & Schrenk, F. (eds.). Oxford University Press, Oxford. O’Brien, E. M. & Peters, C. R. (in prep. c). Climate’s relationship to the species richness of wild fruit trees and shrubs: a useful predictor of their distribution and diversity across the continent of Africa. O’Brien, E. M., Whittaker, R. J. & Field, R. 1998. Climate and woody plant diversity in southern Africa: relationships at species, genus and family levels. Ecography, 21, 267–78. O’Brien, E. M., Whittaker, R. J. & Field, R. (in prep. a). Topography, climate, and diversity in southern Africa. O’Brien, E. M., Field, R. & Whittaker, R. J. (in prep. b). Topography, climate, and a revised interim general model for predicting diversity (working title). Olsen, P. E. & Kent, D. V. 1996. Milankovitch climate forcing in the tropics of Pangaea during the Late Triassic. Palaeogeography, Palaeoclimatology, Palaeoecology, 122, 1–26. Partridge, T. C., Bond, G. C., Hartnady, C. J. H., deMenocal, P. B. & Ruddiman, W. F. 1995a. Climatic EVects of Late Neogene Tectonism and Volcanism. In Paleoclimate and Evolution, with Emphasis on Human Origins, Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, L. H. (eds.), pp. 8–23. Yale University Press, New Haven. Partridge, T. C., Wood, B. A. & deMenocal, P. B. 1995b. The InXuence of Global Climatic Change and Regional Uplift on Large-Mammalian Evolution in East

79

Palaeogeography of the circum-Mediterranean region

80

and Southern Africa. In Paleoclimate and Evolution, with Emphasis on Human Origins, Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, L. H. (eds.), pp. 331–55. Yale University Press, New Haven. Peters, C. R. 1990. African wild plants with rootstocks reported to be eaten raw: the monocotyledons, part 1. Mitteilungen aus dem Institut fu¨r Allgemeine Botanik Hamburg, 23, 935–52. Pettars, S. W. 1991. Regional Geology of Africa. Springer-Verlag, New York. Prell, W. L. & Kutzbach, J. E. 1992. Sensitivity of the Indian monsoon to forcing parameters and implications for its evolution. Nature, 360, 647–52. Pritchard, J. M. 1979. Landform and Landscape in Africa. London, Edward Arnold. Raymo, M. E. & Ruddiman, W. F. 1992. Tectonic forcing of Late Cenozoic climate. Nature, 359, 117–22. Ruddiman, W. F. 1997. Tectonic Uplift and Climate Change. Plenum Press, New York. Ruddiman, W. F. & Kutzbach, J. E. 1989. Forcing of Late Cenozoic Northern Hemisphere climate by plateau uplift in southern Asia and the American West. Journal of Geophysical Research, 94, 18409–27. Ruddiman, W. F. & Kutzbach, J. E. 1991. Plateau uplift and climate change. ScientiWc American, 265, 66–76. Suckling, P. & Doyan, P. 1981. Weather and Climate – Laboratory Manual. Raleigh, NC. Tainton, N. M. & Walker, B. H. 1993. Grasslands of Southern Africa. In Natural Grasslands: Eastern Hemisphere & Resume, R. T. Coupland (ed.), pp. 265–90. London, Elsevier Science Publishers. Thomas, M. F. & Goudie, A. S. (eds.) 1985. Dambos: Small Channelless Valleys in the Tropics – Characteristics, Formation, Utilisation. Zeitschrift fu ¨r Geomorphologie, Supplementband 52. Berlin: Gebruder Borntraeger. Thornthwaite, C. W. & Mather, J. R. 1962. Average Climatic Water Balance Data of the Continents, Part I. Africa. Publications in Climatology, 15(2). Laboratory of Climatology, Centerton, New Jersey. Thornthwaite, C. W. & Mather, J. R. 1964a. Average Climatic Water Balance Data of the Continents, Part VII. USA. Publications in Climatology, 17(3). Laboratory of Climatology, Centerton, New Jersey. Thornthwaite, C. W. & Mather, J. R. 1964b. Average Climatic Water Balance Data of the Continents, Part V. Europe. Publications in Climatology, 17(1). Laboratory of Climatology, Centerton, New Jersey. Thornthwaite, C. W. & Mather, J. R. 1965. Average Climatic Water Balance Data of the Continents, Part VIII. South America. Publications in Climatology, 18(2). Laboratory of Climatology, Elmer, New Jersey. Tiedemann, R., Sarnthein, M. & Shackleton, N. J. 1994. Astronomic timescale for the Pliocene Atlantic &18O and dust Xux records of ODP Site 659. Paleoceanography, 9, 619–38. Tyson, P. D. 1986. Climatic Change and Variability in Southern Africa. Oxford University Press, Oxford. Vincens, A. 1987. Environments botanique et climatique des hominides de l’Est Turkana, Kenya, entre 2,0 et 1,4 millions d’anne´es: Apport de la palynologie. In Palaeoecology of Africa and the Surrounding Islands, J. A. Coetzee (ed.), 18, 257–69. Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, L. H. 1995. Paleoclimate and Evolution, with Emphasis on Human Origins, pp. 8–23. Yale University Press,

Climatic perspectives for Neogene environment

New Haven. Walter, H. 1971. Ecology of Tropical and Subtropical Vegetation (transl. D. Mueller-Dombois), J. H. Burnett (ed.). New York, Van Nostrand Reinhold. Walter, H., Lieth, H. & Rehder, G. 1960–67. Klimadiagramm-Weltatlas. Fischer, Jena. Werger, M. J. A. & Coetzee, B. J. 1978. The Sudano-Zambezian Region. In Biogeography and Ecology of Southern Africa, M. J. A. Werger (ed.), pp. 301–462. Dr. W.Junk, The Hague. White, F. 1983. The Vegetation of Africa: A Descriptive Memoir to Accompany the UNESCO/AETFAT/UNSO Vegetation Map of Africa, p. 356. Paris, UNESCO. White, F. 1984. Vegetation Map of Africa. Africa South of the Sahara 1984–1985, 14th Edition. Europa Publications Ltd. Williamson, P. G. 1985. Evidence for an early Plio-Pleistocene rainforest expansion in East Africa. Nature, 315, 487–9. Yemane, K., BonneWlle, R. & Faure H. 1985. Paleoclimatic and tectonic implications of Neogene microXora from the north western Ethiopian highlands. Nature, 318, 653–6. Yemane, K., Robert, C. & BonneWlle, R. 1987. Pollen and clay mineral assemblages of a Late Miocene lacustrine sequence from the Northwestern Ethiopian Highlands. Palaeogeography, Palaeoclimatology, Palaeoecology, 60, 123–41. Yuretich, R. F. 1982. Possible inXuences upon lake development in the East African Rift Valleys. Journal of Geology, 90, 329–37.

81

XXXXXX

PART II

Miocene mammalian successions

5 A critical re-evaluation of the Miocene mammal units in Western Europe: dispersal events and problems of correlation Jorge Agustı´

Introduction The MN mammalian biochronological classiWcation of the European terrestrial record has been the subject of intensive criticism since its Wrst formulation by Mein (1975). More than twenty years later, the question of its validity is still open, although, unlike previous times, we now have ways for falsifying this ‘hypothesis of mammalian succession’ (as it was deWned by Agustı´ & Moya`-Sola`, 1991). Originally, the MN ‘zonation’ (as it was sometimes wrongly quoted) reXected a pattern of mammalian turnover in the European terrestrial record, mainly based on supposed large overland dispersals and extinctions. To these two criteria, Mein (1975) also added the successive chronospecies of several European mammalian lineages (mainly rodents) as a third criterion for establishing the MN units. Agustı´ & Moya`-Sola` (1991) proposed a restrictive interpretation of the MN units as a regional, Western European mammal scale based on the stratotypical succession of the Neogene mammal stages deWned in a number of Spanish basins (by far, the most complete Neogene succession in Western Europe). A very diVerent approach was that of a reinvented ‘RCMNS working group’ (Bruijn et al., 1992), which proposed to avoid completely the assumed biostratigraphic meaning of the MN units, and to redeWne them on the basis of referencelevels, returning to the system proposed in the 1960s by Thaler (1966), and that the Mein (1975) proposal was supposed to surpass. Each MN unit being deWned only by an isolated reference-locality, this system remained as a selWsh way for paleomammalogists to classify their fossiliferous localities within an assumed chronological background, but without any possibility of falsiWcation or correlation with other biostratigraphic scales, given the absence of a true biostratigraphical background. This anomaly is best exempliWed by the fact that, as established by Bruijn et al. (1992), the MN units lack boundaries by deWnition. Moreover, recent progress in the chronostratigraphic resolution of a number of sections of Spain and Greece enables one to falsify in some cases the chronological assumptions based on this MN succession: localities included in diVerent MN units have in fact the same age, as in the case of the MN 10 localities of Terrassa and the MN 11 localities of Lower Maragheh (Agustı´ et al., 1997; Bernor et al., 1996; Garce´s et al., 1997). In this paper, we propose to re-open the discussion on the MN unit

Miocene mammal units in Western Europe

system and to re-deWne and correlate them in terms of faunal dispersals and extinctions (FAD and LAD) based on the type-sections of the Miocene Mammalian Stage System established for Western Europe (Ramblian, Aragonian, Vallesian, Turolian).

Problems of correlation Correlation among basins is one of the main problems in the study of the mammalian succession in the Eurasian Neogene, especially when we are dealing with rather distant areas. Concentrations of fossil mammals occur in discrete units (‘localities’) which can be more or less directly correlated when they belong to the same basin. However, the problems appear when we try to work at the suprabasinal level. The work undertaken in the Spanish Neogene provides a good basis for an analysis of the main factors involved in the correlations among basins. As is widely known, the Neogene in the Iberian Peninsula appears associated to a number of main tectonic folds (Duero, Tagus, Calatayud-Teruel, Valle`s-Penede`s) which during most of their history acted as endorreic basins. The problems dealing with the correlation among basins can be assembled into Wve categories.

Development of endemic lineages The development of endemic lineages in a given area is one of the factors that can disturb the correlation between supposed synchronous sequences. This problem seems not very important in the case of the large mammalian faunas, although it has a high incidence in the case of the small mammals (mainly rodents). In our opinion, there is a general pattern underlying the appearance of endemic forms. This pattern can be traced both for continental or island environments and, in general terms, could be enunciated in the following way. In a Wrst phase, after an environmental or tectonic change, a given area is normally colonised by opportunistic species which used to display a wide geographic rank. Following this Wrst phase of settlement, the peculiar environmental conditions in each area led to the segregation of the original stock into several vicariant species. Finally, the posterior environmental evolution would introduce progressive modiWcations in the diVerent lineages (‘anagenesis’).

85

Miocene mammalian successions

86

While the possibilities of long distance correlation are high during the Wrst phase, the problems appear in the case of the other two. Especially signiWcant in this way is the example of the late Aragonian rodent faunas from Europe. Therefore, while during the early Aragonian there is a high degree of congruence at the speciWc level among the diVerent bioprovinces, this congruence diminished in the late Aragonian (although it is maintained at the generic level). For instance, the very frequent cricetid Megacricetodon appears split into two or three diVerent lineages (M. crusafonti–M. ibericus in the Iberian peninsula, M. gregarius in Western Europe, M. germanicus in Central Europe; see Aguilar, 1980). An even more signiWcant case is that of Hispanomys, a semi-hypsodont cricetid which in the late Aragonian/Vallesian is usually represented by one or two species per basin. Therefore, endemic species of Hispanomys haven been described in Calatayud-Daroca (H. nombrevillae, H. aragoniensis), Valle`s-Penede`s (H. dispectus, H. thaleri), Central and Southern France (H. decedens, H. bijugatus, H. mediterraneus), etc. Moreover, a variant of Hispanomys, Turkomys, is present in Eastern Europe, also represented by diVerent lineages and species. Frequently, the biostratigrapher faces a picture in which most of the species are endemic variants, while the remaining elements are catholic taxa displaying very low degrees of variation in time and space (and, of course, with a limited value for biostratigraphic correlation).

The problem of the synchronism of the faunas One of the main problems for correlating basins deals with the question of whether two faunas could be considered synchronous or not. This problem is present even in the case of apparently similar faunas, since our work is based on the assumption that the species always change at a constant rate. Nevertheless, the more frequent question is whether two faunas displaying diVerent faunal associations could be considered synchronous or not. This problem is particularly evident in the case of the late Turolian (‘Ventian’) faunas from Southern Europe. Thus, no apparent change took place in the Greek–Turkish province during the Messinian salinity crisis (Bruijn, 1988; Unay & Bruijn, 1984). On the contrary, the late Turolian sites in Western Europe (mainly Spain and Southern France) show a very diVerent picture, with the entry of many eastern immigrants of African or Asiatic origin such as Protatera, Myocricetodon, Calomyscus, Pseudomeriones, etc. (see Aguilar et al., 1983, and Agustı´, 1990). An explanation for this peculiar distribution was proposed by Agustı´ (1989) in the following way: since the connection between the Black Sea and the Eastern Mediterranean remained open even during the late Miocene, a continuous drainage into the last basin persisted during the Messinian. In this context, the humid faunas from the

Miocene mammal units in Western Europe

Greek and Turkish Turolian also persisted during the Messinian. On the other hand, the eVects of the Messinian crisis could have been much harder in the Western Mediterranean basin, isolated both from the Atlantic and the Eastern Mediterranean basin. Nevertheless, there is an alternative explanation based on the possible diachroneity of the faunas from both regions. In fact, the late Turolian, widely represented in many sites of Spain, could be absent in the Greek successions, which in fact would be still middle Turolian. Although several lines of evidence suggest that the Wrst explanation is correct, this second one cannot be discarded as a possibility. This case illustrates very clearly the kind of problems we deal with in trying to test the assumed synchroneity of two faunas. The presence of environmental, latitudinal and altitudinal factors will be discussed below.

Environmental factors Environmental diVerences could cause also many correlation problems. Our analysis will be centred at two diVerent levels.

Intrabasinal level Even at this level, diVerent local environmental conditions can aVect profoundly the composition of the mammalian associations. Thus, in the Valle`s-Penede`s Basin the existence of at least two very diVerent biotopes during the late Aragonian and Vallesian, corresponding to the two sectors (Valle`s and Penede`s) of the basin, was already recognized among the large mammals by Villalta (1952). This was congruent with the very diVerent sedimentological characteristics of these two sectors: Xood plain associated to a braided river system in the case of the Penede`s, palustrine and lacustrine beds associated to distal segments of aluvial fans in the case of the Valle`s. Among the rodent associations, the presence of two diVerent biotopes is illustrated by the almost perfect exclusion between the Cricetinae Democricetodon and Cricetulodon and the Cricetodontinae Megacricetodon ibericus (Agustı´, 1990, table 2). This ecological exclusion can be also referred to other basins such as Calatayud–Teruel (for instance, in the case of the early Vallesian localities of Nombrevilla, with Megacricetodon ibericus, and Pedregueras IIC, with Cricetulodon hartenbergeri).

Suprabasinal level The presence of two biogeographic subprovinces in the Iberian peninsula during the Miocene was established after the analysis of the rodent content

87

Miocene mammalian successions

88

in the Valle`s-Penede`s and other eastern basins, which showed a very diVerent composition when compared with the Iberian Central basins (Calatayud–Daroca, Teruel, Duero, Tagus; Agustı´, 1978). In this way, many Iberian endemic taxa such as Armantomys, Praearmantomys and Tempestia were absent from the Ibero-Levant subprovince. On the contrary, the eastern basins were characterized by many western and central European elements, absent in the former basins: Anomalomys, Eumyarion, Glis, Paraglirulus, Eomuscardinus, etc. This provincial diVerentiation persisted during most of the Aragonian and Vallesian, although a uniformization phase took place at the beginning of the Turolian (Agustı´, 1990). In general, the highest coincidence between the two subprovinces is detected during the dry phases.

Latitudinal factors The faunal succession in diVerent basins can be strongly inXuenced by their latitudinal position. This inXuence appears to be much more accentuated in the moments of climatic change, when a clear gradient in the faunal exchange is sometimes evident. Although the latitudinal component had the higher inXuence in the case of the Pleistocene glaciations, some other examples not directly related to a decrease in the temperature can be quoted. This is the case, for instance, for the Mid Vallesian Crisis, one of the most characteristic crises in the Spanish Neogene. This Vallesian event, placed in the boundary between the early and late Vallesian, aVected many diVerent orders of mammals, involving the disappearance of many of the taxa from the Iberian Miocene record. Thus, the early/late Vallesian transition profoundly aVected the three dominant rodent families in the early and middle Miocene: Cricetids, Eomyids and Glirids. In that moment, up to eight diVerent genera belonging to these families disappeared from the fossil record of the Iberian peninsula. On the other hand, a new element, the murid Progonomys, became very abundant in the diVerent basins. After the Vallesian Crisis, the murids were the dominant element in the rodent taxocenosis during the late Miocene and the early Pliocene. Nevertheless, other orders of micromammals, such as the lagomorphs and some insectivores were slightly or not aVected by this crisis. Among the large mammals, the transition from the early to the late Vallesian involved the disappearance of many of the elements of wet character coming from the middle Miocene. This event, which so profoundly aVected the late Miocene Iberian biotopes, had a very diVerent character when we consider the late Vallesian and early Turolian faunas from other areas in Europe. Therefore, some of the

Miocene mammal units in Western Europe

rodent taxa already extinct in the Iberian peninsula persisted in the early Turolian from Central Europe (for instance, the glirids Paraglirulus and Myoglis and the eomyids Keramidomys and Ecomyops; Franzen & Storch, 1975, and this volume). Thus, the latitudinal character of the Vallesian event seems evident. However, this latitudinal eVect displayed a ‘reversed’ gradient, since the change started in the lower latitudes and went on to the North. Probably, this anomaly can be explained by the fact that the Vallesian crisis was not linked to a drop in the temperature but to the establishment of progressively drier conditions around the Mediterranean. Once the Wrst eVects of the desiccation aVected the Mediterranean basins, the change extended northwards.

Altitudinal factors Besides environment and latitude, a third factor, the altitude, must be considered. This feature is very rarely detected, since most of our localities appear in endorreic basins which in the past acted as lowland plains. Nevertheless, the peculiar associations present in some intramountainous basins can shed some light upon this question. This is the case of the locality of Can Vilella, in the small Pyrenaic basin of La Cerdanya (northern Catalonia). This site delivered a late Turolian faunule containing Apodemus aV. primaevus Mein, Kowalskia aV. lavocati, Epimeriones aV. austriacus Daxner-Ho ¨ ck, Sminthozapus janossy Sulimsky, Muscardinus aV. vireti Mein et Hugueney and Prolagus michauxi Lopez. This association strongly diVers from other late Turolian assemblages in Spain, such as Librilla or La Alberca. For instance, Sminthozapus appear very rarely in some early Pliocene sites in the Teruel basin (Mein et al., 1983). However, the most striking element is Epimeriones, a hypsodont cricetid of supposed gerbilloid aYnities which is present in the Turolian-Ruscinian succession from Central Europe (Bachmayer & Wilson, 1970). Can Vilella is the most southern citation of this genus in Europe. Therefore, we face in Can Vilella a very peculiar assemblage (Epimeriones aV. austriacus, Sminthozapus janossyi, Muscardinus aV. vireti) which seems much closer to the Central European sites than those of the Iberian Peninsula. Given the continental character of this association, we can infer their presence in La Cerdanya to be a consequence of the position of this basin within the Pyrenaic range. Thus, Epimeriones and Sminthozapus would represent cold elements occurring in Can Vilella because of the higher altitude of this locality with respect to other late Turolian sites. Therefore, correlation between basins can be strongly biased by diVerent factors. Among them, the development of endemic lineages in some mo-

89

Miocene mammalian successions

90

ments of their history and the existence of particular environmental conditions have the highest incidence on this kind of analysis. To a lesser degree, altitude and latitude can also produce peculiar faunal associations.

Problems in the early Miocene: the need for an extended Ramblian The biozonation and correlation of the early Miocene European mammalian record appears as one of the main unsolved questions in the frame of the standard biochronologic schemes. The early Miocene MN units were mainly based on the rodent succession and, more precisely, on the evolutionary lineages of the genera Eucricetodon, Melissiodon and RhodanomysRitteneria. In fact, in the case of rodents, most of these units can be viewed as a true succession of ‘range-zones.’ This is the case, for instance, of the units MN 0–MM 1–MN 2 (Rhodanomys transiens – Rh. schlosseri – Ritteneria manca) and MN 2a–MN 2b–MN 3 (Eucricetodon gerandianus – E. aquitanicus – E. infralactorensis). The combination of some of these ‘rangezones’ leads to a formal characterization of the MN units as concurrent range-zones: MN 1: Rhodanomys schlosseri + Melissiodon shro¨deri MN 2a: Eucricetodon gerandianus + Melissiodon schlosseri MN 2b: Eucricetodon aquitanicus + Ritteneria manca MN 3: Eucricetodon infralactorensis + Ligerimys

The Munich Symposium in 1975 proposed the deWnition of a number of mammal stages in which the MN ‘zones’ would be included. Therefore, the stage ‘Agenian’ was proposed for covering the time-span represented by ‘zones’ MN 1 and 2. Nevertheless, since then, no stratotype has been proposed and, in the meantime, Daams et al. (1987) established in the Daroca Basin the new mammal stage ‘Ramblian’ to cover part of the early Miocene. The Ramblian was not proposed as a direct alternative to the unborn ‘Agenian’, since Daams et al. (1987) placed the base of this stage at the base of their zone Z, characterized by the exit of the eomyids of the RhodanomysRitteneria group. This choice excluded MN 1 and part of MN 2 from the range of the Ramblian. It must be pointed out, however, that this restricted conception of the Ramblian was probably related to the fact that the earliest Miocene is lacking in the type area of the Calatayud-Daroca Basin. In an opposite direction, the original criterion for establishing the lower boundary of the middle Miocene Aragonian stage, the entry of Anchitherium, was changed by Daams et al. (1987), since this equid was proven to be very rare in the original sequence of the Calatayud-Daroca type area. Therefore,

Miocene mammal units in Western Europe

Daams et al. (1987) decided to place the Ramblian/Aragonian boundary at the base of their zone B, characterized by the entry of the Wrst ‘Miocene cricetids’ of the genus Democricetodon. In practice, that implied inclusion also of the MN 3 unit within the Ramblian. Therefore, the Ramblian original sequence can be correlated to the upper part of MN 2 and the MN 3. However, there are a number of reasons that would favour an extension of the original Ramblian to the very early Miocene. The main one for proposing an extended Ramblian is the weakness and low biological value of its lower boundary. Although the MN 2–MN 3 boundary represents one of the most signiWcant bioevents in the history of Europe (entry of the Wrst proboscideans), this fact is obscured within the Ramblian sequence, only deWned by the extinction of an old and quite restricted eomyid group (RhodanomysRitteneria) and the gradual transformation of Pseudotheridomys into Ligerimys. But, in fact, other basins in Spain present sections covering the Oligocene–Miocene boundary and the lowermost part of the Miocene which is lacking in the original Ramblian section at Daroca (for instance, Ebro and Loranca basins). Moreover, a high degree of congruence is found among these sections and those previously published in other Western European areas. Thus, a similar pattern of cricetid and eomyid evolution can be found from Central Spain to Western and Central Europe on the basis of the evolutionary changes observed in the genera Eucricetodon, Melissiodon and the Rhodanomys–Ritteneria group. For instance, the genus Eucricetodon, created by Thaler (1966) for the species E. collatus from Ku ¨ ttingen, is a characteristic element of several early Miocene sections in Western Europe. Other early Miocene species assigned to this genus were E. gerandianus (Gervais), E. aquitanicus (De Bonis) and E. infralactorensis (Viret) (the last member of the lineage). The assignment of the specimens from the French locality of Paulhiac (the type locality for MN 1) to E. collatus by Thaler (1966) was rejected by Engesser (1985), who included the material from this site in the new species Eucricetodon hesperius. This species, as well as other members of the Eucricetodon lineage, are also present in the Loranca and Ebro basins (Lacomba, 1988), according to the following succession: Eucricetodon collatus (zone W)–E. hesperius (zone X)–E. gerandianus (zone Y1)–E. cf. aquitanicus (zone Y2). This succession can be complemented by that of the Calamocha area (Sese´, 1987): E. aV. aquitanicus (similar to or slightly larger than that of Bouzigues)–E. aV. infralactorensis. Moreover, the zones X, Y and Z originally deWned by Daams et al. (1987) in the Loranca basin were redeWned by Alvarez (1987) on the basis of the eomyid succession. With respect to the previous succession established in Southern France (which was used as a basis for the early Miocene MN units; see Mein, 1975), this author established the following succession: Rhodan-

91

Miocene mammalian successions

92

omys transiens (zone X)–R. schlosseri (lower Y1)–R. oscensis (upper Y1)– Ritteneria molinae (lower Y2)–R. manca (zone Y2). Therefore, according to the following eomyid and cricetid succession: X: Rhodanomys transiens + Eucricetodon hesperius Y1: Rhodanomys schlosseri (or R. s. oscensis) + Eucricetodon gerandianus Y2: Ritteneria manca (or R. m. molinae) + E. cf. aquitanicus (or E. gerandianus) Z: Ligerimys spp. + E. aV. infralactorensis

The Loranca and Ebro sequences can be integrated within the original concept of the early Miocene MN units by referring the MN 1 unit to zones X and Y1 and the MN 2 unit to zones Y2 and Z. MN 1 In the original scheme of Mein (1975), followed by Bruijn et al. (1992), the deWnition of this zone was based on the faunal content of the French locality of Paulhiac. However, Agustı´ & Moya`-Sola` (1991) proposed an alternative deWnition, by correlating this unit with the Rhodanomys schlosseri zone of the Ebro Basin (Agustı´ et al., 1987) and the zone Y1 of the Loranca Basin (Daams et al., 1987), based on the fact that Rhodanomys schlosseri was one of the most characteristic species of Paulhiac. Later analysis has shown that several localities included in MN 1 still present Rhodanomys transiens and that the cricetid species accompanying Rhodanomys schlosseri was not Eucricetodon hesperius (present at Paulhiac), but the more evolved Eucricetodon gerandianus. Therefore, besides the Rhodanomys schlosseri zone, the MN 1 unit should cover also part of the Rhodanomys transiens zone (or the zones W, X and Y1 of the Loranca Basin). The real problem is that, in contrast with other parts of the Neogene record, there is not a clear criterion for separating the uppermost Oligocene terrestrial levels from those of the lowermost Miocene, the use of particular segments of some evolutionary lineages excepted (but see earlier for a discussion of the problems associated with the use of chronospecies). However, there is a criterion which can be at least useful for Western Europe in order to establish a terrestrial boundary between the two systems, that is, the spread of the glirids of the genus Vasseuromys (the glirids were absent from the ‘resurrected biozonation’ proposed by Bruijn et al., 1992). This genus was created in the locality of Laugnac (included in the MN 2b) to designate small glirids with very complicated dental patterns. Vasseuromys is found in several early Miocene Western European localities such as La Chaux, ranging from MN 1 to MN 2 (Baudelot & de Bonis, 1966; Engesser & Modden, 1997). We consider here the genus Vasseuromys in a restricted way, to include only forms with molarized premolars and complicated dental pattern composed of several

Miocene mammal units in Western Europe

ridges of the same width and continuous endolophid (for instance, we do not include here ‘Vasseuromys’ thenii Daxner-Ho ¨ ck & Bruijn, from the late Vallesian of Eichkogels, which is in fact a convergent form related to Miodyromys aegerci). In the Ebro Basin, where the Oligocene–Miocene boundary has been calibrated by magnetostratigraphic correlation (Barbera` et al., 1994; Agustı´ et al., 1984), a primitive member of Vasseuromys appears abruptly associated to Rhodanomys transiens in the localities of Ballobar-21 and Costa-Sans (Agustı´ et al., 1994). This species is also present in other levels of the Western Ebro Basin (Vasseuromys rugosus = Ebromys autolensis from Bergasa, Autol and Santa Cilia) associated to Rhodanomys oscensis (Cuenca, 1985; Canudo et al., 1994). This is why a Vasseuromys zone was erected by Agustı´ et al. (1984) in order to cover this part of the Ebro sequence. Vasseuromys sp. from the Ebro Basin is not a descendant in situ of other glirids present in the late Oligocene levels of that area (such as Peridyromys murinus or Miodyromys hugueneyae). Its sudden appearance is therefore the result of a dispersal event, as suggested also by the faunal composition of the latest Oligocene faunas in Western Europe. The dispersal of Vasseuromys has also an ecological meaning, as indicated by its complicated dental pattern which suggests a diet based on fruits and soft leaves (‘complicated intermediate group’ of Meulen & Bruijn, 1982) associated to the spread of wooded conditions during the earliest Miocene. The paleomagnetic analysis carried out in the Eastern Ebro Basin (Barbera` et al., 1994; Agustı´ et al., 1994) led to two diVerent magnetobiostratigraphic interpretations, a purely paleomagnetic Option I and a second Option II which Wtted better with the biostratigraphic data from the Swiss Molasse. In both cases, the boundary between the Rhodanomys transiens zone (without Vasseuromys) and the Vasseuromys zone (with Rhodanomys transiens and Vasseuromys) was very close to the Oligocene–Miocene boundary (placed at the base of chron C6Cn.2n). Therefore, the Vasseuromys dispersal event is almost coincident with the Oligocene–Miocene boundary (the base of the Vasseuromys zone is certainly placed at the top of C6Cn.2n, but it could be also placed at the base of this chron if the presence of Vasseuromys in the locality of Torrente de Cinca 68 is conWrmed). Therefore, according to the deWnition proposed here, the base of the MN 1 is coincident with the base of the Miocene in absolute terms.

MN 2a According to the equivalences established earlier, we propose to deWne the MN 2 on the basis of zones Y2 and Z (MN 2a and MN 2b, respectively) of the Loranca and Daroca basins (Daams et al., 1987). The bipartition of this MN

93

Miocene mammalian successions

94

unit (MN 2a and MN 2b) has been a constant since its early deWnition, not being even discarded in the last proposal by Bruijn et al. (1992). This is conWrmed in the Iberian basins by the diVerent composition of zones Y2 and Z and associated large mammal localities (Cetina de Argo ´ n and Valquemado, on the one hand, the Loranca and Navarrete del Rı´o, on the other hand; Adrover, 1978; Daams, 1976; Daams et al., 1987; Made, 1994). The original division of the MN 2 was based on the chronospecies Eucricetodon gerandianus and Eucricetodon aquitanicus. However, while the holotype of Eucricetodon gerandianus is lost, the systematics of the early Miocene Eucricetodon species has proven to be more complicated than initially thought, as demonstrated in the Daroca area and the Swiss Molasse (Sese´, 1987; Engesser, 1985). Therefore, we prefer to base the partition of this unit in the two zones recognized in the Loranca and Teruel basins. In this way, the unit MN 2a includes the localities of the Y2 zone such as Cetina de Arago ´ n, Valquemado and others (Daams, 1976; Daams et al., 1987). We include within the MN 2b unit the localities of the Z zone of the Daroca and Loranca basins: Navarrete del Rio, Loranca, Ramblar 1 to 7, Valhondo 1 and 3a (Adrover, 1978; Daams et al., 1987; Moya`-Sola`, 1987). The localities of the MN 2a unit (= zone Y2) still present members of the genus Ritteneria (R. molinae or R. manca) associated with Eucricetodon cf. aquitanicus. The presence of Ritteneria manca can be recognized over a large geographic range, covering the Ebro and Loranca basins in Spain (Alvarez, 1987) as well as several localities in France (Aguilar, 1981) and the Swiss Molasse (Engesser & Modden, 1997). Among the large mammals, this unit records the FAD of the Bovoidea Andegameryx GINSBURG (Morales & Soria, 1984). Another artiodactyl present in MN 2a, Amphitragulus, is in fact an authochtonous lineage with roots in the Oligocene (Moya`-Sola` & Agustı´, 1990). The localities included in this unit can be correlated with the site of Saint Gerard.

MN 2b Among the small mammals, localities included in the MN 2b unit (= zone Z of the Aragonian type-section; Daams et al., 1987) are characterized by the absence of eomyids of the Ritteneria group, replaced by the Wrst member of the genus Ligerimys (a Western European descendant of Pseudotheridomys; Fahlbusch, 1970; Alvarez, 1987). Besides, large numbers of glirids are present (Peridyromys murinus, Pseudodryomys ibericus, Pseudodryomys simplicidens group and Armantomys) as well as the cricetids Eucricetodon aV. aquitanicus, Eucricetodon aV. infralactorensis and Melissiodon cf. dominans (Daams et al., 1987). Among the large mammals, this unit records the Wrst

Miocene mammal units in Western Europe

entry of the felids of the genus Pseudaelurus (P. transitorius = P. turnauensis from Loranca; Alcala´ et al., 1990). Other elements entering this interval are the anthracotherids of the genus Brachyodus, the suids of the genus Xenohyus and the Wrst giraYds of the genus Teruelia (Moya`-Sola` & Agustı´, 1990). Teruelia adroveri Moya`-Sola` from the locality of Navarrete del Rio is a primitive giraYd of the subfamily ProgiraYnae directly related to ProgiraVa from the Bugti beds in the Siwaliks (Pakistan). On the basis of the common presence of these primitive giraYds, Moya`-Sola` & Agustı´ (1990) proposed the correlation of the Bugti beds with the lower part of the Ramblian (zone Z = MN 2b). The presence of anthracotherids in these beds would also reinforce this correlation (but see Pickford, 1987). Therefore, this unit records a Wrst early Miocene dispersal of Asian elements into Western Europe (Teruelia, Brachyodus) and, probably also in Africa (Wrst true giraYds; Moya`-Sola` & Agustı´, 1990). Dating of the lower boundary of MN 2 is hampered by the absence of paleomagnetic analysis in the corresponding sections of the Loranca and Daroca basins. An ash layer placed immediately over the locality of Tardienta (Y2 or X zone, Ebro Basin) provided an age of 19.6 + / − 0.4 Ma for the lower limit of MN 2b ( = X zone) or the upper part of MN 2a ( = Y2 zone; Canudo et al., 1994).

MN 3 This unit is thought to cover the so called ‘Cricetum vacuum’ in Western Europe (Daams & Freudenthal, 1990), that is, the interval between the last localities with Eucricetodon and the Wrst ones with the Wrst ‘Miocene’ cricetids of the genera Democricetodon and Megacricetodon. Only the peculiar genus Melissiodon is usually present as the representative of this family, although some scarce remains of Eucricetodon infralactorensis can still be found in some localities. Among the small mammals, this unit does not record any signiWcant dispersal and the rodent taxocenosis are still dominated by the eomyids of the genus Ligerimys and the glirids, which are represent by a variety of genera. From this point of view, this unit correlates with the Pseudodryomys ibericus zone of the Valle`s-Penede`s Basin (sections of Molı´ Calopa, Costablanca and Sant Andreu de la Barca; Crusafont et al., 1955; Agustı´, 1982) and the A zone of the Calatayud-Daroca Basin (localities of Agreda, Moratilla; Daams & Freudenthal, 1981). However, a noticeable diVerence in the composition of the glirid faunas is observed between these two Iberian basins. Therefore, the Calatayud-Daroca faunas (Agreda, Moratilla) are dominated by glirids with simple dental pattern of the ‘Asymmetrical molar group’ (Meulen & Bruijn, 1982), indicating dry environmental

95

Miocene mammalian successions

96

conditions: Pseudodryomys (simplicidens-type), Armantomys, Praearmantomys, Altomiramys. In contrast, early Miocene, late Ramblian faunas in the Valle`s-Penede`s Basin (Sant Andreu de la Barca) are characterized by ‘complicated’ glirids of the ‘Flat molar group’ (Glirudinus gracilis, Glirudinus glirulus, Bransatoglis sp.) and the absence of the above reported endemic genera Armantomys, Praearmantomys and Altomiramys, which indicates the existence of humid, wooded conditions similar to those of other Central European localities such as Wintershof-West. The latter conclusion is also supported by the abundant crocodile remains present at Sant Andreu de la Barca. Therefore, the beginning of the diVerentiation between a wet, wooded Ibero-Levantine province and a dry Ibero-Central one (Agustı´, 1978, 1990) can be traced back to the late Ramblian (MN 3). Among the large mammals, this unit records one of the Wrst important faunal dispersals covering Africa and Eurasia. A large number of slender herbivores enter during this time in Europe, including equids (Anchitherium), cervids (Procervulus, Lagomeryx, Acteocemas), tragulids (Dorcatherium), palaeomerycids (Palaeomeryx), suids (Aureliachoerus) and proboscideans (Gomphotheridae). The FAD of gomphoterids at this unit is specially signiWcant since it represents the Wrst step of a number of African dispersals into Europe that will punctuate the early–middle Miocene. Curiously, Bruijn et al. (1992) report the Wrst entry of proboscideans in Europe at MN 4, but the association of gomphotherids with MN 3 rodent faunas is a fact both in the Valle`s-Penede`s (Sant Andreu de la Barca, Agustı´ et al., in press) and Lisbon basins (Unit IVb, Quinta do Narigao; Antunes, 1990). Communication with the African continent is also indicated by the presence of Dorcatherium in the earliest Miocene levels of Songhor, a locality which could be correlated with the Early Aragonian (Moya`-Sola` & Agustı´, 1990). Palaeomerycids are known as long ago as Xina and have proven to be a highly diversiWed group, with several species (and even genera), which makes it diYcult to use for long distance correlation. The spread of new herbivore immigrants into Western Europe resulted in the extinction of most of the previous Ramblian artiodactyls, such as Brachyodus, Xenohyus and Teruelia, and only Andegameryx and Cainotherium persisted. Finally, among the carnivores, this unit records the Wrst entry of the ursids of the genus Hemicyon.

The middle Miocene: the Aragonian The middle Miocene is well represented in several Spanish basins, with long sections which have provided a good biostratigraphic control of the mam-

Miocene mammal units in Western Europe

malian succession. This is the case, for instance, of the type-area of the Aragonian stage, the Calatayud-Daroca Basin (Daams et al., 1987), but also of the Valle`s-Penede`s (Agustı´, 1982; Agustı´ et al., 1997), Tagus (Alberdi et al., 1984) and Duero basins (Garcı´a-Moreno, 1988). The several proposed biozonations display many points of coincidence and are mainly based on the evolution of the genus Megacricetodon (Garcı´a-Moreno, 1987; Daams & Freudenthal, 1988; Agustı´ & Moya`-Sola`, 1991). In spite of the problems reported earlier, some segments of this lineage are common to other parts of Europe: Megacricetodon primitivus, M. collongensis, Megacricetodon gersii (see Aguilar, 1980, and Daams & Freudenthal, 1988). In the Iberian peninsula, Agustı´ & Moya`-Sola` (1991) proposed the establishment of a Megacricetodon–Democricetodon superzone subdivided into a number of zones: Megacricetodon primitivus zone, M. collongensis zone, M. gersii zone, M. crusafonti zone and M. ibericus zone. Regional factors provide elements of uncertainty as in the case of the assignment of MN 5 faunas in Southern and Central Europe but magnetobiostratiographic results provide a method of falsiWcation of the previously proposed MN succession and the correlation between the zoogeographic provinces of Europe.

MN 4 In its original meaning (Mein, 1975), the MN 4 was divided into two subunits, MN 4a and MN 4b. In Western Europe, this interval was characterized by the entry of the so-called ‘miocene cricetids’ of the genera Megacricetodon, Democricetodon, Eumyarion, Fahlbuschia, Cricetodon and Lartetomys (the last two entering in MN 4b), after the ‘Cricetum vacuum’. This event was seen as a global one on the basis of the association found in the French karstic locality of Vieux-Collonges. However, the detailed work undertaken in the Calatayud-Daroca Basin (Daams et al., 1987) showed that the dispersal of these elements was not synchronous but displayed a step-wise pattern. Therefore, Democricetodon appeared before Megacricetodon, the latter before Fahlbuschia and the latter before Cricetodon. In the original scheme by Mein (1975), there was no room for the Megacricetodon primitivus zone recognized between the ‘Cricetum vacuum’ and the M. collongensis zone, since this Megacricetodon species is up to now absent from the French record. However, starting MN 4 with M. collongensis would leave the basal part of the Aragonian, that is, the time when one of the most signiWcant faunal dispersals of the European Neogene takes place, without reference in the MN succession. Moreover, a Megacricetodon primitivus zone has been also recognized in Eastern Europe (Aliveri; Klein Hofmeijer & Bruijn, 1988). We therefore propose to re-deWne the MN 4 in the original meaning of the

97

Miocene mammalian successions

98

MN 4a subunit, including the C zone of the Calatayud-Daroca Basin (Daams et al., 1987) and the Megacricetodon primitivus zone of the Valle`s-Penede`s Basin (Agustı´, 1982). Besides Calatayud-Daroca and Valle`s-Penede`s, in Spain this unit is present also in the Tagus Basin (Co´rcoles) and Bun ˜ol. DeWned as in the above mentioned way, the MN 4 is characterized by the FAD of a signiWcant number of small and large mammals which will become the mostly common elements of the middle Miocene European faunas: cricetids (Megacricetodon, Eumyarion), proboscideans (Deinotherium), perissodactyls (Chalicotheirum), suids (Bunolistriodon, Eurolistriodon) and the Wrst bovids in the continent (Eotragus). The remainder artiodactyl taxocenosis is basically composed of the same elements as the former mammal unit (cervids included), except for the disappearance of Acteocemas and Andegameryx. Some localities still belonging to this unit record the Wrst entry of the rhinoceroses of the genus Hispanotherium. These open-country adapted Elasmotherinae rhinoceroses are of probable Western Asian origin and their dispersal in the Iberian peninsula is probably related to a change in the environmental conditions (Antunes, 1979). Another element of probable Western Asian origin is Eotragus, which has also been reported from the Kamlial beds in Turkey. This genus probably colonized simultaneously Western Europe and Africa during the early Miocene, since its presence is reported also at Maboko and Ombo, although the age of these African localities is probably younger than the MN 4, as indicated by the presence of Listriodon. Therefore, the dispersal events at this time are of mainly Western Asian origin, and only Deinotherium can be quoted as an African immigrant during this time.

MN 5 DeWnition of MN 5 with respect of MN 4 has been another conXicting point in Mein’s original (1975) proposal, since some localities included in this unit (Pont Levoy) were in fact very close to the locality of Vieux-Collonges, placed in the unit MN 4b. Therefore, MN 4b and MN 5 remained as virtually undistinguishable units. In Central Europe, a distinction between MN 4 and MN 5 can be easily established on the basis of the FAD of the cricetid Cricetodon. The Wrst members of this genus (Cricetodon versteegi) are found in Anatolian levels assigned to MN 1 (Bruijn & Unay, 1996). In Western Europe, they appear much later, associated with Megacricetodon collongensis (for instance, in the above mentioned localities of Pont-Levoy and Vieux-Collonges). However, in the Central basins of Spain (Calatayud-Daroca, Duero), Cricetodon is not present in most of the localities with Megacricetodon collongensis (zone D of Daams et al., 1987). Therefore, the base of MN 5 in

Miocene mammal units in Western Europe

Spain was placed at the base of zone E of the Calatayud-Daroca Basin (Daams et al., 1987). This decision led to a signiWcant disagreement between the age and duration of this unit in Spain and Central Europe. Thus, in the Calatayud-Daroca basin this limit was placed at the base of chron C5ACn, with an upper limit at the F zone, at the top of the same chron. In this way, MN 5 was very short within chron C5ACn (in contrast with a very long MN 4). In a opposite way, Steininger (in Steininger et al., 1996) placed the lower boundary of MN 5 at 16.5 Ma. However, in spite of the absence of Cricetodon, the rodent taxocenosis of the Megacricetodon collongensis zone in Spain are very diVerent from the previous MN 4 faunas such as Els Casots, indicating a shift towards dryer conditions. This shift can be recognized in the rodent succession of the Calatayud-Daroca and Valle`s-Penede`s basins and has been also conWrmed by the paleoXoral and pollen analysis in the Valle`sPenede`s Basin (Sanz de Siria, 1985, 1988, 1994; Bessedik, 1985). This environmental change can be extended to other parts of Western Europe and covers most of the Megacricetodon collongensis zone. Therefore, in spite of the regional absence of Cricetodon, we propose to extend the lower boundary of MN 5 to cover also the Megacricetodon collongensis faunas from Spain (this implies that there is a delay in the entry of the Wrst Cricetodon species in the Iberian peninsula). In this way, MN 5 would cover zones D and E of the Calatayud-Daroca Basin. The lower limit of this unit would be established at 16 Ma (top of the chron C5Cn; Krijgsman et al., 1994). This unit is widely present in the Calatayud-Daroca (several localities including some with large mammals such as Valdemoros IA), Tagus (Torrijos), Madrid (O’Donell, Moratines, Puente Vallecas), Duero and Valle`sPenede`s (Vilobı´) basins. Among the large mammals, the most signiWcant events are the Wrst appearance of the boselaphine bovid Miotragocerus (Tarazona), the suid Conohyus and the moschid Micromeryx. In the Iberian peninsula, this zone is characterized by the presence of the elasmotherine rhinocerotid Hispanotherium. The cervids of the genus Lagomeyx are no longer present at this level, replaced by Heteroprox. In Africa, the Libyan locality of Gebel Zelten also records the Wrst entry of Miotragocerus (Hamilton, 1973). Since Gentrytragus (= ‘Caprotragoides’) is still absent from this Northern Africa locality, Moya`-Sola´ & Agustı´ (1990) proposed its correlation with the MN 5 levels of the Iberian peninsula. The Russian locality of Byelometcheskaya also presents Boselaphini, lacking Thetytragus and Lisatriodon, which also suggests a correlation with the MN 5 levels in the Iberian peninsula. This site records the Wrst entry of Asiatic (Hypsodontinae) and African elements (Kubanochoerus and the Wrst modern giraYds). Another African element entering Europe during the MN 5 is Pliopithecus.

99

Miocene mammalian successions

100

MN 6 As in many cases, the MN 6 unit was largely based on the content of its reference-locality Sansan (France). Originally, the most characteristic rodent element at Sansan was said to be the cricetid Megacricetodon crusafonti, described from the Spanish locality of Manchones (Baudelot, 1972). However, Aguilar (1980) showed that the Megacricetodon population from Sansan belonged in fact to a diVerent species, M. gersi, supposed to be a vicariant reply of the Spanish species. Nevertheless, M. gersi or a close population was later recognized in Spain in the Duero (Garcı´a-Moreno, 1987) and Calatayud-Daroca basins (Daams & Freudenthal, 1988), in levels below those with Megacricetodon crusafonti. Therefore, as indicated by the large mammals, Sansan and Manchones represent two separated faunal units. The Wrst faunal unit (MN 6a) would include the faunas of the Megacricetodon gersi zone (Duero Basin; Garcı´a-Moreno, 1988), as well as those of the F zone in the Calatayud-Daroca Basin (Daams et al., 1987). Apart from the stratigraphic range of this species, the base of this zone records the Wrst entry of a second Megacricetodon lineage. Beside the localities included in this zone in the Duero and Calatayud-Daroca basins, also the locality of Paracuellos 5 in the Madrid Basin, can be referred to as MN 6a. Although probably covering a small time span, this subunit records some signiWcant events among the large mammals, such as the replacement of Bunolistriodon by Listriodon and the FAD of the Wrst small hyaenids of the genus Protictiherium. Besides Bunolistriodon, it also records the disappearance of typical early–middle Miocene elements such as the rhinoceros Hispanotherium, the palaeomerycids (Triceromeryx) as well as the artiodactyls Cainotherium and Amphitragulus. Listriodon was probably an Asian immigrant which settled in Europe and Africa as a consequence of the same dispersal event. Assumption of the synchroneity of this event would imply a correlation of the MN 6 localities in Europe with the Maboko beds in Africa (also with Listriodon), as proposed by Moya`-Sola` & Agustı´ (1990). A second MN 6b subunit can be widely recognized in a number of basins of the Iberian peninsula: Calatayud-Daroca (Manchones, Arroyo de Val), Madrid (Paracuellos 3) and Ebro basins (La Ciesma, El Buste). All of them belong to the G 1 or G 2 zones of the Calatayud-Daroca basin or the Megacricetodon crusafonti zone of the Duero basin. The faunal associations from these localities are diVerent from those of the M. gersi zone such as Sansan or Paracuellos not only on the basis of the small mammals, but also on the basis of the large faunal assemblages, which are characterized by the entry of a signiWcant number of artiodactyl taxa: Tethytragus, Hispanomeryx and

Miocene mammal units in Western Europe

Euprox. Among these elements, Tethytragus displays the wider geographic distribution (Europe and Asia, plus Africa if we consider the related genus Gentrytragus). This genus seems to be of Asian origin and it is found in the late Aragonian sites of C ¸ andir and Pac¸alar in Turkey. In Africa, a similar form is known from the levels of Nyakak (Thomas, 1984). If the dispersal of this genus was a synchronous event, Nyakak could be probably correlated with the early late Aragonian localities from Spain. The history and distribution of Hispanomerys parallels that of Tethytragus, since this moschid is again present at C ¸ andir and Pac¸alar (although it never settled in Africa). According to Krijgsman et al. (1996), the lower boundary of this subunit (base of G 1 zone) can be established at chron C5ABn, c. 13.6 Ma.

MN 7 Originally, Mein (1975) created the MN 7 unit for some localities bearing an intermediate character between the faunas of the Sansan kind and the youngest Wssure inWllings from the La Grive complex. However, Bruijn et al. (1992) decided to join MN 7 and 8, since they were unable to recognize such a tripartition of the late Aragonian faunas (MN 6, 7 and 8). They were probably inXuenced by the fact that uppermost Aragonian, pre-Vallesian faunas of the La Grive kind are lacking in the type area of the Aragonian stage, the Calatayud-Daroca Basin (with the possible exception of Nombrevilla 2). However, these faunas are well represented in the Valle`s-Penede`s Basin by a number of localities and sections such as Sant Quirze, Can Almirall and Masquefa. These faunas are characterized by the entry of the hominoid Dryopithecus, the suids Propotamochoerus and Parachleuastochoerus, the cervid Stehlinocerus and the bovid Protragocerus. On the other hand, Heteroprox, Eotragus and Tethytragus became extinct. They can be diVerentiated from the typical MN 8 localities in the basin, such as Castell de Barbera ` , Can Missert, Can Mata (Lower Hostalets) and others, since they lack a number of immigrants that preceded the entry of Hipparion, such as the Wrst giraYds (Palaeotragus), the bovid Austroportax and the hyaenids of the genus Thalassictis. Therefore, a partition of the late Aragonian into three units seems in fact feasible at least in Spain, but probably also in other parts of Europe (Engesser, pers. comm.).

MN 8 This zone is not represented in the original Aragonian section, and only the locality of Nombrevilla 2 can be probably referred to in the Calatayud-

101

Miocene mammalian successions

102

Daroca Basin. However, it has been clearly identiWed in the Valle`s-Penede`s Basin (Fahlbuschia crusafonti zone of Agustı´, 1982) and in the Duero Basin (Megacricetodon ibericus zone of Garcı´a-Moreno, 1988). A number of elements (Cricetodon lavocati, Fahlbuschia crusafonti) indicate that these levels are younger than the classical locality of La Grive M (which is also indicated by the fact that in the Duero Basin they have overlain the levels with Cricetodon albanensis, Garcı´a-Moreno, op. cit.). As deWned in the original MN table (Mein, 1975), the MN 8 is characterized by the entry of Hispanomys, Palaeotragus, Protragocerus and Tetralophodon, while Deperetomys hagni and Democricetodon freisingensis are quoted as characteristic species. Neither of the last two mentioned species are present in the Spanish basins. However, from the point of view of the small mammals, this unit can be easily correlated in Western Europe since it records the last occurrence of Cricetodon and the Wrst occurrence of Hispanomys. Localities belonging to this unit in the Valle`s-Penede`s Basin are Can Missert, Castell de Barbera ` and Hostalets de Pierola (Can Mata and other lower Hostalets levels), characterized, as pointed out, by the entry of the Wrst giraYds (Palaeotragus), the bovid Austroportax and the hyaenids of the genus Thalassictis. The upper limit of MN 8 is formally deWned by the FAD of Hippotherium (base of the Vallesian Mammal Stage). These data strongly support a correlation of this unit to the Megacricetodon ibericus zone, recognized in the Valle`s-Penede`s and Duero basins (Agustı´, 1982; Garcı´a-Moreno, 1988). The problem, in this case, is that the entry of Hippotherium occurs within this rodent zone, the presence of this equid being the only available criterion to recognize a lower Megacricetodon ibericus zone (late Aragonian, without Hippotherium) and an upper Megacricetodon ibericus–Hippotherium zone (Early Vallesian). Therefore the MN 8 corresponds only to the lower part of the Megacricetodon ibericus zone.

The beginning of the late Miocene: the Vallesian The Vallesian stage was established by Crusafont (1950) and records the Wrst entry of eastern immigrants like the widely distributed equid Hippotherium (= ‘early Hipparion’). Other taxa accompanying the dispersal of Hippotherium in Western Europe are the felid Machairodus and the Wrst ‘modern’ ursids (Indarctos) (see Agustı´ et al., 1997, for a revision). The most characteristic feature of the Vallesian with respect to later stages is the coexistence of these immigrants with the previous middle Miocene faunal background dominated by cervids, suids, rhinoceroses and large probos-

Miocene mammal units in Western Europe

cideans of the genera Tetralophodon and Deinotherium. This is observed for instance in the case of Machairodus (coexisting with the nimravid Sansanosmilus) or Indarctos (coexisting with the last amphicyonids of the genus Amphicyon).

MN 9 The Hippotherium dispersal is the Wrst and most outstanding event deWning the beginning of the late Neogene in Eurasia. Until recently, the isochrony or diachrony of this dispersal was the subject of an intensive discussion, but recent paleomagnetic calibrations in the Valle`s-Penede`s Basin established an age of 11.1 Ma for the entry of this equid in Europe (Garce` et al., 1997), an age which is consistent with the correlation established in the Paratethys realm (Vienna Basin; Bernor et al., 1988). The earliest occurrence of this equid in Western (Garce`s et al., 1997) and Eastern Europe (Bernor et al., 1988; Woodburne, 1996) indicates that, at least for Europe (but probably also for Central Asia) the dispersal of Hippotherium was a quick bioevent (the discussion around this event has been extensively detailed in Garce`s et al., 1997). As mentioned in the previous paragraph, the entry of Hipparion did not involve any special change or replacement in the previously existing faunas of the late Aragonian: the latest Aragonian and earliest Vallesian faunas in the Valle`s-Penede`s share the same small and large mammal elements, to the point that they are biostratigraphically undistinguishable if Hippotherium is not found (Agustı´ et al., 1997). A second phase within the early Vallesian can be distinguished in the type-area of the Vallesian, the Valle`s-Penede`s Basin. This second subunit corresponds to the Cricetulodon zone of Agustı´ (1982) and Garcı´a-Moreno (1988; Duero Basin). Among the small mammals, there is a signiWcant change, involving the spread and dominance of the Wrst true cricetines (Cricetulodon), which replace the typical persisting middle Miocene association of Megacricetodon (gersi–crusafonti–ibericus lineage) and Fahlbuschia (F. crusafonti). Among the large mammals, there are no signiWcant extinctions, except in the case of the middle sized felids of the genus Pseudaelurus (P. lorteti, P. quadridentatus) and the slender giraYds ascribed to ‘Palaeotragus’. The latter ones are replaced by the sivatherine giraYds of the genus Decennatherium. In general, the localities placed within this zone (Can Ponsic, Santiga, Ballestar, Can Llobateres 1) reXect a much more forested and humid character than the earliest Vallesian localities. Tapirs, tragulids (Dorcatherium) and hominoids (Dryopithecus) are characteristic elements in these associations. Among the rodents, large beavers (Chalicomys) and Xying squirrels are present, while the glirids attain their maximal diversity,

103

Miocene mammalian successions

104

with several genera such as Bransatoglis, Myoglis, Paraglirulus, Glirulus, Eomuscardinus, Muscardinus, etc.

MN 10 A bipartition of the Vallesian was already recognized by Crusafont, who pointed out the impoverished character of the upper Vallesian faunas of the Valle`s-Penede`s such as Viladecavalls. The existence of a signiWcant decay in the rodent diversity (Agustı´, 1978, 1982) and artiodactyl turnover (Moya`-Sola`, 1983) led Agustı´ & Moya`-Sola` (1990) to identify the existence of an important mammal event at the early/late Vallesian boundary called the ‘Late Vallesian Event’ (sometimes also cited as the ‘Mid Vallesian Event’). The MVC involved the local or global disappearance of most of the humid elements characterizing the former biozone, such as the suid Conohyus, the cervid Amphiprax, the moschid Hispanomeryx, the bovids Miotragocerus and Protragocerus, the rhinoceroses Lartetotherium sansaniense and ‘Dicerorhinus’ steinheimensis, the large carnivores of the families Nimravidae and Amphicyonidae, and the hyaenids of the genera Protictitherium and Progenetta. At the same time, a number of eastern immigrants appear for the Wrst time: the large hyaenids of the genus Adcrocuta and Hyaenictis, the suids Schizochoerus and Microstonyx and the bovids of the genus Tragoportax. Among the rodents, the MVC involved the disappearance of several cricetids and glirids of early or middle Miocene origin (Megacricetodon, Eumyarion, Bransatoglis, Myoglis, Paraglirulus, Eomuscardinus), Xying squirrels (Albanensia, Miopetaurista) and beavers (Chalicomys, Euroxenomys). In Western Europe, the disappearance of these elements coincided with the Wrst spread of the murids (Progonomys sp. at Can Llobateres 2). However, other less diversiWed small mammal groups, such as lagomorphs, and insectivores remained almost unaVected by the MVC. As in the case of the early Vallesian, a bipartition of this unit can be recognized in the Valle`s-Penede`s (but also in the Rhoˆne Basin; see Mein, this volume). Therefore, the lower part of the late Vallesian (MN 10a; Cricetulodon-Progonomys zone; localities of Can Llobateres 2, Viladecavalls, etc.) is characterized by the persistence of some early Vallesian elements, such as Listriodon and Parachleuastochoerus among the suids, Dryopithecus among the hominoids and Cricetulodon among the cricetids. A second subunit within the late Vallesian (Rotundomys zone, MN 10b) is characterized by the end of the previously quoted suid and hominoid genera and the development of a ‘pseudomicrotoid’ trend in the species of the genus Cricetulodon (which led to the genus Rotundomys; localities of the Terrassa

Miocene mammal units in Western Europe

area: Torrent de Febulines, TNA, TSA). Reappearance of beavers (Schreuderia) and Xying squirrels (Pliopetaurista) indicates more humid conditions in this biozone than in the previous Cricetulodon–Progonomys zone.

The latest Miocene: the Turolian The Turolian mammal stage was established by Crusafont (1965) in the Teruel Basin, where rich late Miocene (‘Pontian’) mammalian assemblages were known from the last century. Originally, the Turolian included the typical associations found in classical localities such as Concud or Los Mansuetos. However, the Wrst rigorous biostratigraphic works developed in the area by Weerd (1976), favoured a partition of the Turolian into three rodent zones: Parapodemus lugdunensis zone, Parapodemus barbarae zone and Stephanomys ramblensis zone. This tripartition of the Turolian was strictly followed by Mein’s (1975) proposal, distinguishing MN 11, MN 12 and MN 13 units. While the middle and late Turolian can be easily distinguished on the basis of the small and the large mammal fauna (coinciding with the set of faunal events associated with the Messinian crisis), the diVerentiation between the MN 11 and MN 12 in terms of large mammals was more subtle. However, the work developed in recent years (Alcala´, 1994) has conWrmed this partition in Western Europe. It remains a matter of discussion whether this scheme can be exported to Eastern Europe (see earlier discussion).

MN 11 The lower boundary of this unit can be established on the basis of the FAD of Parapodemus. The mostly primitive species of this genus, Parapodemus lugdunensis, appear widespread in Western Europe as an immigrant. A second characteristic element of this unit, Huerzelerimys vireti, is probably a descendant in situ of a previously existing murid lineage (Progonomys woelferi). In Spain, a third characteristic element is Occitanomys sondaari, evolved from the late Vallesian species Occitanomys hispanicus. In Western Europe (Valle`s-Penede`s and Rhoˆne Valley), this unit also records the extinction of the hypsodont cricetids Rotundomys and Anomalomys. Only the genera Hispanomys and Neocricetodon (= Kowalskia) remain from the once highly diversiWed Miocene cricetid taxocenosis. Among the large mammals, this level records the Wrst entry of the giraYds of the genus Birgerbohlinia and the Wrst true Cervinae (Lucentia). An anthracotherid is also

105

Miocene mammalian successions

106

present in the locality of Crevillente 2. In contrast, a number of artiodactyl taxa disappear: Schizochoerus, Austroportax and the small cervid Euprox. Besides Crevillente-2 (Crevillente Basin), the localities of Piera (Valle`sPenede`s) and Vivero de Pinos and La Cantera (Teruel Basin) belong to this unit.

MN 12 The classical localities of Concud and Los Mansuetos, in the Teruel Basin, and Casa del Acero, in the Fortuna Basin, are included in this unit. Hispanodorcas, Palaeoryx, Gazella and Turiacemas appear at this level in Spain, while Dorcatherium, Micromeryx and Lucentia disappear. Among the rodents, there are no signiWcant changes in diversity, with a generic representation almost identical to the previous unit, the only diVerence being based on the local evolution of the genera Parapodemus, Occitanomys, Huerzelerimys, Hispanomys and Neocricetodon.

MN 13 This unit is represented in several sections from Central and Southern Spain: Teruel (Weerd, 1976; Adrover et al., 1993), La Alberca (Mein et al., 1973), Librilla (de Bruijn et al., 1975; Alberdi et al., 1981), Venta del Moro (Morales, 1984), Fortuna (Agustı´ et al., 1985) and other places. Among the small mammals, it records a very important turnover involving the exit of some of the previous elements (Parapodemus, Huerzelerimys) and the entry of several new lineages which will persist well into the early–middle Pliocene: Apodemus, Stephanomys, Castillomys, ‘Cricetus’. Although most of these elements can be related to previously existing genera (Apodemus to Parapodemus, Stephanomys and Castillomys to Occitanomys, Cricetus to Neocricetodon), a direct in situ relationship cannot be established, most of them appearing as new immigrants in the basins. This unit is also characterized by the entry of several elements of African or Asian character, such as the gerbils Protatera, Pseudomeriones, Epimeriones, Myocricetodon and Calomyscus, the murids Paraethomys and Parasaidomys (Aguilar et al., 1983; Agustı´, 1989; Agustı´ & Llenas, 1996). The dispersal of these elements is probably related to the set of events which appear associated with the Messinian crisis in the Western Mediterranean. Among the large mammals, entry of African elements is also conWrmed by the presence of the Wrst Macaca, Hippopotamus, Paracamelus and the Reduncini, which appear for the Wrst time in Europe (Morales, 1984; Moya`-Sola` et al., 1984; Agustı´ & Moya`-Sola`, 1990). The Wrst Pliocervus and the Wrst graviportal bovids of the

Miocene mammal units in Western Europe

genus Parabos also appear at this time. On the contrary, Microstonyx and Turiacemas are no longer present (Alcala´, 1994).

Concluding remarks Assignment of each MN unit to the corresponding biozones of the Mammal Stage System deWned for Western Europe is the only way to provide stable boundaries and make them correlatable with other non-mammalian biostratigraphic scales (including the oceanic record). Otherwise, as stated, the MN system would remain as a selWsh way of classifying mammalian localities in assumed chronological units which could, in fact, be diachronous (or, as in the previously mentioned MN 10/MN 11 case, could be synchronous but belonging to diVerent MN units). Of course, the description of the MN units provided in this paper has rather limited value outside Western Europe, but it is a matter of discussion as to whether the extension of the MN system over wider areas has a real paleontological meaning, even in terms of purely mammalian events.

Acknowledgements This work was supported by the project DGICYT-PB94-1265 of the Spanish Ministry of Education and Science.

References Adrover, R. 1978. Les Rongeurs et lagomorphes (Mammalia) du Mioce`ne infe´rieur continental de Navarrete del Riı´o (Prov. de Teruel, Espagne). Doc. Lab. Ge´ol. Fac. Sci. Lyon, 72, 3–47. Adrover, R., Mei , P. & Moissenet, E. 1993. Roedores de la transicio´n Mio-Plioceno de la regio´n de Teruel. Paleont. i Evol., 26–27, 47–84. Sabadell. Aguilar, J. P. 1980. Nouvelle interpretation de l’e´volution du genre Megacricetodon au cours du Mioce`ne. Palaeovertebrata, Montpellier, Me´m. Jub. R. Lavocat, pp. 355–64. Aguilar, J. P. 1981. Evolution des Rongeurs Mioce`nes et Pale´ogeographie de la Me´diterrane´e Occidentale. The´se Univ. Sciences et Te´ch. du Languedoc, Montpellier, pp. 1–203, 7 pl. Aguilar, J. P., Brandy, L. D. & Thaler, L. 1983. Les Rongeurs de Salobren ˜ a (Sud de l’Espagne) et le proble`me de la migration messinienne. Paleobiologie Continentale, Montpellier, 14, 2, 3–17. Agustı´, J. 1978. El Vallesiense inferior de la Penı´nsula Ibe´rica y su fauna de roedores. Acta Geol. Hisp., Barcelona, 13, 4, 137–41.

107

Miocene mammalian successions

108

Agustı´, J. 1982. Biozonacio´n del Neo´geno continental de CatalunN023a mediante roedores (Mammalia). Acta Geol. Hisp., 17 (1–2), 221–6. Barcelona. Agustı´, J. 1989. On the peculiar distribution of some muroid taxa in the Western Mediterranean. Bull. Soc. Pal. It., 28 (2–3), 147–54. Agustı´, J. 1990. The Miocene Rodent Succession in Eastern Spain: a zoogeographical appraisal. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 375–404. Plenum Press, New York. Austı´, J., Barbera`, X., Cabrera, L., Pare´s, J. M. & Llenas, M. 1994. Magnetobiostratigraphy of the Oligocene-Miocene transition in the Ebro basin (Eastern Spain): state of the art. Mu¨nchner Geowissensch. Abhand, (A), 26, 161–72. Mu¨nchen. Agustı´, J., Cabrera, L., Garce´s, M. & Pare´s, J. M. 1997. The Vallesian mammal succession in the Valle`s-Penede`s basin (northeast Spain): Paleomagnetic calibration and correlation with global events. Paleogeogr. Paleoclim. Paleoecol., 133 (3–4). Agustı´, J. & Llenas, M. 1996. The late Turolian muroid rodent succession in eastern Spain. Acta Zool. Cracov., 39 (1), 47–56. Kra` kow. Agustı´, J. & Moya`-Sola`, S. 1990. Mammal extinctions in the Vallesian (Upper Miocene). Lecture Notes in Earth Science, 30, 425–32. Berlin-Heidelberg. Agustı´, J. & Moya`-Sola`, S. 1991. Spanish Neogene Mammal succession and its bearing on the continental biochronology. Newslett. Stratigr., 25 (2), 91–114. Stuttgart. Agustı´, J., Moya`-Sola`, S., Gibert, J., Guille´n, J. & Labrador, M. 1985. Nuevos datos sobre la bioestratigrafı´a del Neo´geno continental de Murcia. Paleontologia i Evolucio´, 18. Sabadell. Agustı´, J., Moya`-Sola`, S. & Pons-Moya`, J. 1984. Mammal distribution dynamics in the eastern margin of the Iberian Peninsula during the Miocene. Pale´obiologie Continentale, 14, 33–46. Montpellier. Alberdi, M. T., Hoyos, M., Junco, J., Lo´pez, N., Morales, J., Sese´, C. & Soria, D. 1984. Biostratigraphy and sedimentary evolution of continental Neogene in the Madrid area. Pale´obiologie Continentale, 14, 47–68. Montpellier. Alberdi, M. T., Morales, J., Moya`, S. & Sanchiz, B. 1981. Macrovertebrados (Repitlia y Mammalia) del yacimiento Wnimioceno de Librilla (Murcia). Est. Geol., 37, 307–12. Alcala´, L. 1994. Macromamiferos neo´genos de la fosa Alfambra-Teruel. Inst. Est. Turolenses, pp. 554. Teruel. Alcala´, L., Morales, J. & Soria, D. 1990. El registro fo´sil neo´geno de carnivoros (Creodonta y Carnivora, Mammalia) de Espan˜a. Paleont. i Evol., 23, 55–74. Sabadell. Alvarez-Sierra, M. A. 1987. Estudio sistema´tico y bioestratigra´Wco de los Eomyidae (Rodentia) del Oligoceno superior y Mioceno inferior espan˜ ol. Scripta Geol., 86, 1–207. Leiden. Antunes, M. T. 1979. ‘Hispanotherium fauna’ in the Iberian middle Miocene, its importance and paleogeographical meaning. Ann. Geol. Pays Hellen., Athens, H.s., 1979, 1, 19–26. Antunes, M. T. 1990. The proboscideans data, age and paleogeography: evidence from the Miocene of Lisbon. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 253–62. Plenum Press, New York.

Miocene mammal units in Western Europe

Bachmayer, F. & Wilson, R. 1970. Die Fauna der Alt-Plioza¨nen Ho¨hlen- und Spaltenfu ¨ llungen bei KohWdich, Burgerland (Osterreich). Ann. Natur. Mus. Wien., 74. Wien. Barbera`, X., Pare´s, J. M., Cabrera, L. & Anado´n, P. 1994. High resolution magnetic stratigraphy across the Oligocene-Miocene boundary in Alluvial-Lacustrine succession (Ebro Basin, NE Spain). Mu ¨ nchner Geowiss. Abh. (A), 26, 161–72. Baudelot, S. 1972. etude des Chiropte`res, Insectivores et Rongeurs du Mioce´ne de Sansan. The`se Universite´ Paul Sabatier, Toulouse, pp. 1–364. Baudelot, S. & Bonis, L. de 1966. Nouveaux gliride´s (Rodentia) de l’Aquitanien du bassin d’Aquitaine. C. R. Somm. Se´anc. Soc. Ge´ol. France, 9, 341–2. Bernor, R. L., Kovar-Eder, J., Lipscomb, D., Ro¨gl, F., Sen, S. & Tobien, H. 1988. Systematic, stratigraphic, and paleoenvironmental contexts of Wrst-appearing Hipparion in the Vienna basin, Austria. Journal of Vertebrate Paleontology, 8, 427–52. Bernor, R. L., Solounias, N., Swisher III, C. C. & Van Couvering, J. A. 1996. The Correlation of three classical ‘Pikermian’ Mammal faunas – Maragheh, Samos and Pikermi – with the European MN Unit System. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittman, H.-W. (eds.). Columbia University Press, New York. Bessedik, M. 1985. Reconstitution des Environments Mioce`nes des Regions Nord-Ouest Medite´rrane´ennes a Partir de la Palynologie. Ph. Dr. Thesis Univ. Sc. et Tech. Languedoc, Montpellier, pp. 1–162. Bruijn, H. de 1988. Smaller mammals from the upper Miocene and lower Pliocene of the Strimon basin. Int. Workshop ‘Continental Faunas at the Mio/Pliocene Boundary’. Abstracts, Faenza, 10. Bruijn, H. de, Daams, R., Daxner-Hock, G., Fahlbusch, V., Ginsburg, L., Mein, P. & Morales, J. 1992. Report of the RCMNS working group on fossil mammals, Reisensburg 1990. Newslt. Strat., 26 (2–3), 65–118. Berlin-Stuttgart. Bruijn, H. de, Mein, P., Montenat, C. & Weerd, A. van der. 1975. Correlations entre les gisements de rongeurs et les formations marines du Miocene terminal d’Espagne me´ridionale. Kon. Ned. Akad. v. Wetensch., Ser. B, 78 (4), 1–32. Bruijn, H. & Unay, E. 1996. On the evolutionary history of the Cricetodontini from Europe and Asia Minor and its bearing on the reconstruction of migrations and the continental biotope during the Neogene. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittman, H.-W. (eds.), pp. 227–34. Columbia University Press, New York. Canudo, J. L., Cuenca, G., Odin, G. S., Lago, M., Arranz, E. & Cosca, M. 1994. Primeros datos radiome´tricos de la base del Rambliense (Mioceno inferior) en la cuenca del Ebro. Coms. II Congr. Grup. Esp. Terciario, 73–6. Jaca. Crusafont, M. 1950. La cuestio´n del llamado Meo´tico espan ˜ ol. Arrahona, 1, 3–9. Sabadell. Crusafont, M. 1965. Observations au travail de M. Freudenthal et P.Y. Sondaar sur les nouveaux gisements a` Hipparion d’Espagne. Kon. Ned. Akad. v. Wetens., Proc. Ser. B, 68, 121–6. Amsterdam. Crusafont, M., Vilalta, J. F. de & Truyols, J. 1955. El Burdigaliense continental de la cuenca del Valle`s-Penede`s. Mem. y Coms. Inst. Geol. Diputacio´ n Barcelona`, 12, 1–2, 272.

109

Miocene mammalian successions

110

Cuenca, G. 1985. Los roedores (Mammalia) del Mioceno inferior de Autol (La Rioja). Ciencias de la Tierra, 21, 1–96. Inst. Est. Riojanos. Daams, R. 1976. Miocene Rodents (Mammalia) from Cetina de Arago´n (prov. Zaragoza) and Bun ˜ ol (prov. Valencia), Spain. Proc. Konin. Ned. Akad. Wetensch., Ser. B, 79, 152–82. Daams, R. & Freudenthal, M. 1981. Aragonian: the Stage concept versus Neogene Mammal Zones. Scripta Geologica, 62, 1–17. Leiden. Daams, R. & Freudenthal, M. 1988. Cricetidae (Rodentia) from the type-Aragonian; the genus Megacricetodon. Scripta Geol., Spec. Issue 1, 39–132. Leiden. Daams, R. & Freudenthal, M. 1990. The Ramblian and the Aragonian: limits, subdivision, geographical and temporal extension. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 51–9. Plenum Press, New York. Daams, R., Freudenthal, M. & Alvarez-Sierra, M. 1987. Ramblian, a new stage for continental deposits of early Miocene age. Geologie en Mijnbouw, 65, 297–308. Dordrecht. Engesser, B. 1985. Die Gattung Eucricetodon (Mammalia, Rodentia) im Grenzbereich Oligoza¨n/Mioza¨n. Eclog. Geol. Helv., 78 (3), 669–92. Engesser, B. & Mo¨dden, C. 1997. A new version of the biozonation of the Lower Freshwater Molasse (Oligocene and Agenian) of Switzerland and Savoy on the basis of fossil mammals. Me´m. Trav. E.P.H.E. Inst. Montpellier, 21, 475–99. Fahlbusch, V. 1970. Populationsverschiebungen bei tertiaren Negtieren, eine Studien an oligozanen und miozanen Eomiidae Europas. Bay. Akad. Wissench. Mat.-Naturw. k., Abh. N.F., 145, 1–136. Mu ¨ nchen. Franzen, J. L. & Storch, G. 1975. Die unterplioza¨ne (Turolische) Wirbeltierfauna von Do ¨ rn-Durkheim, Rheinessen (SW-Deutschland). 1: Carnivora. Proboscidea. Rodentia. Senckenbergiana lethaea, 56 (4–5), 233–303. Garce´s, M., Cabrera, L., Agustı´, J. & Pare´s, J. M. 1997. Old World Wrst appearance datum of ‘Hipparion’ horses: late Miocene large mammal dispersal and global events. Geology, 25 (1), 19–22, 2 Wgs. Garcı´a-Moreno, E. G. 1987. El ge´nero Megacricetodon (Cricetidae, Rodentia) en el Aragoniense y Vallesiense de la cuenca del Duero. Relaciones Wlogene´ticas. Colpa, 41, 51–105. Madrid. Garcı´a-Moreno, E. G. 1988. The Miocene rodent biostratigraphy of the Duero Basin (Spain); a proposition for a new Aragonian/Vallesian limit. Paleontologia i Evolucio, 22, 103–12. Sabadell. Hamilton, W. R. 1973. The Lower Miocene ruminants of Gebel Zelten, Libya. Bull. Brit. Mus. Nat. Hist. (Geol.), 21, 73–150. London. Klein Hofmeijer, G. & Bruijn, H. de 1988. The mammals from the Lower Miocene of Aliveri (Island of Evia, Greece). Part 8: The Cricetidae. Proc. Konin. Ned. Akad. Wetensch., Ser. B, 91 (2), 185–204. Krijgsman, M., Garce´s, M., Langereis, C. G., Daams, R., Dam, J. van, Meulen, A. van der, Agustı´, J. & Cabrera, L. 1996. A new chronology for the middle to late Miocene continental record in Spain. Earth Planet. Sci. Lett., 142, 367–80. Krijgsman, M., Langereis, C. G., Daams, R., Dam, J. van & Meulen, A. van der 1994. Magnetostratigraphic dating of the Middle Miocene climate change in the continental deposits of the Aragonian type area in the Calatayud-Teruel basin (Central Spain). Earth Planet. Sci. Lett., 128, 513–26.

Miocene mammal units in Western Europe

Lacombia, J. I. 1988. Estudio de las faunas de micromamiferos del Oligoceno superior y Mioceno inferior en las cuencas de Loranca, Ebro riojano y Ebro aragone´s. Aspectos paleoecolo´gicos. Tesis Doct. Univ. Complut. Madrid, pp. 369, 31 pl. Made, J. van der 1994. Suoidea from the Lower Miocene of Cetina de Arago´n. Rev. Esp. Paleont., 9 (1), 1–23. Mein, P. 1975. Report on activity RCMNS-Working groups (1971–1975), 78–81. Bratislava. Mein, P., Bizon, G., Bizon, J. J. & Montenat, C. 1973. Le gisement de Mammife`res de La Alberca (Murcia, Espagne me´ridionale). Corre´lations avec les formations marines du Mioce`ne terminal. C. R. Acad. Sc. Paris, ser. D, 276, 3077–80. Mein, P., Moissenet, E. & Adrover, R. 1983. L’extension et l’aˆge des formations continentales plioce`nes du fosse´ de Teruel (Espagne). C. R. Acad. Sc. Paris, (ser. II), 296, 1603–10. Meulen, A. van der & Bruijn, H. de 1982. The mammals from the Lower Miocene of Aliveri (Island of Evia, Greece). Proc. Konin. Ned. Akad. Wetensch., Ser. B, 85 (4), 485–524. Morales, J. 1984. Venta del Moro: su Macrofauna de Mamiferos y BioestratigraWa Continental del Mioceno Terminal Mediterra´neo. Ph. Doct. Thesis. Ed. Univ. Complutense Madrid. Morales, J. & Soria, D. 1984. Los artioda´ctilos del Mioceno inferior de las cuencas centrales de Espan ˜ a. Colpa, 39, 51–9. Madrid. Moya`-Sola`, S. 1983. Los Boselaphini (Bovidae, Mammalia) del Ne´ogeno de la penı´nsula Ibe´rica. Pub. de Geologı´a. Univ. Auto´ noma Barcelona, 18, 1–236. Moya`-Sola`, S. 1987. Los Rumiantes (Cervoidea y Bovoidea, Artiodactyla, Mammalia) del Ageniense (Mioceno inferior) deNavarrete del Rı´o (Teruel, Espan ˜ a). Paleont. i Evol., 21, 247–69. Sabadell. Moya`-Sola`, S. & Agustı´, J. 1990. Bioevents and Mammal Successions in the Spanish Miocene. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 357–74. Plenum Press, New York. Moya`-Sola`, S., Agustı´, J. & Pons-Moya`, S. 1984. The Mio-Pliocene insular faunas from the West Mediterranean. Origin and distribution factors. Pale´obiologie Continentale, 14, 2, 347–57. Montpellier. Pickford, M. 1987. Revision des Suiformes (Artidactyla, Mammalia) des Bugti beds (Pakistan). Ann. de Pale´ontologie, 73, 4, 347–57. Sanz de Siria, A. 1985. La Xora mioce´nica de los alrededores de Sant Sadurnı´ d’Anoia (Barcelona). Paleont. i Evol., 18, 161–72. Sabadell. Sanz de Siria, A. 1988. Los vegetales mioce´nicos de Rubı´ (Barcelona). Paleont. i Evol., 22, 71–6. Sabadell. Sanz de Siria, A. 1994. La evolucio´n de las paleoXoras en las cuencas cenozoicas catalanas. Acta Geol. Hisp., 29, 2–4, 169–89. Barcelona. Sese´, C. 1987. Eucricetodon y Melissiodon (Cricetidae, Rodentia) from the Ramblian and Lower Aragonian of the Calamocha area (Calatayud-Teruel Basin, Spain). Scripta Geologica, 83, 1–17. Leiden. Steininger, F. F., Berggren, W. B., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Miocene and Pliocene) Marine-Continental Chronologic Correlations of European Mammal Units. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R., Fahlbusch, V. & Mittman, W. (eds.), pp. 7–46. Columbia Univ. Press, New York.

111

Miocene mammalian successions

112

Thaler, L. 1966. Les Rongeurs fossiles du Bas-Languedoc dans leurs rapports avec l’histoire des faunes et la stratigraphie d’Europe. Me´m. Mus. Hist. Nat., n.s., C, 17. Paris. Thomas, H. 1984. Les GiraVoidea et les Bovidae Mioce`nes de la formation Nyakach (Rift Nyanza, Kenya). Palaeontographica, 183, 64–89. Unay, E. & Bruijn, H. de 1984. On some Neogene rodent assemblages from both sides of the Dardanelles, Turkey. Newsletter Strat., 13, 119–32. Weerd, A. van der 1976. Rodent faunas of the Mio-Pliocene continental sediments of the Teruel-Alsambra region, Spain. Utrecht Micropal. Bull., Spec. Pub., 2, 1–217. Woodburne, M. O. 1996. Precision and resolution in mammalian chronostratigraphy: Principles, practices, examples. Journal of Vertebrate Paleontology, 16, 531–55.

6 Large mammals from the Vallesian of Spain Jorge Morales, Manuel Nieto, Meike Kholer and Salvador Moya` -Sola`

Introduction The fossil record of Neogene macromammals is especially rich in Spain. Localities bearing mammal remains are easily found in several large continental Tertiary basins with long-detailed time series and highly exposed strata. However, the Vallesian record is only well documented and available in two small, marginal basins. 1. The Valle´s-Penede´s Basin (Catalun ˜a), where this mammal age was deWned (Crusafont, 1950; Aguirre et al., 1975; Aguirre, 1981; Moya`-Sola` & Agustı´, 1990a; Agustı´ & Moya`-Sola`, 1991). It has a rich continuous record ranging from the Upper Aragonian till the Lower Turolian. 2. The Calatayud-Teruel Basin (Arago´n), which covers the same time span but within two areas, Daroca up to the end of the Lower Vallesian, and Teruel with fossil sites from the beginning of the Upper Vallesian onwards (Daams et al., 1988; Alcala´, 1994; Dam, 1998). Other basins – lacking a continuous record – present very important isolated Vallesian localities. That is the case of Los Valles de Fuentiduen ˜a from the Upper Vallesian of the Duero Basin (Alberdi et al., 1981), or Cerro de los Batallones site in the Upper Vallesian of the Tagus Basin where an incredible amount of carnivore remains have already been unearthed (Morales et al., 1992).

Fossil faunas The beginning of the Vallesian is marked by the entrance of the immigrant fossil equid Hipparion (see Sen, 1990, for Hipparion event disscusion) that came from North America to replace a previous immigrant equid Anchitherium. Their coexistence has been only registered once, in the fossil site of Nombrevilla 1 in the Calatayud-Teruel Basin. The Aragonian–Vallesian boundary reveals a strong change in the composition of the macromammal fossil faunas. Indeed, the Lower Vallesian was characterised by completely diVerent communities of carnivores and herbivores from those of the Upper Aragonian. Within the large herbivores, signiWcant new taxa were – along with the already mentioned Hipparion – the Sivatherine giraYds and the boselaphine Bovids. Among the carnivore

Miocene mammalian successions

114

associations, the large hyaenids – Thalassictis, Lycyaena and Adcrocuta – and felids – Machairodus and Paramachairodus – were then dominant. Although some common Aragonian taxa still survived – like the small rhino Alicornops simorrense – most of them became very scarce and eventually disappeared, while some rare forms – like the herbivore Dorcatherium – from the Upper Aragonian increased their numbers. The Aragonian–Vallesian transition should be somehow gradual. This is reXected by the strong similarities between the Hostalets de Pierola inferior and Sant Quirze faunas (without Hipparion) and the fauna of localities as Nombrevilla 1, where Hipparion was already present (Table 6.1). The analysis of the Lower Vallesian faunas show a biogeographic component not fully understood that separates the Valle´s-Penede´s area from the other Basins of Spain. In fact, well documented faunas such as Los Valles de Fuentiduen ˜ a in the Duero Basin are clearly diVerent from those coetaneous ones in the Valle´s-Penede´s. Primates, tapirs, modern ursids (Indarctos and Ursavus) and suids are typical components of the Lower Vallesian of the Valle´s-Penede´s localities that are lacking or almost lacking in the rest of the basins (an exception being Nombrevilla 2, an Upper Aragonian locality from the Calatayud-Teruel Basin, where two suids, Listriodon and Barberahyus, and Ursavus have been recorded). This geographical segregation still persists in the Upper Vallesian where all Valle´s-Penede´s sites show more similarity between them than to any other site even when the age is more similar with the later ones. This pattern can be shown with the following example: the Upper Vallesian site of Terrassa has a mammal association more similar to the one of Villadecavalls (earlier according to Agustı´ & Gibert, 1979) or to the Lower Vallesian sites from the same area than to Upper Vallesian dated sites from the CalatayudTeruel Basin like Masia del Barbo and La Roma 2. Alternatively it can be proposed that this pattern responds to a temporal sequence where the Calatayud-Teruel sites were younger than the Catalonian ones and thus more similar to Lower Turolian faunas. By the Lower Turolian, this geographical segregation is not apparent and the faunas from the Valle´s-Penede´s resemble those of the other basins. For example, Piera in the Valle´s-Penede´s has fairly similar fauna to those of Puente Minero (Calatayud-Teruel) or Crevillente 2 (Levante Basin), all being mainly composed of Hipparion remains and one bovid, Tragoportax. In all basins, the beginning of the Lower Turolian meant a similar change in the faunas, the rhinocerotids and suids became rare and the species present were represented by few individuals. All the above mentioned points are illustrated in Fig. 6.1, as well as some new interesting ones. The Wgure is the result of a cluster analysis with data

Large mammals from the Vallesian of Spain

[Figure 6.1] Cluster analysis results on data (Table 6.1) of large mammals from the Upper Aragonian to the Middle Turolian of Spain.

on presence or absence of primates, proboscideans, carnivores, perissodactyls and artiodactyls species in the most representative localities from the Upper Aragonian (MN 6–8) till the Lower Turolian (MN 11–12). Presence or absence data were employed to calculate similarity measurements between sites using Dice index, which stresses the common elements in its way to calculate the similarity matrix. The sites within the matrix were then grouped using UPGMA mode in the SAHN procedure obtaining the tree shown in the Wgure. From the Wgure it is clear that macromammal associations Wt the commonly used MN mammal zones pointing to the temporal closeness of Terrasa and Viladecavalls faunas. Nombrevilla 1 is the only exception because presenting Hipparion and thus being Vallesian (indeed Lower Vallesian, MN 9) it is grouped with Valle´s-Penede´s faunas from the Upper Aragonian dated as MN 8.

Diversity and palaeoenvironmental changes Vallesian faunas have been classically considered as the result of a faunistic interchange associated with a trend towards subtropical conditions. This idea is supported by the fact that some of the immigrant taxa involved in the Late Aragonian–Early Vallesian interchange came from Africa. Moreover, according to Pickford & Morales (1994), the biogeographic limit between the Ethiopic and Palaeoarctic realms would be placed – by the Early Vallesian – in more higher latitudes than today. From the Upper Aragonian to the Vallesian macromammal diversity increased due to the gradual entrance of taxa. This process reached its

115

1 0 0 0

1 0 0 0

0 0 0 1 0 0 1 0 0 0 0 0 0

Carnivora Canis cipio Agnotherium antiquus Pseudarctos sp. Amphicyon major Amphicyon castellanus Thaumastocyon dirus Plithocyon armagnacensis Ursavus depereti Ursavus primaevus Ursavus brevirhinus Indarctos vireti Indarctos atticus Simocyon simpsoni

0 0 0 0 0 0 1 0 0 0 0 0 0

0

0 0 0 0

0

0 0 0 0

Primates Dryopithecus laietanus Dryopithecus crusafonti Anapithecus Pliopitehcus antiquus

AV

Proboscidea Deinotherium giganteum Gomphotherium angustidens Tetralophodon longirostris Anancus arvernensis Zygolophodon turicensis

P3

Taxa

0 0 0 0 0 0 0 0 0 0 0 0 0

1 0 0 0

0

0 0 0 0

T3

0 0 0 0 0 0 0 0 0 0 0 0 0

1 0 0 0

1

0 0 0 0

OT

0 0 0 0 0 0 0 1 0 0 0 0 0

0 0 0 0

0

0 0 0 0

0 0 0 0 0 0 1 0 0 0 0 0 0

0 1 0 0

1

0 0 0 0

NO2 HI

0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

0

1 0 0 0

STQ

0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

0

0 0 0 0

0 0 0 1 1 1 0 0 0 0 0 0 0

0 1 0 0

1

0 0 0 0

NO1 VF

Table 6.1. Presence/absence matrix of the species of macromammals

0 1 1 1 0 0 0 0 1 1 1 0 1

0 1 0 0

1

1 0 0 0

LL

0 0 0 1 0 0 0 0 1 0 1 0 0

0 1 0 0

1

0 1 0 0

PO

0 0 0 0 1 0 0 0 0 0 0 0 0

0 1 0 0

1

1 0 0 1

BB

0 0 0 0 1 0 0 0 0 0 0 0 1

0 1 0 0

0

0 0 0 0

BT

0 0 0 0 0 0 0 0 0 0 1 0 0

0 1 0 0

1

0 0 0 0

VV

0 0 0 0 0 0 0 0 0 0 0 1 0

0 1 0 0

1

0 0 1 0

TR

0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

0

0 0 0 0

R2

0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

0

0 0 0 0

MB

0 0 0 0 0 0 0 0 0 0 0 1 0

0 1 0 0

0

0 0 0 0

PM

0 0 0 0 0 0 0 0 0 0 0 1 0

0 1 0 0

1

0 0 0 0

CR

0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

1

0 0 0 0

PI

1 0 0 0 0 0 0 0 0 0 0 1 0

0 1 1 1

1

0 0 0 0

CD

1 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0

0

0 0 0 0

LM

Simocyon primigenius Ischyrictis petteri Lymnonyx sinerizi Circamustela dechaseauxi Taxodon sansaniensis Sabadellictis crusafonti Trocharion albanense Trochictis narcisoi Promephitis pristinidens Mesomephitis medius Paralutra sp. Adroverictis schmidtkit. Promeles sp. Plesiomeles cajali Paleomeles pachecoi Marcetia santigae Proputorius sp. Martes sp. Martes munki Martes burdigaliensis Martes andersoni Martes mellibula Martes paleosinensis Martes basilii Baranogale adroveri Sivaonyx lluecai Plesiogulo sp. Semigenetta ripoli Herpestes dissimilis Leptoplesictis aurelianensis

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 1 1 1 1 1 1 1 1 0 0 0 0 1 0 0 1 0 0 1 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0

0 0

Viverra sansanensis Proticththerium crassum Plioviverrops guerini Ictitheriinae indet. Ictitherium pannonicum Thalassictis robusta Thalassictis montadai Thalassictis hipparionum Thalassictis sp.1 Thalassictis sp.2 Hyaenictis almerae Lycyaena chaeretis Adcrocuta eximia Sansanosmilus jourdani Stenailurus teilhardi Metailurus major Metailurus parvulus Pseudaelurus quadrident. Pseudaelurus lorteti Machairodus alberdiae Machairodus aphanistus Machairodus giganteus Paramachairodus orientalis Paramachairodus ogygia Felis antediluviana

AV

P3

Taxa

Table 6.1 (cont.).

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

T3

0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

OT

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1 0 0 0 0

NO2 HI

0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

STQ

0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 1 0 1 1 0 0

NO1 VF

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0

LL

0 0

0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0

PO

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0

BB

1 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0

BT

0 1

0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 1 0 0

VV

0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1

TR

0 0

0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0

R2

0 0

0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0

MB

0 0

0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 1

PM

1 0

0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0

CR

0 0

0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0

PI

0 0

0 0 1 0 1 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1 1

CD

0 0

0 0 1 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 0

LM

Artiodactyla Albanohyus pygmaeus Barberahyus castellensis Listriodon splendens Parachleuastochoerus crus.

0 0 1 0

1 0 1 0

0

0 0 1 0 0 0 0 0 0

0

0

1 0 0

1 0 0

Hyracoidea Pliohyrax graecus

0 0

0 0

0 0 0 0 0 0

0 0 0 1

1 0 0 1

Perissodactyla Chalicotherium grande Ancylotherium pentelici Tapirus priscus Lartetotherium sansaniense Stephanorhinus schleierm. Dicerorhinu steinheimensis Alicornops simorrense Alicornops alfambrense Aceratherium tetradactylum Aceratherium incisivum Brachypotherium sp. Anchitherium sp. Anchitherium sampelayoi Hipparion primigenium Hipparion catalaunicum Hipparion mediterraneum Hipparion concudense Hipparion gromovae

0 1 0

0

0 1 1 0 0 0 0 0 0

1 0 0

0 0

0 0 0 0

0 0 1 0

0

0 0 1 0 0 0 0 0 0

1 0 0

0 0

0 0 0 1

0 1 1 0

0

0 0 1 0 0 0 0 0 0

1 0 0

0 0

0 0 0 0

0 0 0 0

0

0 0 1 0 0 0 0 0 0

0 0 0

0 0

0 0 0 0

0 0 0 0

0

0 0 0 0 1 0 0 0 0

0 0 0

0 0

0 0 0 0

1 0 0 0

0

0 0 0 1 1 0 0 0 0

1 0 0

0 0

0 0 0 1

0 0 0 0

0

1 0 0 0 1 0 0 0 0

1 0 0

0 0

0 1 0 0

0 0 1 1

1

1 0 0 0 0 1 0 0 0

1 0 0

1 0

1 0 1 1

0 0 1 0

0

0 0 0 0 0 1 0 0 0

1 0 0

0

1 0 1 1

1 1 1 0

0

0 0 0 0 0 0 0 0 0

1 0 1

0

1 0 0 0

0 0 0 0

0

0 0 0 0 0 0 0 0 0

0 0 0

0

0 0 0 0

0 0 0 1

0

1 0 0 0 0 1 0 0 0

0 0 0

1

1 0 0 0

0 0 0 0

0

1 0 0 0 0 1 1 0 0

1 0 0

1

1 0 0 0

0 0 0 0

0

1 0 0 0 1 0 0 0 0

0 1 0

1

0 0 0 0

0 0 0 0

0

1 0 0 0 1 0 0 0 0

0 0 0

1

0 0 0 0

0 0 0 0

0

0 0 0 0 1 0 0 0 1

0 0 0

1

0 0 0 0

0 0 0 0

0

0 0 0 0 0 0 1 0 0

0 0 0

1

0 0 0 0

0 0 0 0

0

1 0 0 0 0 0 1 0 0

0 0 0

1

1 0 0 0

0 0 0 0

0

1 0 0 0 0 0 0 1 0

0 0 0

1

0 0 0 0

0 0 0 0

0

1 0 0 0 1 0 0 1 0

0 0 0

1

0 0 0 0

1 0 0 0

0

0 0 0 0 0 1 0 1

0 0 1 0 0 0

0 0 0 0

0

0 0 0 0 0 0 0 1

0

0 0

1 0 0 0

0 0 1 0 0 0

0

0 0 0 0 0 1 0 1

0

0

0

0

T3

1

0

Parachleuastochoerus stein. Parachleuastochoerus huen. Propotamochoerus palaeoc Microstonyx major Microstonyx erymanthius Schizochoerus vallesiensis Dorcatherium crassum Dorcatherium naui Micromeryx flourensianus Hispanomeryx duriensis Hispanomeryx aragoniensis Stehlinocerus elegantulus Amphiprox anocerus Euprox furcatus Euprox minimus Euprox dicranoceros Palaeoplatyceros hispanicus Heteroprox larteti Lucentia pierensis Lucentia iberica Turiacemas concudensis

AV

P3

Taxa

Table 6.1 (cont.).

0 0 0 0

0 0 0 0 0 1

0 0 0 1 0 0 0 0

0

0

0

OT

0 0 0 0

0 0 1 0 1 0

0 0 0 0 0 0

0 0 0 0

0 0 0 0 0 0 0 0

0

0

0

0 0 0 0 0 0 0 0

0

0

0

NO2 HI

0 0 0 0

0 0 1 0 0 0

0 0 0 0 0 0 0 0

0

0

1

STQ

0 0 0 0

0 1 0 0 1 0

0 0 0 0 0 0 0 0

0

0

0

0 0 0 0

0 0 0 0 1 0

0 0 0 0 1 1 1 0

0

0

0

NO1 VF

0 0 0 0

0 0 0 0 0 0

0 0 0 0 1 1 0 0

1

0

0

LL

0 0 0 0

0 1 0 0 1 0

0 0 0 0 1 1 0 0

1

1

1

PO

0 0 0 0

1 0 1 0 0 0

0 0 0 0 1 1 0 0

1

1

1

BB

0 0 0 0

0 0 0 0 0 0

1 0 0 0 0 0 1 0

0

0

0

BT

0 0 0 0

0 0 1 1 0 0

0 0 1 0 0 1 0 0

0

0

0

VV

0 0 0 0

0 0 0 1 0 0

1 0 0 0 1 1 0 0

0

0

0

TR

0 0 0 0

0 0 0 0 0 0

1 1 0 0 0 1 0 0

0

0

0

R2

0 0 0 0

0 0 0 0 0 0

1 0 0 0 0 0 0 0

0

0

0

MB

0 0 0 0

0 0 0 0 0 0

1 0 0 0 0 1 0 0

0

0

0

PM

0 0 1 0

0 0 0 0 0 0

1 0 0 0 1 0 0 0

0

0

0

CR

0 0 0 0

0 0 0 0 0 0

1 0 0 0 0 0 0 0

0

0

0

PI

0 1 0 1

0 0 0 0 0 0

1 0 0 0 0 0 0 0

1

0

0

CD

0 0 0 1

0 0 0 0 0 0

1 0 0 0 0 0 0 0

0

0

0

LM

1 0 0

0 0 0 0 0 0 1 1 0 0 0 0 0

1 0 0

0 0 0 0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 0 0 0 1 0 0 0 0

1 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0

0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0

0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0

0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0

0 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0

0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0

0 0 0 1 0 0 0 0 1 0 0 0 1 1 1 0

0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0

0 0 0

Localities: P3 = Paracuellos 3; AV = Arroyo del Val; T3 = Toril 3; OT = Cerro del Otero; NO2 = Nombrevilla 2; STQ = San Quirze; HI = Hostalets de Pierola inferior; NO1 = Nombrevilla 1; VF = Los Valles de Fuentiduen˜a; LL = Can Llobateras; PO = Can Ponsic; BB = Castel de Barbera´; VV = Villadecavaus; TR = Terrassa; R2 = La Roma 2; MB = Masia del Barbo; PM = Puente Minero; CR = Crevillente 1; PI = Piera; CD = Concud; LM = Los Mansuetos; BT = Batallones. Taxonomic data from Alberdi & Alcala´ (1990); Azanza et al. (1997); Cerden˜o (1992); Mazo (1977, 1981); Moya`-Sola` & Agustı´ (1990a,b); Nieto et al. (1997); Fraile et al. (1997); Made (1997).

Palaeomeryx magnus Palaeotragus sp. Decennatherium pachecoi Birgerbohlinia schaubi Miotragocerus monacensis Miotragocerus pannoniae Protragocerus chantrei Austroportax latifrons Tragoportax gaudryi Eotragus sansanensis Tethytragus langai Samotragus pilgrimi Palaleoryx pallasi Gazella deperdita Hispanodorcas torrubiae Aragoral mudejar

Miocene mammalian successions

122

maximum during the Lower Vallesian, when the highest mammal diversity of the whole Spanish Neogene was reached. This would mean that the two dispersal events (B and C) identiWed by Moya`-Sola` & Agustı´ (1990b) for the Late Aragonian and the Early/Late Vallesian boundary were part of a unique and vast Late Aragonian–Late Vallesian faunistic turnover. Two diversity patterns can be recognised in the studied faunas (Fig. 6.2). 1. A positive value for the Wrst/last records ratio during the Upper Aragonian and the beginning of the Lower Vallesian lead to a continuous increase of the diversity. 2. A negative value for the above mentioned ratio is recorded from the Lower Vallesian onwards. As a consequence diversity fell, reaching a minimum in the Upper Vallesian of the Calatayud-Teruel Basin, when almost no Wrst records are registered while last ones keep on being numerous. This trend remains during the Early and Middle Turolian although diminished by a greater entrance of immigrants, mainly during the Middle Turolian. Cenograms – Wrst developed by Valverde (1967) and adapted to palaeontological studies by Legendre (1986; 1987) – are another source of information about diversity. They also yield information about the trophic structure of the non-carnivorous mammal community. Best represented communities for the studied time span are analysed by this means. Their structure presents strong diVerences in the various localities, pointing to changes in the environmental conditions (Fig. 6.3). Starting in the Upper Aragonian – represented by Paracuellos 3 fauna – its cenogram presents a steep slope due to low diversity and strong size differences. Few small species are represented in a short line that could be interpreted as the consequence of moderate temperatures. Lines corresponding to middle and large species are almost continuous indicating high humidity (Legendre, 1987) which increased from the arid conditions of the Middle Aragonian (Calvo et al., 1993). Lower Vallesian patterns represented by the cenograms of Los Valles de Fuentiduen ˜ a and Can Llobateras are very diVerent, reXecting an environmental change. They present smoother slopes due to a larger amount of species within a continuous range of sizes. Long large and middle sized species lines reXect a wet, warm, nearly tropical climate. In the Upper Vallesian, Terrassa’s cenogram shows no signiWcative diVerences with the previous ones, but it is clearly diVerent from those of Puente Minero and Masia del Barbo (Alcala´, 1994). The latter have steeper slopes and short lines for the large and small species range separated by an abrupt transition. This pattern indicates open, arid, and less warm environments.

Large mammals from the Vallesian of Spain

[Figure 6.2] Chronostratigraphic distribution of Upper Aragonian to Middle Turolian localities bearing large mammals (Primates, Proboscidea, Carnivora, Perissodactyla, Hyracoidea and Artiodactyla). MN biozonation after Bruijn et al. (1992). Diversity, first and last records for all large mammals excluding Carnivora were based on the associations recognised in the cluster analysis (in Roman numerals) plus the temporal range of the localities (based on their micromammal composition).

Diversity and cenogram analysis point to a climatic optimum for the development of macromammls reached during the Vallesian. By the end of it, an abrupt return to the previous, colder and drier, conditions is indicated by the Calatayud-Teruel faunas and the disappearance of all ape taxa. In clear agreement with these results, analysis on the Teruel-Calatayud micromammals (Dam, 1998) points to a wet period during MN 9–10 that became drier during the boundary between MN 10–MN 11 and in the later zone. In both macro- and micromammals, humidity changes strongly agree, reXecting that these changes are easier to detect than those of temperature when studying mammal faunas in middle latitudes.

Conclusions Spanish Vallesian faunas are fairly continuous, as shown by the gradual changes between the diVerent mammal associations during that time. Faunas without the Hipparion equid resemble others where this species was already present, allowing the hypothesis about a rapid inmigration of this

Miocene mammalian successions

124

[Figure 6.3] Cenograms of the localities of Paracuellos 3 (MN 6, Tajo Basin), Los Valles de Fuentiduen˜a (MN 9, Duero Basin), Can Llobatera (MN 9, Valle´s-Penede´s Basin), Terrassa (MN 10, Valle´s-Penede´s Basin), Masia del Barbo (MN 10, Calatayud-Teruel Basin) and Puente Minero (MN 11, Calatayud-Teruel Basin).

form. Anyway, Vallesian macromammal faunas from Spain were not homogeneous. Marked biogeographic, biostratigraphic and palaeoenvironmental diVerences are manifest. This age is characterised by a strong replacement of taxa with a maximum during the end of the Early Vallesian; afterwards the extinction of these immigrants marks the beginning of the Late Vallesian when all faunas were much poorer. According to all these considerations, four assemblages can be recognised within the Vallesian localities here studied.

Large mammals from the Vallesian of Spain

References Aguirre, E. 1981. El Vallesiense en la meseta castellana. Estudios Geolo´gicos, 37, 339–41. Aguirre, E., Alberdi, M. T. & Pe´ rez-Gonza´lez, A. 1975. Vallesian. In Stratotypes of Mediterranean Neogene Stages, Steininger, F. F. & Neveskaya, L. A. (eds.). Slovak Acad. Sci. Bratislava, 2, 153–7. Agusti, J. & Gibert, J. 1979. Micromamı´feros fo´siles del Mioceno superior de Terrasa (Barcelona, Espan ˜ a). Estudios Geolo´gicos, 35, 493–6. Agustı´, J. & Moya`-Sola`, S. 1991. Spanish Neogene Mammal succession and its bearing on continental biochronology. Newslett. Strat., 25, 91–114. Alberdi, M. T. & Alcala´, L. 1990. El ge´nero Hipparion en la fosa de Alfambra-Teruel. Paleontologı´a i Evolucio´, 23, 105–9. Alberdi, M. T., Lo´pez, N., Morales, J., Sese´, C. & Soria, D. 1981. Bioestratigrafı´a y biogeografı´a de la fauna de mamı´feros de los Valles de Fuentiduen˜a (Segovia). Estudios Geolo´gicos, 37, 503–11. Alcala´, L. 1994. Macromamı´feros Neo´genos de la Fosa de Alfambra-Teruel. Instituto de Estudios Turolenses – Museo Nacional de Ciencias Naturales, 554 pp. Azanza, B., Nieto, M., Soria, D. & Morales, J. 1997. El registro neo´geno de los Cervoidea (Artiodactyla, Mammalia) de Espan˜a. In Avances en el Conocimiento del Terciario Ibe´rico. Calvo, J. P. & Morales, J. (eds.), pp. 41–4. Univ. Complutense-C.S.I.C. Bruijn, H de., Daams, R., Fahbusch, V., Ginsburg, L., Daxner-Ho¨ck, G., Mein, P. & Morales, J. 1992. Report of the RCMNS working group on fossil mammals. Newslett. Stratigr., 26, 65–118. Calvo, J. P., Daams, R., Morales, J., Lo´pez-Martı´nez, N., Agustı´, J., Anado´n, P., Armenteros, I., Cabrera, L., Civis, J., Corrochano, A., Dı´az-Molina, M., Elizaga, E., Hoyos, M., Matı´n-Sua´rez, E., Martı´nez, J., Moissenet, E., Mun ˜ oz, A., Pe´rez-Garcı´a, A., Pe´rez-Gonza´lez, A., Portero, J.M., Robles, F., Santisteban, C., Torres, T., Van der Meulen, A., Vera, J.A. & Mein, P. 1993. Up-to-date Spanish continental Neogene synthesis and paleoclimatic interpretation. Rev. Soc. Geol. Espan ˜ a, 6, 29–40. Cerden ˜ o, E. 1992. Spanish Neogene Rhinoceroses. Palaeontology, 35, 297–308. Crusafont, M. 1950. El sistema mioceno en la depresio´n espan ˜ ola del Valle´s-Penede´s. Proceed. 18 Internat. Geol. Congress, London, 1948, 11, 33–42. Daams, R., Freudenthal, M. & Meulen, A. van der 1988. Ecostratigraphy of micromammal faunal from the Neogene of the Calatayud-Teruel Basin. Scripta Geologica Special Issue, 1, 287–302. Dam, J. A. van 1998. The small mammals from the upper Miocene of the Teruel-Alfambra region (Spain): Paleobiology and Paleoclimatic reconstructions. Geologica Ultraiectina, 156, 204 pp. Fraile, S., Pe´rez, B., Miguel, I. & Morales, J. 1997. Revisio´n de los carnı´voros presentes en los yacimientos del Neo´geno espan ˜ ol In Avances en el Conocimiento del Terciario Ibe´rico, Calvo, J. P. & Morales J. (eds.), pp. 77–80. Univ. Complutense-C.S.I.C. Legendre, S. 1986. Analysis of mammalian communities from the late Eocene and Oligocene of Southern France. Palaeovertebrata, 16, 191–212.

125

Miocene mammalian successions

126

Legendre, S. 1987. Les comunaute´s de mamife`res d’Europe occidentale de l’Eocene supe´rieur et Oligoce`ne: structures et milieux. Mu ¨ nchner Geowiss. Abh., 10, 301–12. Made, J. van. 1997. Los Suoidea de la penı´nsula Ibe´rica. In Avances en el Conocimiento del Terciario Ibe´rico, Calvo, J. P. & Morales, J. (eds.), pp. 109–12. Univ. Complutense-C.S.I.C. Mazo, A. V. 1977. Revisio ´ n de los mastodontes de Espan ˜ a. Tesis doctoral. Univ. Complutense, 440 pp. Mazo, A. V. 1981. Estudio taxono´mico de los mastodontes (proboscidea, Mammalia) de la provincia de Teruel (Espan ˜ a). Teruel, 65, 169–94. Morales, J., Capita´n, J., Calvo, J. P. & Sese´, C. 1992. Nuevo yacimiento de vertebrados del Mioceno superior al Sur de Madrid. Geogaceta, 12, 77–80. Moya`-Sola`, S. & Agustı´, J. 1990a. The Vallesian in the type area (Valle`s-Penede`s, Barcelona, Spain). Ann. Inst. Geol. Publ. Hung., 70, 93–9. Moya`-Sola`, S. & Agustı´, J. 1990b. Biovents and mammals successions in the Spanish Miocene. In European Neogene Mammal Chronology, Lindsay, E. H. et al. (eds.), pp. 357–73. Plenum Press. Nieto, M., Azanza, B., Soria, D. & Morales, J. 1997. El registro fo´sil neo´geno de los Bovoidea (Artiodactyla, Mammalia) de Espan ˜ a. In Avances en el Conocimiento del Terciario Ibe´rico, Calvo, J. P. & Morales, J. (eds.), pp. 137–40. Univ. Complutense-C.S.I.C. Pickford, M. & Morales, J. 1994. Biostratigraphy and palaeobiogeography of East Africa and the Iberian peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 112, 297–322. Sen, S. 1990. Hipparion datum and its chronologic evidence in the Mediterranean area. In European Neogene Mammal Chronology, Lindsay, E. H. et al. (eds.), pp. 495–505. Plenum Press. Valverde, J. A. 1967. Estructura de una Comunidad Mediterra´nea de Vertebrados Terrestres. C.S.I.C., 218 pp.

7 Trends in rodent assemblages from the Aragonian (early–middle Miocene) of the Calatayud-Daroca Basin, Aragon, Spain Remmert Daams, Albert J. van der Meulen, Pablo Pelaez-Campomanes and Maria A. Alvarez-Sierra

Introduction Previous studies on rodent faunas from the Calatayud-Daroca Basin are numerous. An extensive enumeration of these papers is given by Daams et al. (in press). In this paper we will limit ourselves to the ones on paleoecology and paleoclimatology. The latest rodent databases used in paleoecological analysis of this basin are from Daams & Freudenthal (1988) and van der Meulen & Daams (1992). In the 1990s, many more faunas have been sampled and are included in the present study. Faunas with less than 100 rodent M1–2 are excluded from the present analysis since they are considered not to be representative enough. The faunal succession covers the latest Ramblian, Aragonian and the early Vallesian (MN3–MN9; 16.8–10 Ma). Stratigraphic control and dense magnetostratigraphic sampling allowed correlation to the chronostratigraphic and numerical time scales (Krijgsman et al., 1994, 1996). In this paper some corrections are introduced as far as the correlation of the lower part of our succession to the numerical time scale is concerned. Numerical ages are assigned to the individual faunas using the paleomagnetic data in combination with average sedimentation rates of the predominantly lacustrine sediments.

Association structure In order to describe the association structure, and more precisely its variation through time, we calculated the reciprocal of Simpson’s index (Peet, 1974) since it depends only slightly on the number and frequency of rare species. Thus, the eVect of a possible sampling bias is minimized. We decomposed the changing diversity of the mainly Aragonian rodent faunas into two components: equitability (Fig. 7.2) and species richness (Fig. 7.5). The results are discussed in the following paragraphs.

Miocene mammalian successions

128

[Figure 7.1] Diversity curve calculated from the reciprocal Simpson’s index for the latest Ramblian to early Vallesian rodent faunas from the Daroca area.

Diversity The highest diversity of the rodent associations from the Daroca area is found during early MN4, followed by relatively low values during late MN4, as shown in Fig. 7.1. In earliest MN5 times a recovery is shown, followed by a gradual decrease to very low diversity in early MN7/8 around 12.9 Ma. During MN7/8 there is an increase toward more diverse faunas. The early Vallesian faunas are relatively diverse, although they do not reach the high diversity levels of the early Aragonian.

Equitability The equitability (Fig. 7.2) is the component of the diversity that describes the abundance distributions of the diVerent species of a community. The index used represents the proportion of the maximum diversity possible for a given association in which all species have the same abundance. Faunas

Trends in rodent assemblages

[Figure 7.2] Equitability curve calculated from the reciprocal Simpson’s index for the latest Ramblian to early Vallesian rodent faunas from the Daroca area. Faunas with an equitability close to 1 contain species with evenly distributed abundances, while faunas with an equitability close to 0 have generally one predominant species accompanied by rare ones.

with an equitability close to 1 contain species with evenly distributed abundances, while faunas with an equitability close to 0 have generally one predominant species accompanied by rare ones. The observed equitability for the rodent faunas of the Daroca-Calatayud area shows roughly a similar pattern to that of the diversity, although some diVerences are observed. Early MN4 has the most equitable faunas, rapidly followed by an interval of low equitability during late MN4. A recovery is observed during earliest MN5, after which a decrease takes place with a low between 13.75 and 12.6 Ma (MN6–early MN7/8). The early Vallesian faunas are equitable again, but less than the earliest MN4 ones. Three main periods can be recognized from 15 Ma onwards: 1) a period with rapidly changing equitabilities between 0.3 and 0.5 (15–13.75 Ma, late MN5), 2) a period with equitability close to 0.2 (13.75–11.5 Ma, MN6–7/8) and 3) a period with equitability close to 0.4 (10.9–10 Ma, MN9).

129

Miocene mammalian successions

130

[Figure 7.3] Scatter diagram of number of rodent species versus sample size. Straight lines connect test samples with final ones. Curves represent the function relating the species number to the sample size calculated after Sanders’ rarefaction method (1968).

Number of species and sample size Any richness measure, understood as an indicator of the wealth of species in an association, depends on the sample size, i.e. the larger the sample the larger the expected number of species. Several richness indices have been proposed which are , theoretically, independent of the sample size (Peet, 1974). However, two conditions are implicit when using such indices: 1) the functional relationship between the expected number of species and the sample size has to be equal among the studied communities, and 2) the functional relationship has to be known. In our succession these conditions are not met, as can be observed in Fig. 7.3. COL-D, a sample with a relatively even species distribution has a functional relationship between number of species and sample size that diVers considerably from the low equitable fauna of Solera (SOL). Since we can not assume that our samples have similar abundance distributions we use the species counts as a measure of species richness. Direct comparison of species counts needs equal sample size, which is not the case here (Fig. 7.3). To eliminate this inequality we

Trends in rodent assemblages

used the rarefaction method of Sanders (1968) corrected by Hurlbert (1971) which calculates the expected number of species of each sample reducing it to a standard size of 100 M1–2. This index is what we call species richness (Fig. 7.5). Localities with less than 100 M1–2 have not been included in this study. The rarefaction method has the problem that when comparing associations with variable equitability the richer communities such as the one from COL-D (Fig. 7.3) may have a lower species richness than another less rich community (e.g. SOL, Fig. 7.3) when the theoretical sample size decreases. In the given example of SOL and COL-D the critical value is situated at some 600 M1–2; above this value SOL has a higher species richness than COL-D which is reXected by the larger number of species of SOL, whereas below this value SOL has a lower species richness. It is further remarkable that the test sample of COL-D is situated on the theoretical curve constructed on the basis of the amount of M1–2 of the Wnal sample only. In order to minimize the sample size problem, and to show the maximum number of species that we found in each period the actual number of species (Fig. 7.4) per locality is also presented. The test sample of COL-D contains 58 M1–2 distributed over nine species, whereas the Wnal sample of this locality has 1987 M1–2 over twelve species. The three new species have a relative abundance of less than 1%. The test sample of VA8A has 238 M1–2 and eleven species whereas the numbers of the Wnal sample are 770 and twelve respectively. The only new species is again less abundant than 1%. The test sample of LUM3 yielded 178 M1–2 and Wve species, whereas the numbers of the Wnal sample are 1124 and eight. For the test sample and the Wnal one of VA7C these numbers are 182–seven and 1536–nine. In all these cases the new species in the large sample have a relative abundance of less than 1%.

Number of species The number of species (Fig. 7.4) is largest in the lower part of our succession (MN4–earliest MN5) and there is a strong drop at about 15.5 Ma. From late MN5 until 13.75 Ma species numbers increase in two steps towards relatively high values during MN6 between 13.6 and 13.2 My. During early MN 7/8 there is a low whereas late MN7/8 and MN9 faunas have a higher species number again, but not as high as during the early Aragonian.

131

Miocene mammalian successions

132

[Figure 7.4] Number of rodent species per fauna for the latest Ramblian to early Vallesian rodent faunas from the Daroca area.

Species richness The species richness (Fig. 7.5) is highest at the beginning of the Aragonian. During MN4 there is a progressive decrease followed by a slight recovery during early MN5 and a sudden drop at 15.4 Ma approximately. The low in species richness continues until 14.4 Ma and represents the minimum for MN5. A rapid recovery is observed toward 14 Ma followed by a general trend towards low species richness in early MN7/8. During the early Vallesian the species richness is higher than in MN7/8 of the late Aragonian.

Relative humidity and temperature The habitat preference of the various rodent genera and species are after van der Meulen & Daams (1992) with the following changes (see Figs. 7.6 and 7.7). All hamsters with complex dental pattern are considered to be inhabitants of relatively wet biotopes. One of the most important dental features for assignation of habitat is the length of the mesoloph(id)s. When these ridges

Trends in rodent assemblages

[Figure 7.5] Species richness per rodent fauna from the latest Ramblian to early Vallesian of the Daroca area in which the bias of sample size is eliminated by reducing each fauna to a common size of 100 using the rarefaction method of Sanders (1968) corrected by Hurlbert (1971).

are long in more than 25% of M2 the species is considered as an inhabitant of humid environments such as was observed in extant populations from Latin America by Hooper (1952). It implies that our as-yet unnamed Democricetodon species of small size and Megacricetodon collongensis in the lower part of our succession are considered to be ‘wet’ rodents. This does not refer to D. hispanicus which has medium long mesoloph(id)s. In our Wgures all cricetids with shorter mesolophs in M2 are considered to be inhabitants of relatively dry habitats. The lowermost part of our faunal succession (latest MN3–early MN4, 16.8–16.5 Ma) is characterized by a relatively dry phase which is evident from the predominance of ‘dry’ rodents, both qualitatively and quantitatively. The late MN4 (between 16.5 and 16 Ma) is thought to be a relatively wet phase on the basis of the increased frequencies of ‘wet’ rodents (Fig. 7.6). During this time the forest dwelling Anchitherium (Equidae) is a common constituent of our faunas. It is, however, curious that during this wet phase there are more dry taxa than wet ones (Fig. 7.7). An important dry phase started at about 15.5 Ma which is also evident in the large mammal faunas. Hispanotherium (running rhinoceros) is common in Spain during this time (Cerden ˜ o, 1989). At approximately

133

Miocene mammalian successions

134

[Figure 7.6] Relative abundance of rodents according to their habitat preference during the latest Ramblian to early Vallesian from the Daroca area.

14 Ma a strong cooling trend is observed in stable oxygen isotopes in the marine record (Flower & Kennett, 1993). From that time on, the fossil rodent record shows immigrants from north (eastern) Europe such as the Gliridae Myomimus dehmi, Muscardinus, Microdyromys complicatus, Paraglirulus werenfelsi, Bransatoglis, Myoglis; the Cricetidae Cricetodon, Megacricetodon rafaeli, M. minor, Democricetodon gaillardi; and the Eomyidae Eomyops. MN6 (c. 13.8–13 Ma) appears to be a relatively wet period. In late MN7/8 a new wet phase was initiated during which the ratio between the number of wet species and their relative abundance is more balanced.

Earliest and Latest Known Records Instead of using First and Last Appearance Datums (FAD and LAD) as deWned by Berggren & Van Couvering (1974), or FO and LO (First and Last

Trends in rodent assemblages

[Figure 7.7] Percentage of the number of rodent species according to their habitat preference during the latest Ramblian to early Vallesian from the Daroca area.

Occurrence), we prefer to use Earliest and Latest Known Record (EKR and LKR), as proposed by Pickford & Morales (1994). Since we are dealing with the restricted fossil record of the Calatayud-Daroca Basin only, we abstain from attributing regional importance to Earliest and Latest Known Records of taxa, such as implied by the terms FAD/LAD and FO/LO respectively. The total numbers of EKR and LKR are given in the middle of each local biozone. Fig. 7.8 shows that during late MN4 there is a signiWcant peak of events and that there is a minimum around 15.1 Ma during early MN5. The period covered by late MN5–MN7/8 is characterized by a high number of faunal events. Another minimum occurs during the earliest Vallesian when there are only three LKRs.

Discussion There is a positive correlation between the number of species and humidity, especially during MN4 and MN5 (Figs. 7.4 and 7.6). Diversity and

135

Miocene mammalian successions

136

[Figure 7.8] Cumulative representation of Earliest and Latest Known Records (EKR and LKR) of rodent faunas from latest Ramblian to early Vallesian in the Daroca area. Count of these fenomena are represented in the middle of the local zones, such as defined by Daams et al. (in prep.).

equitability (Figs. 7.1 and 7.2) on the contrary seem to be negatively correlated to humidity. The equitability is positively correlated to the frequency of dry rodents from late MN3 to MN6 (Figs. 7.2 and 7.6). On the whole the structure of the late MN3–MN6 Aragonian rodent communities is characterized by the long term presence of a group of dry representatives such as Heteroxerus, Atlantoxerus (Sciurinae), FahlbuschiaPseudofahlbuschia (Cricetidae), Armantomys and several other Myomiminae (Gliridae) with simple dental patterns (van der Meulen & Daams, 1992). Humid rodents are immigrants in the studied area and have limited statigraphical ranges. Their increase in frequency is accompanied by drops in diversity and equitability, in spite of the simultaneous increase of the species number. Therefore, we assume that the basical environment during the above mentioned part of the Aragonian is relatively dry, and that the more humid periods are perturbations that aVect the structure of the rodents’ community. The humid period that starts at the beginning of MN6 produces the minimum in equitability. The species which constitute the core of the dry group of early and middle Aragonian rodent faunas, disappear or are poorly represented. Therefore this drop in equitability seems to represent

Trends in rodent assemblages

the collapse of the old community structure, and a new community with immigrants from north(eastern) Europe is established. This faunal turnover is among others evidenced by the sudden replacement of Megacricetodon collongensis by Megacricetodon gersii which are the most frequent species in their respective faunas. This event is observed in the Aragonian type section between the fossiliferous levels of LUM20 and LUM21 (Krijgsman et al., 1994) and must have had a duration of less than 50 000 years since the attributed ages of these levels are 13.77 and 13.72 Ma respectively (Daams et al., in press). From late MN6 onwards humid faunas are more equitable than dry faunas. On the basis of the frequencies of wet rodents, the MN6–MN9 environment is more humid than during most part of the MN5 and the humid rodents such as Megacricetodon minor-debruijni, Myoglis meini, Microdyromys complicatus, Muscardinus thaleri-hispanicus, and Paraglirulus werenfelsi have longer time ranges than the humid representatives from MN4/5. Focussing on certain periods such as late MN5 (15–14 Ma) we may mention the inXuence of temperature changes such as observed in the marine record (Flower & Kennett, 1993). Initially the diversity is relatively high , then it drops to a low at approximately 14.4 Ma after which there is an evident recovery with a high between 14 and 14.2 Ma. This recovery is produced by immigration of species of northeastern origin of more temperate regions and is probably due to a steep temperature drop. This high is followed by a gradual decrease. Evident trends in equitability and relative dryness are absent. The number of earliest known records (EKR) is the highest of our faunal sequence. This produces a net increase of species richness and number of species (Figs. 7.4 and 7.5). The absence of any clear trend in the abundance distribution of the rodent associations caused by the temperature decrease during this dry phase implies that apparently the community was quite resilient (Putman, 1994). During MN6–7/8 (13.7–11.1 Ma) our faunas are predominantly dry, as inferred from the quantitative data (Fig. 7.6). It is striking, however, that the proportion of wet species is relatively high (Fig. 7.7) which is in contrast with the early Aragonian when the relative humidity is much higher for the same proportions of wet species. Therefore we can conclude that the proportion between number of ‘dry’ and ‘wet’ species is less reliable than their frequencies to infer humidity conditions. The early Vallesian (11–10 Ma) lacks the dense information such as we have for previous periods, and general conclusions are therefore more hazardous. The diversity, equitability, the number of species, and the species richness are higher than during the late Aragonian. The humidity

137

Miocene mammalian successions

138

increases as deduced from the frequencies of the wet rodents, and the number of wet rodent species increases as well.

Acknowledgements Financial support by the D.G.I.C.Y.T., Spain, projects No. PB92-0013 and PB95-114 for the members of the Spanish team is gratefully acknowledged. The Dutch member (AJvdM) thanks the State University of Utrecht for Wnancing.

References Berggren, W. A. & Van Couvering, J. A. 1974. The Late Neogene; biostratigraphy, geochronology, and paleoclimatology of the last 15 million years in marine and continental sequences. Palaeogeography, Palaeoclimatology and Palaeoecology, 16, 1–216. Cerden ˜ o, E. 1989. Revisio´n de la Sistema´ tica de los Rinocerontes del Neo´geno de Espan ˜ a. Ph. D. Thesis, Universidad Complutense de Madrid, Spain; 429 pp. Daams, R. & Freudenthal, M. 1988. Synopsis of the Dutch-Spanish collaboration program in the Neogene of the Calatayud-Teruel Basin. 1976–1986. In Biostratigraphy and Paleoecology of the Neogene Micromammalian Faunas from the Calatayud-Teruel Basin (Spain), M. Freudenthal (ed.). Scripta Geol., Spec. Issue 1, 3–18. Daams, R., van der Meulen, A. J., Alvarez-Sierra, M. A., Pela´ez-Campomanes, P., Calvo, J. P., Alonso-Zarza, M. A. & Krijgsman, W. Stratigraphy and sedimentology of the Aragonian in its type area (Prov. Zaragoza, Spain). Newsletters on Stratigraphy (in press). Flower, B.P. & Kennett, J. P. 1993. Middle Miocene ocean climate transition: high resolution oxygen and carbon isotopic record from deep sea drilling project site 588A, southwest PaciWc. Paleoceanography, 8, 6, 811–43. Hooper, E. T. 1952. A systematic review of the harvest mice (Genus Reithrodontomys) of Latin America. Museum of Zoology, University of Michigan, 77, 1–255. Hurlbert, S.H. 1971. The nonconcept of species diversity: a critique and alternative parameters. Ecology, 52, 4, 577–86. Krijgsman, W., Garce´s, M., Langereis, C. G., Daams, R., van Dam, J., van der Meulen, A. J., Agustı´, J. & Cabrera, L. 1996. A new chronology for the Middle to Late Miocene continental record in Spain. Earth and Planetary Science Letters, 142, 367–80. Krijgsman, W., Langereis, C. G., Daams, R. & van der Meulen, A. J. 1994. Magnetostratigraphic dating of the Middle Miocene climate change in the continental deposits of the Aragonian type area in the Calatayud Teruel Basin (Central Spain). Earth and Planetary Science Letters, 128, 513–26. Meulen, van der A.J. & Daams, R. 1992. Evolution of Early-Middle Miocene rodent

Trends in rodent assemblages

communities in relation to long term paleoenvironmental changes. Palaeogeography, Palaeoclimatology, Palaeoecology, 93, 227–53. Peet, R.K. 1974. The measurement of species diversity. Ann. Rev. Ecol. Syst., 5, 285–307. Pickford, M. & Morales, J. 1994. Biostratigraphy and palaeobiogeogarphy of East Africa and the Iberian Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 112, 297–322. Putman, R. J. 1994. Community ecology. Chapman & Hall, 178 pp. Sanders, H. L. 1968. Marine benthic diversity: a comparative study. Amer. Natur., 102, 243–82.

139

8 The Late Miocene small mammal succession from France, with emphasis on the Rhoˆ ne Valley localities Pierre Mein

Introduction Upper Miocene localities in France are relatively numerous, but not equally distributed through time. Only three localities are allocated to the Lower Vallesian. There are abundant local faunas from the Upper Vallesian and Lower Turolian, but the Middle and Upper Turolian are poorly represented. The fossiliferous sites occur in diVerent areas (Fig. 8.1 and Table 8.1). Near Lyon, in the Rhoˆne Valley (Ain County code number 01), the oldest fossiliferous locality Jujurieux (Mein, 1985), along the banks of the Ecotay river, is embedded in Serravallian marine sands, with Foraminifera, and occurs immediately above the last marine stratum. A preliminary list was given by Mein (1985). This fauna which is more recent than La Grive L3, is characterised by new species for the Rho ˆ ne Valley such as a new form of Democricetodon and a small species of Megacricetodon; the large form of Megacricetodon has disappeared. In spite of the absence of typical Vallesian elements, the strata of Jujurieux may be equivalent to the marine sands at Saint Fons (in the south suburb of Lyon) which has yielded a Hipparion mandible. An updated faunal list is given in the appendix. The locality Priay 2 (Welcomme et al., 1991) also lies just above marine sediments, at the basis of the ‘Marnes de Bresse’ Formation and has been referred to the Lower Vallesian. The locality of Douvre (Mein, 1984) is slightly younger. The sites of Amberieu 1, 2 and 3 belong to three diVerent levels in the same formation (Farjanel & Mein, 1984). In the same area, Soblay is a doline deposit which has yielded large mammals (Viret & Mazenot, 1948) and small mammals (Mein, 1970, 1984). On the opposite side of the Ain river, on the ‘coˆtie`re des Dombes’ escarpment, we Wnd a succession of localities: Sermenaz (Viret, 1937; Mein, Truc & Ballesio, 1972), Chasnes (Vilain, Mein & Truc, 1989), Mollon (Viret, 1939; Mein, Truc & Ballesio, 1972) South of Lyon, in Droˆme County (code number 26), we mention two diVerent localities: Les Bourbons near Chabeuil (Guerin & Mein, 1971) and Bernardie`re along the banks of the Galaure river (not yet published). In the Ise`re County (code number 38), in Xuviatile sands, lies the very rich locality of Dionay (Mein, 1984). In the Vaucluse County, we Wnd a lot of interesting sites: Lobrieu (Mein &

The Late Miocene small mammal succession

[Figure 8.1] Map of France showing the counties with fossiliferous localities.

141

Table 8.1. Succession of French micromammal localities during the Upper Miocene MN Zones

Main succession

13

69 – Lissieu 84 – Ratavoux

Possible correlations 06 – La Tour?

12 11

10

9

84 – Cucuron stade 07 – St Bauzile 01 – Mollon 84 – Lobrieu 01 – Ambe´rieu 3 26 – Bernardie`re 38 – Dionay 01 – Ambe´rieu 2 a 01 – Ambe´rieu 2 c 01 – Ambe´rieu 1 01 – Sermenaz 34 – Montredon 01 – Douvre 01 – Priay 2 49 – Doue´ la Fontaine 01 – Jujurieux (Ecotet)

66 – Vives 66 – Castelnou – 1? 01 – Chasnes 84 – Valre´as 84 – Pertuis 66 – Lo Fournas 6 26 – Les Bourbons 66 – Lo Fournas 7? 66 – Lo Fournas 1993

The number placed before the name of the localities refers to French counties.

Truc, 1966; Mein, 1984); Valre´as CD-46, near Dars farm (not yet published); Pertuis (Aguilar & Clauzon, 1982); Cucuron stade (Mein & Michaux, 1979); Ratavoux (Aguilar & Clauzon, 1982). In a diatomitic deposit of Arde`che County (code number 07), Saint Bauzile (Mein et al., 1984; Demarcq et al., 1989; Mein & Romaggi 1991), is famous for yielding complete skeletons.

Miocene mammalian successions

142

In the Durance Valley, the site of La Tour (code number 06), known through several levels, may represent the uppermost part of the Miocene (Aguilar, Calvet & Michaux, 1989). Near the Pyre´ne´es mountains, the locality of Vives (code number 66) has yielded some micromammals (Clauzon et al., 1982) and Montredon is an important locality in the west part of He´rault County (code number 34); its fauna, with large and small mammals has been revised in 1982 and 1988. The latest faunal list was given by Michaux (1988) (some new taxa are added here: Megaderma, Glirulus, Pannonicola, Neocricetodon). Unique in the west of France and embedded in marine sediments, the locality of Doue´ La Fontaine is typical of the Lower Vallesian having yielded a few rodents (Ginsburg et al., 1979; Aguilar, 1981). Besides sedimentary deposits, some Wssure Wllings contain faunas which may not be homogenous: such as Lissieu (code number 69) in the vicinity of Lyon (Mein, 1964; Hugueney & Mein, 1965; Mein, 1988) and, in the Pyre´ne´es Orientales County (code number 66), the Wssure Wllings of Lo Fournas 6 and Lo Fournas 7 (Aguilar et al., 1986; Aguilar & Michaux, 1996). Lo Fournas 1993 is a large block of breccia with no geological context, having been transported by road builders of the Eastern Pyrene´es. J. Aymar (personal communication) noticed one of these blocks (about 400 kg), which was richly fossiliferous (more than 20 000 teeth). When asking where the block came from, he was informed that it came from the quarry Lo Fournas (near Baixas). The fauna collected is close to that of Lo Fournas 7, but it diVers by the absence of Occitanomys faillati and Huezelerimys vireti. It is possible that this block came from the same Wssure but from an earlier deposit. In the same County, near Thuir, the localities of Castelnou 1 (Aguilar et al., 1995) and Castelnou 3 (Aguilar et al., 1991) show strange associations suggesting mixed faunas.

Primates For the Upper Miocene of France, our knowlegde about Primates is poor; only three localities having been listed: Pliopithecus, at Doue´-la-Fontaine and Priay 2, and Mesopithecus at Mollon.

Insectivora In contrast, Insectivora (Table 8.2) are well represented and diverse. Among Erinaceidae, Galerix is practically always present; Schizogalerix zapfei,

9

10

13 12 11

Lissieu Cucuron stade Mollon Chasnes Lobrieu Ambe´rieu 3 Bernardie`re Dionay Ambe´rieu 2 a Ambe´rieu 2 c Ambe´rieu 1 Soblay Lo Fournas 1993 Montredon Douvre Jujurieux

Galerix

•

• • •

• • •

• •

• • •

Schizogalerix

• • • •

•

Lanthanotherium

• • •

• •

Postpalerinaceus • • • •

•

•

Plesiodimylus

•

•

• • •

•

Dinosorex

•

•

•

•

• • • • • • • • • • • •

Crusafontina

INSECTIVORA

•

• • •

• •

•

Petenyia

Table 8.2. Insectivora from the French Upper Miocene

Paenelimnoecus • • • • • • • • •

Miosorex •

•

cf. Blarinoı¨des •

Talpa • •

• • • • • • •

• •

Desmanella •

•

• • • • • • • •

• • •

•

• • •

• • •

• •

Archaeodesmana

Miocene mammalian successions

144

[Plate 8.1] Vasseuromys pannonicus – Dionay, M1sin. (1.37 ; 1.55), Fig. 1; M1sin. (1.35 ; 1.23), Fig. 2; Glirulus aff.diremptus – Montredon,M1sin (0.83 ; 0.90), Fig. 3; Progonomys sp. – Dionay, M1dext. (1.55 ; 1.09), Fig. 4; Heteroxerus sp. – Valre´as,M3sin (1.80 ; 1.62), Fig. 5; Echinosoricinae indet. – Lobrieu, M3dext. (2.26 ; 3.04 ) or aberrant M2, Fig. 6; Occitanomys sp. – Lobrieu,M1sin. (2.45 ; 1.68), Fig. 7; Occitanomys hispanicus – Sermenaz,M1sin. (1.86 ; 1.22), Fig. 8; Pannonicola brevidens – Montredon, M2sin (1.65 ; 1.32) or aberrant Rotundomys, Fig. 9; Neocricetodon skoflexi – Lobrieu, M2sin (1.37 ; 1.11), Fig. 10; Ramys multicrestatus – Lobrieu, M2sin (1.29 ; 1.47), Fig. 11; M2sin (1.46 ; 1.31), Fig. 12.

known in Switzerland and in Spain at the beginning of the Upper Miocene, occurs only in the upper part of MN 10. Lanthanotherium sanmigueli is less frequent, but is present during MN 10 and MN 11. Postpalerinaceus vireti, present in MN 10, is absent in MN 11 but reappears during MN 13. One last upper molar from Lobrieu (Plate 8.1, Fig. 6) may represent an unknown Echinosoricinae or an aberrant M2 from Galerix. The Dimylid Plesiodimylus chantrei occurs at the beginning of MN 11, very similar to the form from La Grive. The Heterosoricid Dinosorex disappears from France in the lower part of MN 10. Paenelimnoecus and Petenyia are less frequent. Among Soricids, Crusafontina kormosi is the most common Insectivora during MN 10 and MN 11. Miosorex disappears in the lower part of MN 10.

The Late Miocene small mammal succession

New forms of Soricids appear at Lissieu (cf. Blarinoides and Soricinae indet.). Two Talpids are very frequent: Desmanella and Archaeodesmana; Talpa is rarer and Wnally only the oldest locality of Jujurieux yields an Urotrichus cf. dolichochir.

Chiroptera Chiroptera are rare, except in Soblay and in karst Wssures such as Lo Fournas 1993; they are often represented by isolated teeth whose determination is not easy. Megaderma is present at Montredon, Lo Fournas 1993, Soblay and Lissieu. Rhinolophus is known by a relatively large species R. csakvarensis in many localities and by the small R. lissiensis present only at Lissieu. Hipposideros occurs at Lo Fournas 1993 and Asellia at Lo Fournas and Soblay. During the Upper Miocene, Vespertilionidae were more diverse than in the Middle Miocene. Eptesicus cf. campanensis and Eptesicus cf. noctuloı¨des are present at Lo Fournas 1993 and Lobrieu. Pareptesicus may be present in Soblay; Pipistrellus appears at Dionay and Bernardie`re. It is the Wrst mention of this genus in Miocene epoch. Myotis is represented by several species and the small Miniopterid M. fossilis occurs in Lo Fournas and Lissieu. Some rare bats such as Taphozous at Lo Fournas 1993, Otonycteris at Soblay, Mormopterus at Douvre have been recorded at only one locality.

Lagomorphs During this period, the Lagomorphs are essentially represented by the species Prolagus crusafonti which is present and abundant in each locality. The large Eurolagus fontannesi occurs only at Soblay. At Lissieu, we note the Wrst record of Leporids testiWed by the presence of the genus Alilepus; in this particular instance, the Lagomorphs, even Prolagus, are under-represented.

Sciuridae and Castoridae The terrestrial Sciurids (see Table 8.3) are represented by the genus Spermophilinus, present until MN 12. Sometimes this genus is accompanied by a smaller form referred to as Tamias. Heteroxerus is rarer; three diVerent lineages occur. At the locality of Valre´as this genus is represented by an unnamed species which has very small and high crowned teeth, without

145

Miocene mammalian successions

Table 8.3. Scuridae and Castoridae from the French Upper Miocene

12 11

11

10

9

Cucuron Mollon Chasnes Valre´as Lobrieu Ambe´rieu 3 St Bauzile Dionay Ambe´rieu 2 a Ambe´rieu 2 c Ambe´rieu 1 Soblay Lo Fournas 1993 Montredon Douvre Priay 2 Doue´ la Fontaine Jujurieux (Ecotet)

• •

•

Trogontherium

Dipoides

Chalicomys

Miopetaurista

Albanensia

CASTORIDAE Pliopetaurista

Hylopetes

Blackia

Tamias

Heteroxerus

SCIURIDAE Spermophilinus

146

• •

• • • • • • • • • • •

• •

•

• •

•

• • • • •

•

• •

•

•

•

•

• •

• • • • • • •

• • •

• • •

•

• •

• •

• • • • •

•

entolophid (Plate 8.1, Fig. 5). Flying squirrels are diverse. We note the last appearance of Albanensia in MN 10 and Miopetaurista in MN 11. Blackia persists; Hylopetes and Pliopetaurista appear at the beginning of MN 10; they are unknown during the Middle and Upper Turolian but reappear at the beginning of the Ruscinian. Three diVerent Castorids (Table 8.3) are found during the Upper Miocene: Trogontherium minutum is the smallest and the most abundant. Dipoides is poorly represented at the end of MN 10 and the beginning of MN 11.

Gliridae The Glirids (Table 8.4) are still diverse during MN 10; they decrease later. Muscardinus is the most frequent Glirid in the Upper Miocene. We note the last records of Glirudinus and Eomuscadinus at Jujurieux (MN 9); it is the same for Microdyromys, Myoglis, and Paraglirulus, which are still present in the lower part of MN 10 (Douvre and Soblay). Eliomys appears in MN 10 in

The Late Miocene small mammal succession

Table 8.4. Gliridae from the French Upper Miocene

147

13 12 11

10

9

Lissieu Cucuron Mollon Valreas Lobrieu Ambe´rieu 3 St Bauzile Bernardie`re Dionay Ambe´rieu 2 a Ambe´rieu 2 c Ambe´rieu 1 Soblay Lo Fournas 1993 Montredon Douvre Jujurieux

•

•

Ramys

Vasseuromys

Myomimus

Paraglirulus

Graphiurops

Microdyromys

Glirulus

Eliomys

Glirudinus

Myoglis

Muscardinus

Eomuscardinus

Glis

GLIRIDAE

• •

• •

• •

• • • •

•

• • • • • • • • •

• • • • •

• • • • • • • •

• • • • • • •

•

• •

•

•

•

the South-West of France, but is absent in the Rhoˆne Valley until MN 12. Glirulus has a long chronological range with two forms, G. cf. diremptus MN 10 (Plate 8.1, Fig. 3) and G. lissiensis MN 10–13. In one locality, Ambe´rieu 1, these two species occur simultaneously (Farjanel & Mein, 1984). Graphiurops (Plate 8.2, Figs. 1–3) is known in the Rhoˆne Valley during the upper part of MN 10 and the beginning of MN 11. Myomimus, Vasseuromys (Plate 8.1, Figs. 1–2) and Ramys (Plate 8.1, Fig. 11) occur in few localities and disappear in MN 11 or MN 12. Glis appears only in MN 13. During the Vallesian and Lower Turolian, the diversity of Glirids in France is higher than it is in Spain and is comparable to that of Central Europe but with some diVerences.

Eomyidae – Zapodidae The Eomyids change little during the Upper Miocene. Two forms are present: Eomyops catalaunicus and Keramidomys; the second genus is rarer (Table 8.5).

Miocene mammalian successions

148

[Plate 8.2] Graphiurops austriacus – Bernardie`re, M1sin. (0.88 ; 0.96), Fig. 1; M1sin. (0.94 ; 0.91), Fig. 2; Graphiurops cf. austriacus – Ambe´rieu 3, M2sin. (1.08 ; 1.12), Fig. 3; Epimeriones austriacus – Ambe´rieu 3, Figs. 4–10; M1sin. (1.95 ; 1.27), Fig. 4; M2dext. (1.36 ; 1.12), Fig.5; M3sin. (1.06 ; 1.08), Fig. 6; M3sin (1.16 ; 1.07), Fig.7; M1sin. (2.11 ; 1.23), Fig. 8; M2dext. (1.47 ; 1.24), Fig. 9; M2dext (1.33 ; 1.15), Fig. 10; Arvicolid. indet. – Lobrieu M1dext. (-x 1,75-h.post: 3,50); lingual view, Fig. 11a; occlusal view, Fig. 11b; Microtocricetus molassicus – Douvre M1−2 sin. frag.(-x 1,09), Fig. 12.

9

10

13 12 11

Lissieu Cucuron Mollon Valre´as Lobrieu Ambe´rieu 3 Bernardie`re Dionay Ambe´rieu 2 a Ambe´rieu 2 c Ambe´rieu 1 Soblay Lo Fournas 1993 Montredon Douvre Priay 2 Doue´ la Fontaine Jujurieux

Hispanomys

•

• • • • •

•

•

•

Ruscinomys

•

Megacricetodon

• •

Eumyarion •

Democricetodon • •

Neocricetodon or Cricetulodon • • • • • • • • •

• • • • • •

Rotundomys • • • • • • • • •

•

Microtocricetus • •

Epimeriones •

•

•

• • • • • • • •

• •

•

Eomyops

EOMYIDAE

• • •

•

Keramidomys

CRICETIDAE

• •

•

•

•

• •

ANOMALOMYIDAE

ZAP.

Eozapus

Table 8.5. Eomyidae, Zapodidae, Cricetinae and Anomalomyidae from the French Upper Miocene

Anomalomys

•

• • • • • • •

•

Prospalax

Miocene mammalian successions

150

The Zapodids, as in Spain and Central Europe, are present with Eozapus intermedius; in the Rhoˆne Valley, it occurs in the upper part of MN 10; absent in MN 11 and MN 12, it reappears during MN 13.

Cricetidae – Anomalomyidae The Cricetids (Table 8.5) are represented by Middle Miocene genera such as Megacricetodon, Democricetodon, Eumyaryon lasting during MN 9; Hispanomys persists till MN 10 in the Rhoˆne Valley but becomes rare during the Lower Turolian: just one incisor is known from Mollon; Ruscinomys schaubi is known in MN 12. During the Upper Miocene, two related genera of Cricetinae occur in Europe; the former is Cricetulodon, which may be a direct successor of Democricetodon and which is essentially known in the Lower Vallesian. The latter, classically called Kowalskia is relatively frequent and diversiWed from the Upper Vallesian to the Ruscinian. After the new description of Neocricetodon grangeri by Daxner-Ho¨ck et al. (1996), it appears that the name Kowalskia is a junior synonym of Neocricetodon Schaub, 1934. This genus Neocricetodon is present in France throughout the Upper Miocene, with several lineages and sometimes occurs with some survivors of Cricetulodon (see faunal list for Dionay). Neocricetodon skoXeki (Kordos, 1987), formerly known from Central Europe, is present in several localities of the Rhoˆne Valley (one tooth from Lobrieu is illustrated in Plate 8.1, Fig. 10). A new species of Neocricetodon is represented in Douvre and Soblay and another at Ambe´rieu. Two diVerent species occur simultaneously in several localities: Ambe´rieu 1, Ambe´rieu 2a and 2c, Ambe´rieu 3, Lobrieu. At Lissieu two forms of diVerent size occur also. The holotype of Neocricetodon lavocati (Hugueney & Mein, 1965) was thought to be an upper M2 of a very small species; it now seems that this tooth represents an upper M3 of a medium size form. A revision of French Cricetinae is just issued by Freudenthal et al. (1998). Rotundomys is very common during MN 10 and persists, but is rare, during MN 11. Microtocricetus is known only along the edge of the Jura mountains at the end of MN 9 and at the base of MN 10 in Priay 2 and Douvre (Plate 8.2, Fig. 12). The gerbilloid Cricetid Epimeriones is present in two localities: relatively abundant at Ambe´rieu 3 (Plate 8.2, Figs. 4–10), this form is very rare at Lissieu. The Anomalomyids are represented by two genera during MN 10: Anomalomys in the South West of France and Prospalax in the Rhoˆne Valley. Prospalax is known until the top of MN 11 (Mollon). From this locality and

10

Douvre

Occitanomys hispanicus

•

•

Ambe´rieu 2 c

Ambe´rieu 1 Soblay Sermenaz Montredon

•

• Occitanomys sondaari

Dionay Ambe´rieu 2 a

Ambe´rieu 3

• Occitanomys clauzoni

• •

• Occitanomys sp.

• Huerzelerimys minor

• •

• Huerzelermys vireti

•

Lobrieu

•

•

• Stephanomys stadi

Mollon

• Occitanomys adroveri

11

• Rhagapodemus primaevus

Cucuron stade

• Stephanomys ramblensis

12

• Occitanomys cf. adroveri

Lissieu

13

MN zones Localities

• Progonomys cathalaı¨

• Prog. cf. cathaı¨ •

•

Species

Table 8.6. Succession of Upper Miocene Muridae stored in the University of Lyon

• Progonomys sp.

• Castromys sp.

• Parapodemus cf. lugdunensis

• Parapodemus lugduneisis

•

• •

•

•

Parapodemus barbarae •

• Parapodemus sp.

•

• Parapodemus pasquierae

•

•

• Apodemus gudrunae

Miocene mammalian successions

152

for this genus Kretzoi has coined a new genus Allospalax with no valid support. The Lower Turolian site of Lobrieu yielded a very damaged hypsodont tooth (Plate 8.2, Fig. 11) which diVers from Plio-Pleistocene Arvicolids by its synclines, the depth of which are more developed in the centre of the tooth than in the peripheral parts. This strange tooth is very important because it is the oldest record of an Arvicolid in Western Europe.

Muridae Murids (Table 8.6) are the most valuable elements for biochronology; they are under study and numerous works are still in press or have just been issued (Aguilar & Michaux, 1996; de Bruijn et al., 1996; Michaux et al., 1997; Martin Suarez & Mein, 1998). The oldest Murid known in France is Progonomys cathalai; Wrst described at Montredon, it appears in some older localities such as Priay 2 and Douvre. Occitanomys hispanicus appears sometimes later at Sermenaz where it is the only Murid found (but the number of teeth is low); this animal survived until the top of MN 10; later it is evolved into O. sondaari, characteristic of MN 11. At Soblay, there is an increase in the number of Murids to three species. Progonomys becomes a little bigger and it is called P. cf. cathalai; a second smaller form, with a connection between t6 and t9, represents a Parapodemus: P. cf. lugdunensis; a third lineage of tiny Murid is called Parapodemus species because the present material is not suYcient to allow greater precision. In the same area, but a short time later, we note four Murids at Ambe´rieu 1: Progonomys cf. cathalai, a little larger than at Soblay Parapodemus lugdunensis, smaller than at Soblay, the small rare Parapodemus sp., a large Occitanomys (without connection between t6 and t9): O. clauzoni. This level does not contain O. hispanicus, but this form reappears at Ambe´rieu 2c; we observe a great size change in Progonomys cf. cathalai: the new form has been described (Mein et al., 1993) under the name of Huerzelerimys minor. It seems to be a local descendant of Progonomys and is unlikely to be a migrant: a bore hole at Ambe´rieu joining Ambe´rieu 2 level to Ambe´rieu 1 level yielded several teeth of this lineage throughout the section (Farjanel & Mein, 1984).

The Late Miocene small mammal succession

Three metres above, at Ambe´rieu 2a, we observe four diVerent Murids, with a new lineage of a large Parapodemus, slightly smaller than Parapodemus barbarae; this new lineage, formerly called P. cf. barbarae, has been recently described as Parapodemus pasquierae (Aguilar & Michaux, 1996). The anagenesis between Parapodemus lugdunensis and Parapodemus barbarae may be challenged by the occurrence of two diVerent species coexisting during the uppermost part of the Vallesian and Lower Turolian. The locality of Dionay represents the highest known Vallesian fauna in France. The Murids consist of two forms with a very small Progonomys (Plate 8.1, Fig. 4), known only by one tooth which is wide and short as in the genus Occitanomys, but cusp 1 is not deplaced backwards. There is absolutely no connection between cusps 6 and 9 as we can see in Parapodemus sp. from Soblay. It may represent an unknown lineage of Murid. At the top of the Vallesian, we have recognised six diVerent lineages but the maximum number of species found in one locality is only four. Thus, in Western Europe, the Upper Vallesian is characterised by a large increase in Murid diversity; among the diVerent lineages, Huerzelerimys shows the most rapid size variation. The karstic Wssure Wlling of Lo Fournas 6 seems very near in age to Dionay, perhaps slightly older, taking into consideration the size of O. clauzoni.

Discussion and conclusions Some localities of this area with an abundant and diversiWed fauna may yield evidence for local climatic or environmental changes. In this case, the most interesting locality is probably Lissieu, which shows an overwhelming majority of Murids such as Apodemus–Rhagapodemus and a low representation of the group Stephanomys–Occitanomys. Glirids are represented by two species of Muscardinus, one of Glirulus and one of Glis; this grouping of species indicates a humid environment. The complete lack of Sciurids might mean a biotope which was too wet for the terrestrial Sciurids and not hot enough for Xying squirrels. In contrast, these squirrels are present but not abundant, at the locality of La Tour, where they are associated with Celadensia and Stephanomys– Occitanomys; this grouping suggests a drier and warmer environment. In addition, the fossiliferous localities of the Upper Miocene in France may oVer information of a biostratigraphic interest. Here, the Upper Miocene is very unequally represented. So, it is not possible to recognise a good succession in the Lower Vallesian, Middle and Upper Turolian. Only the Upper Vallesian and Lower Turolian are well represented.

153

Miocene mammalian successions

154

However, a subdivision of MN 10 into two or three levels is easy; an Upper level with Ambe´rieu 2, Dionay and Lo Fournas 6 is especially well characterised, showing a great diversity among Murids. The boundary between MN 10 and MN 11 may be observed at the Ambe´rieu section. The faunal event characteristic of MN 11 is, here, the appearance of Huerzelerimys vireti, and Occitanomys sondaari and the decline of Cricetids. In Spain, at this time, we note the appearance of Ruscinomys, but the Wnds of this lineage remains rare in France; at Mollon, only one lower incisor is known which is insuYcient to distinguish Ruscinomys from Hispanomys. In the Rhoˆne Valley, we do not observe the major extinction at the end of the Vallesian which is reported for Spain and Central Europe. Finally, this area may be compared with both Central Europe and Spain, which thereby, allows us to investigate the variations of populations in Europe during the Upper Miocene.

References Aguilar, J. P. 1981. Evolution des rongeurs mioce` nes et pale´oge´ographie de la Me´diterrane´e occidentale. The`se Science Montpellier, pp. 1–203. Aguilar J. P. 1982. Contribution a` l’e´tude des micromammife`res du gisement mioce`ne supe´rieur de Montredon (He´rault) -2-Les Rongeurs. Palaeovertebrata, Montpellier, 12 (3), p. 81–117, 12 Wg., 2 pl. Aguilar, J. P., Calvet, M. & Michaux, J. 1986. De´couvertes de faunes de micromammife`res dans les Pyre´ne´es–Orientales (France), de l’Oligoce`ne supe´rieur au Mioce`ne supe´rieur; espe`ces nouvelles et re´Xexion sur l’e´talonnage des e´chelles continentale et marine. C. R. Acad. Sc. Paris, t.303, Se´r. II, N°8, pp. 755–60, 2 pl. Aguilar, J. P., Calvet, M. & Michaux, J. 1989. La limite Mio-Plioce`ne dans le sud de la France d’apre`s les faunes de Rongeurs ; e´tat de la question et remarques sur les datations a` l’aide des rongeurs. Bolletino della Societa Paleontologica Italiana, 28 (2–3), pp. 137–45, 4 Wg., 3 tabl. Aguilar, J. P., Calvet, M. & Michaux, J. 1995. Les rongeurs du gisement karstique mioce`ne supe´rieur de Castelnou 1 (Pyre´ne´es Orientales, France). Geobios, 28 (4), pp. 501–10, pl. 66, 1 Wg. Aguilar, J. P. & Clauzon, G. 1982. Evolution ge´odynamique de la Provence septentrionale au cours du Mioce`ne supe´rieur et terminal d’apre`s les faunes de Rongeurs. C. R. Acad. Sc. Paris, t.294, Se´r. II, pp. 915–20, 1 pl. Aguilar, J. P. & Michaux, J. 1996. The beginning of the age of Murinae (Mammalia: Rodentia) in southern of France. Acta Zoologica Cracoviensia, 39 (1), 35–45. Aguilar, J. P., Michaux, J., Bachelet, B., Calvet, M. & Faillat, J. P. 1991. Les nouvelles faunes de Rongeurs proches de la limite mio-plioce`ne en Roussillon. Implications biostratigraphiques et bioge´ographiques. Palaeovertebrata, Montpellier, 20 (4), pp. 145–72, 2 pl.

The Late Miocene small mammal succession

Bruijn, H., de, van Dam, J. A., Daxner-Ho¨ck G., Fahlbusch, V. & Storch, G. 1996. The genera of the Murinae, endemic insular forms excepted of Europe and Anatolia during the Late Miocene and Early Pliocene. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 220–6. Columbia University Press, New York. Clauzon, G., Aguilar, J. P. & Michaux, J. 1982. De´couverte de rongeurs d’aˆge mioce`ne supe´rieur dans le bassin tertiaire de Ce´ret (Pyre´ne´es orientales ): implications stratigraphiques, structurales et pale´oge´ographiques. Bull. BRGM, (2), I, n°4, pp. 285–93, 4 Wg. Daxner-Ho¨ck, G., Fahlbusch, V., Kordos, L. & Wu, W. 1996. The Late Neogene Cricetid Rodent genera Neocricetodon and Kowalskia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 220–6. Columbia University Press, New York. Demarcq G., Mein, P., Ballesio, R. & Romaggi, J. P. 1989. Le gisement d’Andance (Coiron, Arde`che, France) et le Mioce`ne supe´rieur de la valle´e du Rhoˆne: un essai de corre´lation, marin-continental. Bull. Soc. Ge´ol. France, Paris, (8), t.V n°4, pp. 797–806, 1 pl., 3 Wg., 3 tabl. Farjanel, G. & Mein, P. 1984. Une association de mammife`res et de pollens dans la formation continentale des ‘Marnes de Bresse’, d’aˆge mioce`ne supe´rieur, a` Ambe´rieu (Ain). Ge´ol. France, n°1–2, pp. 131–48, 3 Wg., 3 pl. Freudenthal M., Mein P. & Martin Suarez, E. 1998. Revision of the Late Miocene and Pliocene Cricetinae from France and Spain. Treb. Mus. Geol. Barcelona, 7, 5–82. Ginsburg, L., Janvier, P., Mornand, J. & Pouit, D. 1979. De´couverte d’une faune de mammife`res terrestres d’aˆge valle´sien dans le falun mioce`ne de Doue´-la-fontaine (Maine et Loire). C. R. Somm. Soc. Ge´ol. Fr., fasc. 5–6, pp. 223–7. Guerin, C. & Mein, P. 1971. Les principaux gisements de mammife`res mioce`nes et plioce`nes du domaine rhodanien. Docum. Lab. Geol. Univ. Lyon, HS., pp. 131–70, 1 Wg., 1 tabl. Hugueney, M. & Mein, P. 1965. Lagomorphes et Rongeurs du Ne´oge`ne de Lissieu (Rhoˆne). Trav. Labo. Ge´ol. Fac. Sc. Lyon, N. S., no 12, pp. 109–23. Martin Suarez, E. & Mein, P. 1997. Revision of the genera Parapodemus, Apodemus, Rhagamys and Rhagapodemus (Rodentia, Mammalia). Geobios, 31 (1), pp. 87–97. Mein, P. 1964. Chiroptera (Mioce`ne) de Lissieu (Rho ˆ ne). 89e`me Congr. Soc. sav., Lyon, Avril 1964, pp. 237–53, 18 Wg. Mein, P. 1970. Les sciuropte`res (Mammalia, Rodentia) ne´oge`nes d’Europe occidentale. Geobios, Lyon, vol. 3, fasc. 3, pp. 7–77. Mein, P. 1975. Une forme de transition entre deux familles de rongeurs. Coll. Intern CNRS n°218, Paris 1973. Proble`mes actuels de Pale´ontologie-Evolution des Verte´bre´s, pp. 759–63, 9 Wg. Mein, P. 1984. Composition quantitative des faunes de mammife`res du Mioce`ne moyen et supe´rieur de la re´gion lyonnaise. Pale´obiol. Continent., Montpellier, XIV, pp. 339–46, 2 tabl. Mein, P. 1985. A new direct correlation between marine and continental scales in Rhodanian Miocene. VIIIth RCMNS, Budapest, 1986,Abstracts; Hunghar. Geol. Survey, pp. 377–9, 1 tabl.

155

Miocene mammalian successions

156

Mein, P. 1988. Le gisement karstique de Lissieu (Rhoˆne). Intern. Workshop ‘Continental Faunas at the Miocene/Pliocene Boundary’, Faenza, 1 p. Mein, P., Martin Suarez, E. & Agusti, J. 1993. Progonomys (Schaub 1938) and Huerzelerimys gen.nov. (Rodentia Mammalia) ; their evolution in western Europe. Scripta Geol., 103, pp. 1–18, 1 pl., 5 Wg. Leiden. Mein, P., Meon, H., Rommagi, J. P. & Samuel, E. 1984. La vie en Arde`che au Mioce`ne supe´rieur d’apre`s les documents trouve´s dans la carrie`re de la Montagne d’Andance. Nouv. Arch. Mus. Hist. nat. Lyon, fasc. 21 suppl., pp. 37–44, 2 Wg.1 pl., Lyon, 1983. Mein, P. & Michaux, J. 1979. Une faune de petits mammife`res d’aˆge turolien moyen (Mioce`ne supe´rieur) a` Cucuron (Vaucluse): donne´es nouvelles sur le genre Stephanomys (Rodentia) et conse´quences stratigraphiques. Geobios, n°12, fasc. 3 Lyon, 1979, pp. 481–5, 1 pl. Mein, P. & Rommagi, J. P. 1991. Un Gliride´ (Mammalia Rodentia) planeur dans le Mioce`nes supe´rieur de l’Arde`che : une adaptation non retrouve´e dans la nature actuelle. Geobios, Lyon, 1991, nouv. se´r. n°13, pp. 45–50, 1 pl., 2 Wg. Mein, P. & Truc, G. 1966. Facies et association faunique dans le Haut-Comtat Venaissin. Trav. Lab. Ge´ol. Lyon, nouv. se´r., n°13, 1966, pp. 273–6. Mein, P., Truc, G. & Ballesio, R. 1972. Age des formations de la Coˆtie`re de Dombes, a` la lumie`re d’e´le´ments pale´ontologiques nouveaux. C. R. Acad. Sc. Paris, t.274, 1972, pp. 2016–18. Michaux, J. 1988. Contributions a` l’e´tude du gisement mioce`ne supe´rieur de Montredon (He´rault). Les grands Mammife`res. Dix-Conclusions ge´ne´rales. Palaeovertebrata, Montpellier, Me´m. extra., pp. 189–92. Michaux, J., Aguilar, J. P., Montuire, S., WolV, A. & Legendre, S. 1997. Les Murinae ne´oge`nes du Sud de la France: e´volution et pale´oenvironnements. Geobios, Lyon, Me´m. Sp. n°20. Vilain, R., Mein, P. & Truc, G. 1989. Le contact Quaternaire-Mioce`ne (Marnes de Bresse) au camp de Chasnes, commune de Be´ligneux (Ain, France). Bull. Mens. Soc. Linn. Lyon, 58 (3), pp. 111–20, 3 Wg., 5 pl. Viret, J. 1937. Une coupe dans la ‘Coˆtie`re de Dombes’ au niveau de Sermenaz. Ann. Soc. Linn. Lyon, t.80, pp. 1–13, 2 Wg. Viret, J. 1939. Note pre´liminaire sur la formation bressane de Mollon (Ain) et sur de nouvelles faunes de Verte´bre´s et d’Inverte´bre´s qui s’y rencontrent. C. R. Soc. Ge´ol. France. Paris, fasc. 2, pp. 7–9. Viret, J. & Mazenot, G. 1948. Nouveaux restes de mammife`res dans le gisement de lignite pontien de Soblay (Ain). Ann. Paleont., Paris, 34, pp. 17–59, 3 Wg., 2 pl. Welcomme, J. L., Aguilar, J. P. & Ginsgurg, L. 1991. De´couverte d’un nouveau Pliopithe`que (Primates, Mammalia) associe´ a` des rongeurs dans les sables du Mioce`ne supe´rieur de Priay (Ain, France) et remarques sur la pale´oge´ographie de la Bresse au Valle´sien. C. R. Acad. Sci. Paris, t.313, Se´r. II, pp. 723–9.

The Late Miocene small mammal succession

Appendix Updated Faunal Lists For The Main Local Faunas MN 9 F-O1 Jujurieux

Insectivora Galerix cf. socialis (H. von Meyer, 1865) – Galerix sp. – Urotrichus cf. dolichochir (Gaillard, 1899) – Desmanella stehlini Engesser, 1972 – Talpa cf. minuta Blainville, 1840 – Plesiodimylus chantrei Gaillard, 1899 – Crusafontina endemica Gibert, 1975 – Miosorex cf. grivensis (Depe´ret, 1892) – Paenelimnoecus crouzeli Baudelot, 1972 – Dinosorex pachygnatus Engesser, 1972 – Lagomorpha Prolagus ‘œningensis (Kœnig, 1825) Rodentia Spermophilinus cf. bredai (H. von Meyer, 1848) – Heteroxerus grivensis (Major, 1903) – Trogontherium minutum (H. von Meyer, 1838) – Eomuscardinus cf. sansaniensis (Lartet, 1851) – Muscardinus hispanicus de Bruijn, 1966 – Glirudinus undosus Mayr, 1979 – Paraglirulus werenfelsi Engesser, 1979 – Ramys multicrestatus (de Bruijn, 1966) – Eomyops catalaunicus (Hartenberger, 1966) – Hispanomys bijugatus Mein & Freudenthal, 1971 – Megacricetodon cf. freudenthali Garcia-Moreno, 1986 – Democricetodon cf. nemoralis Agusti, 1981 – Anomalomys cf. gaudryi Gaillard, 1900 – Eumyarion cf. latior Schaub & Zapfe, 1953. MN 10 F-01 Douvre

Insectivora Plesiodimylus cf. chantrei Gaillard, 1899 – Talpa gilothi Storch, 1978 – Archaeodesmana vinea Storch, 1978 – Dinosorex cf. pachygnathus Engesser, 1972 – Crusafontina kormosi (Bachmayer & Wilson, 1970 ) – Chiroptera Myotis indet. – Mormopterus (Hydromops) helveticus (Revilliod, 1920) – Lagomorpha Prolagus crusafonti Lopez-Martinez, 1975 – Rodentia Spermophilinus cf. bredai (H. von Meyer, 1848) – Blackia miocaenica Mein, 1970 – Hylopetes sp. – Pliopetaurista bressana Mein, 1970 – Albanensia grimmi (Black, 1966) – Chalicomys jaegeri Kaup, 1832 – Trogontherium rhenanum Franzen & Storch, 1975 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Myoglis meini (de Bruijn, 1966) – Glirulus cf. diremptus (Mayr, 1979) – Eomyops catalaunicus (Hartenberger, 1967) – Rotundomys cf. montisrotundi (Schaub, 1944) – Microtocricetus cf. molassicus Falbusch & Mayr, 1965 – Cricetulodon bugesiensis Freudeuthal et al., 1998 – Prospalax petteri Bachmayer & Wilson, 1970 – Progonomys cathalai Schaub, 1938.

157

Miocene mammalian successions

158

F-01 Sermenaz

Insectivora Archeodesmana vinea Franzen & Storch, 1975 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Lagomorpha Prolagus crusafonti Lopez Martinez, 1975. Rodentia Occitanomys hispanicus ( Michaux, 1971).

F-66 Lo Fournas 1993

Insectivora Galerix iberica Mein & Martin Suarez , 1993 – Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Postpalerinaceus vireti Crusafont & Villalta, 1947 – Talpa gilothi Storch, 1978 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970) – Miosorex cf. grivensis (Depe´ret, 1892) – Chiroptera Megaderma cf. vireti Mein, 1964 – Rhinolophus csakvarensis Kretzoı¨, 1951 – Hipposideros cf. collongensis De´peret, 1892 – Asellia cf. mariaetheresae Mein, 1958 – Eptesicus cf. campanensis Baudelot, 1972 – Eptesicus cf. noctuloı¨des Lartet 1851 – Myotis antiquus Gaillard, 1899 – Myotis murinoı¨des Lartet, 1851 – Myotis cf. boyeri Mein, 1964 – Miniopterus sp. – Taphozous sp. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. bredai (H. von Meyer, 1848) – Heteroxerus huerzeleri Stehlin & Schaub, 1951 – Muscardinus hispanicus de Bruijn, 1966 – Eliomys sp. – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Microdyromys cf. complicatus de Bruijn, 1966 – Paraglirulus werenfelsi Engesser, 1972 – Myomimus dehmi de Bruijn, 1966 – Eomyops catalaunicus (Hartenberger, 1967) – Hispanomys mediterraneus Aguilar, 1982 – Neocricetodon sp. – Rotundomys montisrotundi (Schaub, 1944) – Anomalomys gaillardi Viret & Schaub, 1946 – Progonomys castillae Aguilar & Michaux, 1996.

F-01 Soblay

Insectivora Galerix cf. socialis (H. von Meyer, 1865) – Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Postpalerinaceus vireti Crusafont & Villalta, 1947 – Plesiodimylus cf. chantrei Gaillard, 1899 – Desmanella sp. – Talpa gilothi Storch, 1978 – Archeodesmana vinea Storch, 1978 – Dinosorex cf. pachygnatus Engesser, 1972 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Petenyia dubia Bachmayer & Wilson, 1970 – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970) – Chiroptera Megaderma cf. vireti Mein, 1964 – Rhinolophus csakvarensis Kretzoı¨, 1951 – Asellia mariaetheresae Mein, 1958 – Otonycteris sp. – Pareptesicus

The Late Miocene small mammal succession

priscus Zapfe, 1950 – Myotis (Myotis) sp. – Myotis antiquus Gaillard, 1899 – Myotis cf. murinoides (Lartet, 1851). Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Eurolagus fontannesi (Deperet, 1887). Rodentia Spermophilinus cf. bredai (H. von Mayer, 1848) – Heteroxerus grivensis (F. Major, 1903) – Blackia miocaenica Mein, 1970 – Hylopetes sp. – Albanensia grimmi (Black, 1966) – Pliopetaurista bressana Mein, 1970 – Chalicomys jaegeri Kaup, 1832 – Trogontherium rhenanum Franzen & Storch, 1975 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Myoglis meini (de Bruijn, 1966) – Glirulus cf. diremptus (Mayr, 1979) – Paraglirulus werenfelsi Engesser, 1972 – Eomyops catalaunicus (Hartenberger, 1967) – Hispanomys mediterraneus Aguilar, 1982 – Cricetulodon bugeysiensis Freudeuthal et al., 1998 – Rotundomys bressanus Mein, 1975 – Prospalax petteri Bachmayer & Wilson, 1970 – Progonomys cf. cathalai Schaub, 1938 – Parapodemus cf. lugdunensis Schaub, 1938 – Parapodemus sp. –

F-01 Ambe´ rieu 1

Insectivora Schizogalerix zapfei (Rabeder, 1973) – Talpa gilothi Storch, 1978 – Desmanella sp. – Archeodesmana vinea Storch, 1978 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Petenyia dubia Bachmayer & Wilson, 1970 – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970) – Chiroptera Myotis sp. – Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. bredai (H. von Mayer, 1848) – Tamias sp. – Blackia miocaenica Mein, 1970 – Pliopetaurista bressana Mein, 1970 – Chalicomys jaegeri Kaup, 1832 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Graphiurops austriacus Bachmayer & Wilson, 1970 – Glirulus cf. diremptus (Mayr, 1979) – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Eomyops catalaunicus (Hartenberger, 1967) – Keramidomys pertesunatoi (Hartenberger, 1967) – Eozapus intermedius Bachmayer & Wilson, 1970 – Hispanomys cf. mediterraneus Aguilar, 1982 – Neocricetodon ambarreusis Freudenthal et al., 1998 – Neocricetodon skoXeki (Kordos, 1987) – Rotundomys bressanus Mein, 1975 – Prospalax petteri Bachmayer & Wilson, 1970 – Progonomys cf. cathalai Schaub, 1938 – Parapodemus lugdunensis Schaub, 1938 – Parapodemus sp. – Occitanomys clauzoni Aguilar et al., 1986.

F-01 Ambe´ rieu 2 c

Insectivora Galerix sp. – Schizogalerix zapfei (Rabeder, 1973) – Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Talpa gilothi Storch, 1978 –

159

Miocene mammalian successions

Desmanella sp. – Archeodesmana vinea Storch, 1978 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Petenyia dubia Bachmayer & Wilson, 1970 – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970).

160

Chiroptera Rhinolophus csakvarensis Kretzoi, 1951 – Vespertilionidae indet. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. bredai (H. von Meyer, 1848) – Tamias sp. – Pliopetaurista bressana Mein, 1970 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Graphiurops austriacus Bachmayer & Wilson, 1980 – Eomyops catalaunicus (Hartenberger, 1967) – Keramidomys pertesunatoi Hartenberger, 1967 – Eozapus intermedius (Bachmayer & Wilson, 1970) – Hispanomys cf. mediterraneus Aguilar, 1982 – Neocricetodon ambarrensis Freudenthal et al., 1998 – Neocricetodon skoXeki (Kordos, 1987) – Rotundomys bressanus Mein, 1975 – Prospalax petteri Bachmayer & Wilson, 1970 – Parapodemus lugdunensis Schaub, 1938 – Huerzelerimys minor Mein et al., 1993 – Occitanomys hispanicus (Michaux, 1971).

F-01 Ambe´ rieu 2 a

Insectivora Galerix sp. – Schizogalerix zapfei (Rabeder, 1973) – Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Talpa gilothi Storch, 1978 – Desmanella sp. – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970). Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. bredai (H. von Mayer, 1848) – Hylopetes sp. – Pliopetaurista bressana Mein, 1970 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Graphiurops austriacus Bachmayer & Wilson, 1980 – Eomyops catalaunicus (Hartenberger, 1967) – Keramidomys pertesunatoi Hartenberger, 1967 – Rotundomys bressanus Mein, 1975 – Neocricetodon ambarrensis Freudenthal et al., 1998 – Neocricetodon skoXeki (Kordos, 1987) – Prospalax petteri Bachmayer & Wilson, 1970 – Huerzelerimys minor Mein et al., 1993 – Parapodemus lugdunensis Schaub, 1938 – Parapodemus pasquieri Aguilar & Michaux, 1996 – Occitanomys clauzoni (Aguilar et al., 1986).

F-38 Dionay

Insectivora Galerix sp. – Schizogalerix zapfei (Rabeder,1973) – Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Postpalerinaceus vireti Crusafont & Villalta, 1947 – Plesiodimylus cf. chantrei Gaillard, 1899 – Talpa gilothi

The Late Miocene small mammal succession

Storch, 1978 – Desmanella crusafonti Ru ¨ mke, 1974 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970). Chiroptera Rhinolophus csakvarensis Kretzoi, 1951 – Myotis boyeri Mein, 1964 – Pipistrellus sp. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus turolensis de Bruijn & Mein, 1968 – Heteroxerus cf. huerzeleri Stehlin & Schaub, 1951 – Pliopetaurista bressana Mein, 1970 – Chalicomys jaegeri Kaup, 1832 – Dipoides problematicus Schlosser, 1902 – Trogontherium minutum Franzen & Storch, 1975 – Muscardinus austriacus Bachmayer & Wilson, 1970 – Glirulus cf. diremptus (Mayr, 1979) – Graphiurops austriacus Bachmayer & Wilson, 1980 – Vasseuromys pannonicus (Kretzoi, 1978) – Eozapus intermedius (Bachmayer & Wilson, 1970) – Hispanomys mediterraneus Aguilar, 1982 – Neocricetodon skoXeki (Kordos, 1987) – Cricetulodon cf. sabadellensis Hartenberger, 1966 – Rotundomys cf. bressanus Mein, 1975 – Prospalax petteri Bachmayer & Wilson, 1970 – Progonomys sp. – Huerzelerimys minor Mein et al., 1993 – Parapodemus lugdunensis Schaub, 1938 – Occitanomys hispanicus (Michaux, 1971) – Occitanomys clauzoni (Aguilar et al., 1986).

MN 11 F-26 Bernardie`re

Insectivora Lanthanotherium sanmigueli Villalta & Crusafont, 1944 – Desmanella cf. crusafonti Ru ¨ mke, 1974 – Talpa gilothi Storch, 1978 – Archaeodesmana vinea (Storch, 1978) – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970). Chiroptera Pipistrellus sp. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Graphiurops austriacus Bachmayer & Wilson, 1980 – Rotundomys bressanus Mein, 1975 – Prospalax petteri Bachmayer & Wilson, 1970 – Parapodemus lugdunensis Schaub, 1938 – Occitanomys sondaari Weerd, 1976.

F-01 Ambe´ rieu 3

Insectivora Galerix sp. – Plesiodimylus cf. chantrei Gaillard, 1899 – Archaeodesmana vinea Storch, 1978 – Desmanella crusafonti Ru ¨ mke, 1974 – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Petenyia dubia Bachmayer & Wilson, 1970 – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970).

161

Miocene mammalian successions

162

Chiroptera Myotis sp. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. turolensis de Bruijn & Mein, 1968 – Blackia miocaenica Mein, 1970 – Pliopetaurista bressana Mein, 1970 – Miopetaurista cf. crusafonti (Mein, 1970) – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Graphiurops austriacus Bachmayer & Wilson, 1980 – Eomyops catalaunicus (Hartenberger, 1967) – Neocricetodon skoXeki (Kordos, 1987) – Neocricetodon sp. – Epimeriones cf. austriacus Daxner-Ho ¨ ck, 1972 – Prospalax petteri Bachmayer & Wilson, 1970 – Parapodemus lugdunensis Schaub, 1938 – Huerzelerimys vireti (Schaub, 1938) – Occitanomys sondaari Weerd,1976. F-84 Valre´as

Insectivora Desmanella crusafonti Ru ¨ mke, 1974. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975. Rodentia Heteroxerus sp. – Muscardinus austriacus Bachmayer & Wilson, 1970 – Myomimus dehmi (de Bruijn, 1966) – Neocricetodon skoXeki (Kordos, 1987) – Parapodemus lugdunensis Schaub, 1938 – Parapodemus pasquieri Aguilar & Michaux, 1996 – Huerzelerimys vireti (Schaub, 1938). F-84 Lobrieu

Insectivora Galerix sp. – Lanthanotherium sanmigueli Crusafont & Villalta, 1944 – Echinosoricinae indet. – Archaeodesmana vinea Storch, 1978 – Desmanella cf. crusafonti Ru¨mke, 1974 – Anourosorex kormosi Bachmayer & Wilson, 1970 – Petenyia dubia Bachmayer & Wilson, 1970 – Paenelimnoecus repenningi (Bachmayer & Wilson, 1970). Chiroptera cf. Eptesicus noctidoides (Lartet, 1851) – cf. Eptesicus campanensis Baudelot, 1970 Lagomorpha Prolagus crusafonti Lopez Martinez, 1975. Rodentia Spermophilinus turolensis de Bruijn & Mein, 1968 – Trogontherium rhenanum Franzen & Storch, 1975. – Muscardinus austriacus Bachmayer & Wilson, 1970 – Ramys multicrestatus (de Bruijn, 1966) – Eomyops catalaunicus (Hartenberger, 1967) – Hispanomys cf. peralensis Weerd, 1976 – Neocricetodon skoXeki (Kordos, 1987) – Neocricetodon sp. – Arvicolidae indet. – Parapodemus lugdunensis Schaub, 1938 – Parapodemus pasquieri Aguilar & Michaux, 1996 – Huerzelerimys vireti (Schaub, 1938) – Occitanomys clauzoni (Aguilar, Calvet & Michaux, 1986) Muridae indet.

The Late Miocene small mammal succession

F-01 Chasnes

Insectivora Crusafontina kormosi (Bachmayer & Wilson, 1970) – Chiroptera Rhinolophus csakvarensis Kretzoi, 1951 – Rodentia Pliopetaurista bressana Mein, 1970 – Neocricetodon sp. – Parapodemus lugdunensis Schaub, 1938 –

F-01 Mollon

Primates Mesopithecus sp. Insectivora Galerix sp. – Crusafontina kormosi (Bachmayer & Wilson, 1970) – Desmanella cf. crusafonti Ru ¨ mke, 1974 – Archaeodesmana vinea ( Storch, 1978) Chiroptera Myotis sp. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Rodentia Spermophilinus cf. turolensis de Bruijn & Mein, 1968 – Tamias sp. – Pliopetaurista bressana Mein, 1970 – Trogontherium rhenanum Franzen & Storch, 1975 – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Hispanomys indet. – Neocricetodon skoXeki (Kordos, 1987) – Rotundomys sp. – Prospalax petteri (Bachmayer & Wilson, 1970) – Parapodemus lugdunensis Schaub, 1938 – Huerzelerimys vireti (Schaub, 1938) –

F-07 Saint Bauzile

Lagomorpha Prolagus crusafonti Lopez Martinez, 1975. Rodentia Chalicomys jaegeri Kaup, 1832 – Dipoides problematicus Schlosser, 1902 – Trogontherium rhenanum Franzen & Storch, 1975 – Glirulus cf. lissiensis Hugueney & Mein, 1965 – Parapodemus lugdunensis Schaub, 1938 – Parapodemus pasquieri Aguilar & Michaux, 1996 –

MN 12 F-84 Cucuron stade

Insectivora Galerix (Parasorex) sp. – Archaeodesmana cf. pontica (Schreuder, 1940) – Desmanella crusafonti Ru ¨ mke, 1974 – Talpa gilhoti Storch, 1978 – Chiroptera Small Vespertilionidae indet. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 –

163

Miocene mammalian successions

164

Rodentia Spermophilinus cf. turolensis de Bruijn & Mein, 1968 – Eliomys truci Mein & Michaux, 1970 – Vasseuromys pannonicus (Kretzoi, 1978) – Ruscinomys schaubi Villalta & Crusafont, 1956 – Neocricetodon seseae Aguilar et al., 1995 – Parapodemus barbarae Weerd, 1976 – Occitanomys adroveri (Thaler, 1966) – Stephanomys stadii Mein & Michaux, 1979. MN 13 F-69 Lissieu

Insectivora Galerix sp. – Postpalerinaceus vireti Grusafont & Villalta, 1947 – Desmanella crusafonti Ru ¨ mke, 1974 – Talpa gilhoti Storch, 1978 – Petenyia dubia Bachmayer & Wilson, 1970 – cf. Blarinoides sp. – Soricini indet. Chiroptera Megaderma vireti Mein, 1964 – Rhinolophus csakvarensis Kretzoi, 1951 – Rhinolophus lissiensis Mein, 1964 – Myotis boyeri Mein, 1964 – Myotis sp. – Miniopterus cf. fossilis Zapfe, 1950. Lagomorpha Prolagus crusafonti Lopez Martinez, 1975 – Alilepus sp. – Rodentia Glis minor Kowalski, 1956 – Muscardinus vireti Hugueney & Mein, 1965 – Muscardinus davidi Hugueney & Mein, 1965 – Eomyops catalaunicus (Hartenberger, 1967) – Keramidomys cf. pertesunatoi Hartenberger, 1967 – Eozapus intermedius (Bachmayer & Wilson, 1970) – Neocricetodon lavocati (Hugueney & Mein, 1965) – Neocricetodon sp. – Epimeriones cf. austriacus Daxner-Ho¨ck, 1972 – Rhagapodemus primaevus (Hugueney & Mein, 1965) – Apodemus gudrunae Weerd, 1976 – Castromys sp. – Occitanomys cf. adroveri Thaler, 1964 – Stephanomys ramblensis Weerd, 1976 –

9 Late Miocene mammals from Central Europe Jens Lorenz Franzen and Gerhard Storch

Introduction The Vallesian and Early Turolian constitute a period of considerable faunal changes in Central Europe and even show a complete faunal turnover in some families. On the genus level, this period represents the time when the Middle Miocene fauna – still predominant during MN 9 – vanishes and immigrants appear which provide a modern stamp and often include ancestors of the extant fauna. Zone MN 10 is actually the crucial period of the faunal turnover. Therefore, the biostratigraphical subdivision does not follow the same standard throughout the present range charts: zone MN 9 is not further subdivided although there is not much doubt that Go¨tzendorf takes a position pretty close to the MN 9 – MN 10 boundary (see Ro¨gl et al., 1993) while Hammerschmiede and Vo¨sendorf occupy a more basal position within MN 9. On the other hand zone MN 10, although obviously a rather short period in absolute chronology, is subdivided for micromammals into an earlier and later component in order to assess faunal changes during that critical interval. The penecontemporaneous Early Turolian localities Eichkogel and Dorn-Du ¨ rkheim 1 are listed for micromammals separately to provide a basis for inferences about latitudinal diVerences.

Large mammals There are six groups of Vallesian macromammal sites known from Central Europe that are geographically, stratigraphically, and/or taphonomically diVerent. All are located south of the Main River (Fig. 9.1). 1. The Xuviatile ‘Sands with Deinotherium’ (‘Dinotheriumsande’) in the Mainz Basin southwest of the city of Mainz: localities of Eppelsheim, Bechtheim, Bermersheim, Dintesheim, Esselborn, Kettenheim, Oppenheim, Sprendlingen ( = Napoleonsho ¨ he = Steinberg), Vendersheim, Wahlheim, Westhofen, Wissberg ( = Gau-Weinheim), and Wolfsheim (Bartz, 1936; Gue´rin, 1980; Mein, 1986, 1989; Tobien, 1980a, 1983). The ‘Dinotheriumsande’ are the sediments of the earliest Rhine River. The fossils are accumulated in conglomeratic layers near the basis of the sequence. Some fossils are abraded and may have been reworked. Some genera comprise a very broad taxonomic spectrum (e.g. Deinotherium bavaricum alongside D. giganteum) that is otherwise

Miocene mammalian successions

[Figure 9.1] Distribution of Late Miocene (Vallesian and Turolian) macromammal sites in Central Europe. Symbols:  = Early Vallesian sites (MN 9),  = Late Vallesian sites (MN 10),  = Early Turolian sites (MN 11) of Dorn-Du¨rkheim 1 (Mainz Basin) and Eichkogel (Vienna Basin).

known from the Bavarian Upper Freshwater Molasse from very diVerent stratigraphic levels (Gra¨f, 1957). A leaf Xora became known recently from a basal clayish layer at the Napoleonsho ¨ he near Sprendlingen (Meller, 1989). 2. The Xuviatile deposits of the Elsgau (Ajoie, Switzerland; sites of Bois de Raube, Charmoille, Fregie´court, Lugnez, Montchabeut, Neux-Champs), located west of Basel, derive from the Vosges Mountains in the North. Therefore, they are also named ‘Vogesensande’. The fauna is still poorly known. Only the beavers have been described in some detail. Beside the mammals that come from the gravels there are also molluscs and plant remains embedded in intercalated marls (Schaefer, 1966). 3. The Wssure Wllings from the Suebian Alb, localities of Melchingen, Salmendingen, and TrochtelWngen, have been exploited for fossils since the nineteenth century, but not very systematically. Therefore, almost nothing is known about the taphonomy and the local stratigraphy of the sites (Mein, 1986, 1989; Schlosser 1902). 4. The volcanic lake of Ho¨wenegg northwest of Lake Constance was intensively excavated during the 1950s and early 1960s, and more recently in 1986 and 1992 (Tobien, 1980b; Woodburne et al., 1996). The locality is famous for its complete mammal skeletons (Hippotherium

Late Miocene mammals from Central Europe

primigenium, Aceratherium incisivum, Miotragocerus pannoniae). Recently it was absolutely dated to 10.3 Ma (Swisher, 1996). It also contains some pollen and leaves stemming from herbs as well as trees and lianas. 5. The Xuviolacustric deposits of the Vienna Basin, e.g. Gaiselberg, Vo¨sendorf, Go¨tzendorf (Ro¨gl & Daxner-Ho¨ck, 1996), and Mariathal (Mein, 1986, 1989; Thenius, 1982) are situated around the city of Vienna. These sites belong to a stratigraphically controlled sequence of the Pannonian that oVers the possibility to trace the evolutionary developments throughout the Vallesian. The site of Vo¨sendorf has also produced some molluscs and vegetational remains (Kovar-Eder et al., 1996). 6. The cave and Wssure Wllings of KohWdisch, located in the southern Burgenland (Austria) southwest of the city of Vienna, have delivered a rich micromammal fauna (Bachmayer & Wilson, 1970, 1978, 1980) as well as macromammals, mostly hyaenas (Ictitherium). Percrocutids (‘Percrocuta’ = Adcrocuta eximia) and Deinotherium giganteum are very rare. Concerning the macromammals, the fossil lagerstaette may have formed as a hyaena den whereas the micromammals are obviously coming from owl pellets (Bachmayer & Zapfe, 1969, 1972). The sites of the Bavarian Molasse (Upper Freshwater Molasse: Marktl, Hammerschmiede), and from the Czech Republic (Hovorany) delivered only small mammals. They are not considered here.

Biostratigraphy Most of the sites are early Vallesian in age (MN 9), except for KohWdisch (MN 10), and the Wssure Wllings from the Suebian Alb that are at least partly younger (Melchingen = MN 10). The argument for the stratigraphic classiWcation of KohWdisch is the appearance of the rodents Parapodemus, Kowalskia, Epimeriones, Pliopetaurista, and Progonomys (Ro¨gl & DaxnerHo¨ck, 1996). The stratigraphic position of Melchingen as MN 10 is deWned by the appearance of Adcrocuta eximia (Morlo, 1997), Ursavus depereti (Werdelin, 1996), and Tapiriscus pannonicus (Franzen, 1981, and in prep.) while hominoids as well as Brachypotherium goldfussi still persist (Fig. 9.2). Contrasting with the Wssure Wllings from the Suebian Alb, the ‘Sands with Deinotherium’ are very rich in taxa as well as individuals. They are represented by isolated teeth and bone fragments only, except for the more southerly situated localities (e.g. Eppelsheim, Kettenheim, Wahlheim) where skulls, mandibles, and complete postcranials were also found. In

167

Miocene mammalian successions

168

[Figure 9.2] First appearances (FADs) and last appearances (LADs) of Primates, Carnivora, and Proboscidea at the Vallesian/Turolian turnover (MN 9–11) in Central Europe. Dashed lines represent so-called Lazarus taxa. Data from Andrews et al. (1996); Bachmayer & Zapfe (1969, 1972); Gaziry (1997); Kuss (1965); Morlo (1997); Roth & Morlo (1997); Tobien (1980a).

Late Miocene mammals from Central Europe

spite of almost 180 years of research no micromammals have been discovered as yet (except for beavers). Therefore the stratigraphic position of the various localities of the ‘Dinotheriumsande’ remains somewhat ambiguous. It is not clear if they are all contemporary. The hominoids Now let us consider the hominoids. The complete femur of Paidopithex rhenanus (Pohlig, 1895) was discovered in 1820 as the world’s Wrst fossil hominoid in Eppelsheim. It is now regarded as cf. Dryopithecus sp. (Andrews et al., 1996) or as a pliopithecid (Begun, 1989). A small canine from the same locality initially identiWed as a cercopithecid and named Semnopithecus eppelsheimensis (Haupt, 1935) was later included in the new genus Rhenopithecus by von Koenigswald (1956). It is now considered as an undeterminable pliopithecid (Andrews et al., 1996). The Wissberg near Gau-Weinheim has delivered two isolated molars determined by von Koenigswald (1956) as Paidopithex rhenanus, and Rhenopithecus eppelsheimensis respectively. In spite of their size diVerences, both are considered as Dryopithecus sp. by Andrews et al. (1996). Among the Wssure Wllings, it is from Salmendingen that one lower molar comes, the type specimen of Dryopithecus brancoi (Schlosser, 1902), and a last lower milk molar, determined by Schlosser (1902) as Dryopithecus rhenanus (Pohlig sp.). The latter taxon is also known from Melchingen by two upper and two lower molars, and from TrochtelWngen by one lower molar. Andrews et al. (1996) regard the specimens from Salmendingen as well as those from Melchingen as Dryopithecus sp., whereas Begun & Kordos (1993) retain the species name Dryopithecus brancoi for the type specimen from Salmendingen as well as the large dryopithecine from Rudaba´nya. A deciduous molar from the same locality is regarded by Begun (1989) as an M1 of the large crouzeliine Anapithecus cf. hernyaki (Kretzoi, 1975), otherwise known from Rudaba´nya. In the Vienna Basin it is Go¨tzendorf (MN 9/10) that has delivered seven isolated hominoid teeth (I1d, I1d, C inf. d, P4s, M1d, M3 s + d) determined by Zapfe (1989) as belonging to Anapithecus hernyaki. From the Vallesian (MN 9) locality of Mariathal (Austria) comes an isolated molar (Mein 1986; Thenius 1982) considered by Andrews et al. (1996) as Dryopithecus carinthiacus (Mottl, 1957). It is strange, however, that the locality of Ho¨wenegg, well known for its articulated skeletons, has not produced any hominoid up to the time of writing.

169

Miocene mammalian successions

170

Palaeoecology As far as is recognisable, all Vallesian sites of Central Europe contain faunas that are forest-dominated although some open areas may have existed in the neighbourhood. Such a scenario is corroborated for the Napoleonsho ¨ he (‘Sands with Deinotherium’) where besides the mammals a Xora rich in leaves was found (Meller, 1989). This points to a gallery forest, and a moderately warm rain climate with a mean temperature for the warmest month of more than 22 °C, and annual precipitation of 1000–1200 mm. Somewhat diVerent conditions were obviously present at Vo¨sendorf and at the Ho¨wenegg. While the Xora from Vo¨sendorf is said to indicate a warm temperate climate of Mediterranean character (Thenius, 1954), the pollen and leaves of three species from the Ho¨wenegg point to marshy as well as dry environments (Gregor, 1982).

Toward a macromammalian biostratigraphy of the Late Miocene The biostratigraphic position of Dorn-Du ¨ rkheim 1, the one and only Turolian locality of Germany, is unquestionable. A fauna of more than 40 micromammal species indicates without any doubt an Early Turolian age (MN 11; Franzen & Storch, 1975; Storch et al., in prep.). The co-occurrence of micro- and macromammals at this site oVers a unique possibility to work out a biostratigraphy for the Late Miocene of Central Europe based on large mammals. Four groups are distinguished (Figs. 9.2 and 9.3). 1. Taxa that remain more or less unchanged from MN 9 to MN 11. Those are the carnivores Eomellivora wimanni, Paramachaerodus ogygius, Machairodus aphanistus, Ursavus primaevus, Indarctos arctoides, and Thalassictis robusta (Morlo, 1997; Roth & Morlo, 1997); the mastodont Tetralophodon longirostris (Gaziry, 1997); the perissodactyls Hippotherium primigenium, Tapirus priscus, Aceratherium incisivum, Lartetotherium schleiermacheri, and Chalicotherium goldfussi (Bernor & Franzen, 1997; Cerden ˜ o, 1997; Franzen, in prep.); and the artiodactyls Dorcatherium naui, Amphiprox anocerus, and Miotragocerus pannoniae (Fig. 9.3; Franzen, 1981; Azanza & Franzen, in prep). 2. Taxa that evolve in place. Particularly interesting is the genus Deinotherium (Fig. 9.4). It displays a lineage characterised by a considerable size increase from Deinotherium bavaricum (MN 4–5) through D. levius (MN 7–8) and D. giganteum (MN 9–10) to D. gigantissimum, as it is known, from the

Late Miocene mammals from Central Europe

171

[Figure 9.3] First appearances (FADs) and last appearances (LADs) of Perissodactyla and Artiodactyla at the Vallesian/Turolian turnover (MN 9–11) in Central Europe. Dashed lines represent so-called Lazarus taxa. Data from Azanza & Franzen (in prep.); Bernor & Franzen (1997); Cerden˜o (1997); Franzen (in prep.); Gue´rin (1980); Heissig (1996); Tobien (1980a); Van der Made (1997).

Miocene mammalian successions

172

[Figure 9.4] Size increase within the Deinotherium lineage from D. bavaricum (MN–4–5, ?9) through D. levius (MN 7–8) and D. giganteum (MN 9–10) to D. gigantissimum (MN 15) as demonstrated by the maximum lengths and the maximum breadths of the M2. The species of Deinotherium from Dorn-Du¨rkheim 1 fits perfectly within between D. giganteum and D. gigantissimum (data from Gra¨f, 1957; Stefanescu, 1995; Tobien, 1988, and own measurements).

Ruscinian (MN 15) of Romania and Bulgaria (Franzen & Storch, 1975; Tobien, 1988; Franzen, in prep.). At Dorn-Du ¨ rkheim 1 its record consists of a sample of cheek teeth that perfectly insert in size between D. giganteum and D. gigantissimum (Fig. 9.4). Also the new species of mastodonts, Stegotetrabelodon lehmanni and Stegolophodon caementifer, seem to evolve in place of Stegotetrabelodon gigantorostris, and Stegolophodon wahlheimensis respectively (Gaziry, 1997). 3. Taxa that become extinct in Central Europe. At the end of MN 9 these are most of the amphicyonids (Pseudocyon sansaniensis, Amphicyon major, and Pseudarctos bavaricus; Kuss, 1965); Gomphotherium angustidens among the mastodonts (Gaziry, 1997); the horse Anchitherium aurelianense (Tobien, 1980a); and almost all suids (Propotamochoerus palaeochoerus, Hyotherium soemmeringi, Listriodon splendens, Conohyus simorrensis, Microstonyx antiquus; Van der Made, 1997) and cervoids (Palaeomeryx sp., Euprox furcatus, Euprox dicranocerus, Heteroprox larteti, Dicroceros elegans; Franzen, 1981; Azanza & Franzen, in prep.). Many of the remaining Vallesian macromammals disappear at the

Late Miocene mammals from Central Europe

end of MN 10. These are all hominoids (dryopithecines and pliopithecids; Andrews et al., 1996; Begun, 1989), the last amphicyonids (Arctamphicyon sp.; Kuss, 1965), and the rhinoceros Brachypotherium goldfussi (Heissig, 1996). 4. Taxa that immigrate into Central Europe. This occurs over some time interval. The Wrst immigrants of the Vallesian/Turolian turnover arrived in MN 10. Those are the hyaena Adcrocuta eximia (Morlo, 1997), the dwarf tapir Tapiriscus pannonicus (Franzen, 1981; Franzen, in prep.), and the bear Ursavus depereti (Werdelin, 1996). A later wave appears in MN 11 at Dorn-Du ¨ rkheim 1. Those are the badger Promeles palaeatticus, the mustelid Baranogale cf. adroveri, the sabre-tooth Paramachaerodus orientalis, the bear Indarctos atticus, the hyaena Protictitherium crassum, the percrocutids Allohyaena kadici and Dinocrocuta sp., the cat Felis attica (Morlo, 1997; Roth & Morlo, 1997), the mastodont Anancus arvernensis (Gaziry, 1997), the rhinoceros Alicornops alfambrense (Cerden ˜ o, 1997), the suid Microstonyx erymanthius (Van der Made, 1997), the roe deer cf. Procapreolus, and the dwarf deer Cervavitulus mimus as well as a new species of Micromeryx (Azanza & Franzen, in prep.), and the antelope Tragocerus sp. (Franzen, 1981). Of particular interest in this context is Alicornops alfambrense known from Spain in MN 9 (La Roma). It appears in southern France (Montredon) in MN 10, and Wnally arrives at Dorn-Du ¨ rkheim 1 in MN 11. A comparable case seems to be Microstonyx erymanthius brevidens from Dorn-Du ¨ rkheim 1. This taxon is morphologically, chronologically, and geographically intermediate between M. major from MN 10 of Spain and M. erymanthius from MN 12 of Pikermi and Samos (Van der Made, 1997). Also Promeles palaeatticus and Paramachaerodus orientalis occur in Spain in MN 10 while they appear in East and Southeast Europe not earlier than MN 12 (Morlo, 1997). Ursavus depereti occurs in Spain in MN 9 of Nombrevilla 1, and perhaps even earlier (Werdelin, 1996). It arrives in southern France in MN 10 at Soblay and Luzinay, and at about the same time at Melchingen (Werdelin, 1996). Other immigrants came evidently from the East and Southeast. Those are Indarctos atticus, Allohyaena kadici, Dinocrocuta sp., Adcrocuta eximia, Tapiriscus pannonicus, and Cervavitulus mimus (Bernor et al., 1996; Roth & Morlo, 1997; Werdelin, 1996; Franzen, 1981, and in prep.). Anancus arvernensis is the earliest representative of the genus and species (Gaziry, 1997); it may have come from the East too.

173

Miocene mammalian successions

174

Admittedly, it is questionable, how complete the documentation of macromammals is in MN 10. The amount of ‘Lazarus taxa’ (taxa known from below and above but not from the intercalated level) leads to the assumption that the mammal record is rather incomplete. Therefore many Wrst appearances (FADs) or last appearances (LADs) of taxa are not exactly determinable as yet.

The faunal turnover at the Vallesian/Turolian boundary Taken altogether, there was a considerable turnover of the macromammalian fauna from MN 9 to MN 11. Particularly remarkable are the disappearance of all amphicyonids, and the diminishment of the suids and the cervoids. Among the carnivores more than 50% of the taxa are replaced. Remarkable is also the arrival of hyaenas and percrocutids. It is evidently at the MN 10/11 boundary that the hominoids vanish completely from Central Europe (Fig. 9.2). Apparently two lineages of hominoids exist into MN 10 (Begun, 1989), a dryopithecine (Dryopithecus sp.) and a pliopithecid (Anapithecus cf. hernyaki or Rhenopithecus eppelsheimensis). The mammal turnover is the more surprising because, as the fauna of Dorn-Du ¨ rkheim 1 shows, forests remained the ecological background in Central Europe. So it may have been a decrease of annual temperature and/or the disappearance of certain plants as diet which may have caused the disappearance of the herbivores mentioned, particularly the hominoids. Maybe it was the withdrawal of evergreens to the beneWt of deciduous elements of the Xora at this time (Kovar-Eder et al., 1996) that played the decisive role. No Xora, however, is known up till now from the Early Turolian of Central Europe. On the other hand, Central Europe evidently became a refuge area for forest dwelling animals at the Vallesian/Turolian turnover. Dryer climatic conditions expanding at this time from the Southwest as well as the East and Southeast were obviously pushing back the forest dwelling fauna toward Central Europe. Such a scenario may explain the immigration of taxa on one side from the Southwest and on the other side from the East and/or Southeast of Europe (Fig. 9.5).

Small mammals Central European micromammalian faunas of Vallesian and Early Turolian age (MN 9–MN 11) are rare and moreover, some Early Vallesian localities

Late Miocene mammals from Central Europe

175

[Figure 9.5] Palaeogeography and migrations of macromammals during the Vallesian and Early Turolian (MN 9–11) of Europe (palaeogeographic map from Ro¨gl & Steininger, 1984).

from southern Germany and Switzerland are not yet completely studied or published. The present biochronological sequence of insectivores and rodents is based on local faunas from the upper Rhinegraben, western Bohemia, the foothills of the Bavarian Alps, the Vienna Basin, and SE Austria. The localities Hammerschmiede (Mayr & Fahlbusch, 1975), Go¨tzendorf (Ro¨gl et al., 1993), Richardhof (Daxner-Ho¨ck, 1996), Eichkogel (Daxner-Ho¨ck, 1980; Rabeder, 1989), and Dorn-Du ¨ rkheim 1 (Franzen & Storch, 1975; Storch, 1978) are freshwater deposits, the site Vo¨sendorf (Bachmayer & Wilson, 1985) is a shallow brackish water deposit, and the localities Suchomasty (Fejfar, 1989) and KohWdisch (Bachmayer & Wilson, 1970, 1978, 1980) are cave and karst Wssure Wlls. In order to assess latitudinal ecological variation and faunal provinciality during the Late Miocene of Europe, a brief characterisation of palaeoenvironmental conditions of the local faunas will be given, with emphasis on Dorn-Du ¨ rkheim 1 in the very north of the area under consideration.

Comments on nomenclature and taxonomy The range charts do not include taxa beyond the genus. Thus, species names or questionable tribal assignments such as neomyine? shrew (at KohWdisch) and scalopine? mole (at Vo¨sendorf) (Bachmayer & Wilson, 1985) are not taken into consideration. Local occurrences of more than one species within a single genus (e.g., two species of Schizogalerix from KohWdisch and of Pliopetaurista from Dorn-Du ¨ rkheim 1) are not specially indicated in the

Miocene mammalian successions

176

tables. Some of the genus names, however, which are applied in the present paper may need some comments. Asoriculus: Episoriculus Ellerman and Morrison-Scott 1951 is considered either a distinct genus or a subgenus of Soriculus Blyth 1854 and refers to extant Himalayan species. Hutterer (1994) showed that the European fossil species of Episoriculus actually do not belong to this living (sub)genus and should be better assigned to the fossil genus Asoriculus Kretzoi 1959. Petenyia: Petenyia dubia Bachmayer & Wilson 1970 was originally described from KohWdisch and later transferred to the genus Blarinella Thomas 1911 by Reumer (1984). However, on the basis of rich comparative material, Storch (1995) strongly advocated the original generic allocation, and Storch & Qiu (1991, 1996) considered the genus Blarinella as indigenous to East Asia and unknown as fossil much beyond this area. Thus, it represents by no means a Tertiary relic. Archaeodesmana: The Late Miocene of Europe includes a widely distributed desmanine lineage with a chequered nomenclatural history. First referred with reservation to ‘Desmana’ (e.g. Storch, 1978), it was later allocated to a distinct new genus, Dibolia Ru¨mke 1985. Dibolia then turned out to be not available and a homonym of Dibolia Latreille 1829, a genus of beetles. Accordingly, the desmanine genus name was replaced by the new name Ruemkelia Rzebik-Kowalska and Pawlowski 1994, before Hutterer (1995) eventually advocated the validity of Archaeodesmana Topachevski and Pashkov 1983. The latter genus includes pontica Schreuder 1940 as type species and pontica unquestionably represents the lineage under discussion. Palaeomys: Kaup (1832) described three new genera and species of beavers from the Early Vallesian of Eppelsheim, Germany: Palaeomys castoroides, Chalicomys jaegeri, and Chelodus typus. All three taxa were later recognised as synonyms, based on teeth of diVerent stages of wear (e.g., von Meyer, 1838; Stirton, 1935; Hu¨nermann, 1966) and the single remaining species is recorded from various Late Miocene European localities. There remains, however, a problem of nomenclature. Franzen & Storch (1975) incorrectly claimed page priority to establish the validity of Palaeomys castoroides. Weerd (1976) correctly applied the ‘Principle of the First Reviser’ (ICZN, 3rd ed., Art. 24) but incorrectly followed Hu¨nermann (1966) in his choice of the valid name Chalicomys jaegeri. Actually, Stirton (1935) has to be considered the Wrst reviser and he selected Palaeomys castoroides, diagnosed the genus Palaeomys, and synonymised Chalicomys jaegeri and Chelodus typus. Accordingly, Palaeomys castoroides is used throughout the present paper. Glis: Brisson’s work Regnum Animale, Edn. 2, from 1762 is not available

------------------*----------

----------

------------------*----------

*-------------------

* = first appearance date, + = last appearance date.

Galerix Schizogalerix Lanthanotherium Erinaceus Plesiodimylus Dinosorex Crusafontina Paenelimnoecus Asoriculus Petenyia Talpa Scaptonyx Desmanella ‘Desmana’ Archaeodesmana

MN 9 Hammerschmiede, Go¨tzendorf, Vo¨sendorf

----------

----------

-------------------

-------------------

-------------------

-------------------

----------

-------------------

Eichkogel

-------------------

----------

----------------------------

Upper MN 10 Kohfidisch

---------------------------------------------------------------*-------------------------------------

----------

Lower MN 10 Suchomasty, Richardhof

----------------------------

----------+ ----------+ ----------+ -------------------------------------

----------+ ----------+

Dorn-Du¨rkheim

MN 11

Table 9.1. Succession of insectivore genera (Erinaceidae, Dimylidae, Soricidae, and Talpidae) in the Late Miocene of Central Europe

Miocene mammalian successions

178

because the principle of binominal nomenclature is not consistently applied. An application to the CINZ by A. Gentry to conserve 11 of Brisson’s generic mammal names, including Glis, and to reject Brisson’s work formally found emphatical support as well as categorical rejection (various comments to the BZN). Pending decisions by the Commission, Glis Brisson 1762 is used in the present paper instead of Myoxus Zimmermann 1780, which is supported by some students as the valid name. Eozapus: The type material of Protozapus Bachmayer & Wilson 1970 is from KohWdisch and the genus name was widely applied to European late Miocene zapodid samples. Fahlbusch (1992), on the basis of comparisons with the extant monospeciWc genus Eozapus Preble 1899, concluded that Protozapus is a junior synonym of this living East Asian genus. He retained, however, Sminthozapus Sulimski 1962 from various Pliocene sites as a separate genus which apparently represents a European branch of short time range.

The micromammal succession: the insectivores The insectivore fauna includes collectively an archaic component (Table 9.1). The genera Lanthanotherium, Plesiodimylus, and Dinosorex represent Middle Miocene groups and have their youngest occurrence in the lower MN 11 zone of Dorn-Du ¨ rkheim 1 and Eichkogel (the latter without Dinosorex). Two genera provide a modern stamp in that they make their Wrst appearance in Central Europe in the Vallesian and remain successful or become more diverse thereafter: Petenyia and Archaeodesmana. Two genera eventually seem to be restricted to, and thus characteristic for, zones MN 9 to MN 11 in Central Europe: Schizogalerix and Crusafontina (= ‘Anourosorex’). Schizogalerix has its oldest record at Vo¨sendorf and the latest known records are from Eichkogel and Dorn-Du ¨ rkheim 1. Crusafontina is Wrst documented at Hammerschmiede (Syn. Angustidens Mayr & Fahlbusch, 1975; see Mein, 1989) and thus far unknown beyond MN 11 of Eichkogel and Dorn-Du ¨ rkheim 1. Paenelimnoecus, Talpa, and Desmanella have extended ranges in Central Europe, spanning the period from Early Miocene (in Desmanella even Late Oligocene) through the Pliocene (Desmanella, Paenelimnoecus) and Recent (Talpa). The range chart suggests some slight increase in soricid diversity. Yet we are far away from Ruscinian ‘shrews’ paradise’ when local faunas may include around a dozen of taxa or more (Reumer, 1984; Rzebik-Kowalska, 1994). This Early Ruscinian enrichment of Central European soricid faunas

Albanensia Spermophilinus Pliopetaurista Hylopetes Blackia Miopetaurista Palaeomys Trogontherium Castor Dipoides gen. indet. Megacricetodon Microtocricetus Democricetodon Eumyarion Kowalskia Pseudomeriones Epimeriones Promimomys Collimys Cricetulodon

----------+ *----------------------------

-------------------------------------------------------

-------------------

MN 9 Hammerschmiede, Go¨tzendorf, Vo¨sendorf

----------+ ----------+ ----------+ *-------------------

----------

----------+ ---------*----------------------------

Lower MN 10 Suchomasty, Richardhof

-------------------+ ----------+

-------------------

*-------------------

----------

----------+ ----------------------------------------------------------------------------------

Dorn-Du¨rkheim

----------

----------

-------------------------------------

Eichkogel

MN 11

----------

----------

-------------------

Upper MN10 Kohfidisch

Table 9.2. Succession of rodent genera (Sciuridae, Castoridae, and Cricetidae) in the Late Miocene of Central Europe

Microdyromys Myoglis Paraglirulus Eliomys Glirulus Muscardinus Graphiurops Glis Myomimus Vasseuromys Dryomys Eomyops Keramidomys Parapodemus Progonomys Eozapus Anomalomys Prospalax Hystrix

*-------------------

----------

----------

----------+ ------------------*----------------------------

MN 9 Hammerschmiede, Go¨tzendorf, Vo¨sendorf

------------------*---------*----------------------------

----------+ ----------+ ---------------------------*-------------------

Lower MN 10 Suchomasty, Richerdorf

*-------------------

-------------------------------------+ ----------

------------------------------------*-------------------

Upper MN 10 Kohfidisch

-------------------+ ----------

-------------------

----------+ ----------+

----------------------------+

Eichkogel

----------

----------

*----------------------------+ ----------

----------

----------

Dorn-Du¨rkheim

MN 11

Table 9.3. Succession of rodent genera (Gliridae, Eomyidae, Muridae, Zapodidae, Anomalomyidae, Hystricidae) in the Late Miocene of Central Europe

Late Miocene mammals from Central Europe

was mainly caused by immigrations which had their starting point in East Asia. According to our recent results (Storch, 1995, and in prep.), East Asia itself received this immigrational stock of various taxa from North America not too long before the Wrst appearance in Central Europe. Thus, we should be cautious to explain changes in faunal richness and composition only by local or regional ecological changes. The actual cause may be geographically far away and ecologically unrelated to regional conditions – in this case the intermittent feasibility of the Bering corridor.

The micromammal succession: the rodents The succession of rodent genera is presented in Tables 9.2 and 9.3. Albanensia is a typical Middle Miocene petauristine sciurid and its last occurrence dates from lower MN 10 of Richardhof Niedero ¨ stereich. Miopetaurista, another Central European giant petauristine and the minute Blackia have very extended Neogene ranges and are recorded from the Early Miocene (Blackia even the Late Oligocene; Werner, 1994) through the Villanyian. A modern stamp is provided by the genus Pliopetaurista which makes its Wrst appearance at Suchomasty and is characteristic of the Central European Turolian and Pliocene; it continues in Eastern Europe into the Early Pleistocene. Hylopetes (including Pliopetes; see Bouwens & de Bruijn, 1986) has a rather poor record in Central Europe and is currently known to occur from MN 8 to MN 15b (de Bruijn, 1995). The sciurine squirrel Spermophilinus is rather common in the Central European Miocene, with records dating from MN 4 to MN 11. Beavers are generally characterised by the co-occurrence of Palaeomys, Trogontherium and Castor. Besides, Dipoides is known from the Late Vallesian locality of Salmendingen on the Suebian Alb and from DornDu ¨ rkheim 1. The latter locality even produced a Wfth beaver taxon of uncertain generic allocation. The hamsters exhibit an almost complete faunal turnover during the Late Vallesian. The typical Middle Miocene genera Megacricetodon, Democricetodon and Eumyarion date back in Central Europe to around MN 4 and they become extinct successively during MN 9 and the lower part of MN 10. Microtocricetus, Kowalskia, Epimeriones, and Collimys Wrst appear in the Late Miocene and contribute markedly to the cricetid faunal turnover. Microtocricetus seems to be conWned to the Vallesian. Collimys and Epimeriones Wrst appear in the Late Vallesian and Early Turolian and their association is characteristic of MN 11 in Central Europe. While Collimys does not seem to survive the Turolian, Epimeriones continues into the Early Ruscinian (Bohemia). The record of Cricetulodon at Dorn-Du ¨ rkheim 1

181

Miocene mammalian successions

182

marks its latest occurrence in Central Europe. The time range of Kowalskia starts in Central Europe at lower MN 10 and the genus vanishes there during the Ruscinian. The records of Promimomys and Pseudomeriones (determined in fact as aV. Pseudomeriones; Fejfar, 1989) are based on very poor material and in need of corroboration. The glirids, too, show a distinct shift in genus composition and richness during the Late Miocene. They can be basically subdivided into three types. First, genera with a long Miocene – in Microdyromys even Oligocene – record in Central Europe, and which successively become extinct during the Vallesian and Early Turolian: Microdyromys, Myoglis, Paraglirulus, and Vasseuromys. Second, a succession of immigrants which are mainly members of living genera and contribute to the modernisation of the Central European glirid fauna: Eliomys, Graphiurops, Myomimus, and Dryomys. The Late Miocene occurrence of Graphiurops and Myomimus as currently known is rather short-lived. The record of Dryomys at Dorn-Du ¨ rkheim 1 is the Wrst appearance date of this genus in Europe. Third, extant genera with an extended biostratigraphical range, dating back in Central Europe to the Early and Middle Miocene: Glis (known since MN 1), Glirulus and Muscardinus (including Eomuscardinus). The range of the eomyid Keramidomys spans the period from MN 5 to MN 11 (Eichkogel and Dorn-Du ¨ rkheim 1), while Eomyops continues into the late Villanyian (MN 17; Schernfeld). The small number of Central European murid genera during the Late Miocene is striking. Parapodemus and Progonomys appear Wrst in the lower part of MN 10 and Progonomys is conWned to MN 10. Eozapus does still exist as a relic in China. The genus makes its Wrst appearance in the Early Vallesian and continues into the Late Ruscinian of Central Europe. The anomalomyid Anomalomys is a characteristic Middle Miocene genus and has its latest record in the Early Turolian of Eichkogel. The genus Prospalax, as here understood, Wrst appears in the upper part of MN 10 and continues into the Central European Pliocene. There is only a single record of Hystrix known (at KohWdisch) from the localities under discussion.

Late Vallesian–Early Turolian rodent succession by individual frequencies Fig. 9.6 shows the percentage of molar numbers in rodent families from MN 10 to MN 11 of Central Europe. Suchomasty still represents a typical Middle Miocene pattern of frequency with a glirid apex and high individual

Late Miocene mammals from Central Europe

[Figure 9.6] Number of molars (in %) of rodent families from Late Vellesian–Early Turolian localities.

numbers of eomyids. These high frequencies of occurrence are replaced during the upper part of MN 10 and MN 11 of KohWdisch, Eichkogel and Dorn-Du ¨ rkheim 1 by the predominance of murids. KohWdisch, by the individual number of dormice, shows a somewhat intermediate position between older Vallesian and Turolian patterns. The turnover from a glirid to a murid apex is not correlated with a general change in species richness since the murid apex is caused at Eichkogel and Dorn-Du ¨ rkheim 1 by a single species while glirids still number Wve and three. The present data also do not support the view that the predominance of murids by individual numbers was related to an increase of dense forest cover. The palaeoenvironment of KohWdisch, for instance, can obviously be characterised on the basis of the thermophilous and xerophilous herpetofauna as largely open country with well-drained woodland and prairie (Bachmayer & Wilson, 1985).

Late Vallesian–Early Turolian succession by species numbers Fig. 9.7 shows the number of species of the same rodent families and fossil sites as in Fig. 9.6. Compared to Fig. 9.6, the pattern diVers considerably. At Suchomasty, KohWdisch, and Eichkogel the peaks are built by glirids while one or two murid species had dominated in individual numbers. There is a decline in glirid species numbers in the course of time which cannot, however, be directly related to a decline of forest cover. Dorn-Du ¨ rkheim 1 with three glirid species represents a well-watered and richly structured woodland biotope, much more so than either Suchomasty with its eight glirids (according to Fejfar (1989) forested country with shrubs, balanced

183

Miocene mammalian successions

184

[Figure 9.7] Number of species of rodent families in Late Vallesian–Early Turolian localities.

with open savannah-like landscape during the time of accumulation) or KohWdisch with six glirids. Dorn-Du ¨ rkheim 1, located in the north of the area under consideration, departs strikingly from the pattern of the other local faunas by high numbers of Xying squirrels and beavers (Wve species each).

Note on palaeoenvironments The micromammalian fauna of Dorn-Du ¨ rkheim 1 in the Upper Rhinegraben mirrors an ecosystem of diverse woodland areas and a great variety of water bodies as clearly indicated by the variety of Xying squirrel and beaver taxa. Among Dorn-Du ¨ rkheim 1 insectivores, the semiaquatic watermole Archaeodesmana provides about half of the specimens and the shrew Crusafontina, like its extant relative Anourosorex most likely an inhabitant of moist forest Xoors, almost one third (Fig. 9.8). The locality was obviously part of a temperate deciduous forest belt in the north of Central Europe, which was adjacent to a Turolian sclerophytic forest zone further south in Europe and represented some sort of a continuation of the once widespread Early Vallesian woodland. Early Vallesian (MN 9) Xoras from the Upper Rhinegraben (Meller, 1989), the Upper Freshwater Molasse in southern Germany (Jung & Mayr, 1980) and the Vienna Basin (Draxler & Zetter, 1993) represent mainly mesophytic deciduous forest which is rich in Fagus and shows a dominance of temperate tree families such as Aceraceae, Ul-

Late Miocene mammals from Central Europe

[Figure 9.8] Number of M1 of insectivore genera from Dorn-Du¨rkheim 1.

maceae, Tiliaceae and Betulaceae. The corresponding survival of oldtimers among micromammals is exempliWed by the last occurrence date of a typical Middle Miocene genus such as Dinosorex.

References Andrews, P., Harrison, T., Delson, E., Bernor, R. L. & Martin, L. 1996. Distribution and Biochronology of European and Southwest Asian Catarrhines. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 168–207. New York, Columbia University Press. Bachmayer, F. & Wilson, R. W. 1970. Die Fauna der altplioza¨nen Ho¨hlen- und Spaltenfu ¨ llungen bei KohWdisch, Burgenland (O ¨ sterreich). Annalen des Naturhistorischen Museums Wien, 74, 533–87. Bachmayer, F. & Wilson, R. W. 1978. A second contribution to the small mammal fauna of KohWdisch, Austria. Annalen des Naturhistorischen Museums Wien, 81, 129–61. Bachmayer, F. & Wilson, R. W. 1980. A third contribution to the small mammal fauna of KohWdisch (Burgenland), Austria. Annalen des Naturhistorischen Museums Wien, 83, 351–86. Bachmayer, F. & Wilson, R. W. 1985. Environmental signiWcance and stratigraphic position of some mammal faunas in the Neogene of eastern Austria. Ann. Naturhist. Mus. Wien, 87A, 101–14. Bachmayer, F. & Zapfe, H. 1969. Die Fauna der altplioza¨nen Ho¨hlen- und Spaltenfu ¨ llungen bei KohWdisch, Burgenland (O ¨ sterreich). Geologische und biostratinomische Verha¨ltnisse der Fundstelle, Ausgrabungen. Annalen des Naturhistorischen Museums Wien, 73, 123–39. Bachmayer, F. & Zapfe, H. 1972. Die Fauna der altplioza¨nen Ho¨hlen- und Spaltenfu ¨ llungen bei KohWdisch, Burgenland (O ¨ sterreich). Proboscidea. Annalen des Naturhistorischen Museums Wien, 76, 19–27.

185

Miocene mammalian successions

186

Bartz, J. 1936. Das Unterplioca¨n in Rheinhessen. Mitteilungen des Oberrheinischen Geologischen Vereins, 25, 119–226. Begun, D. R. 1989. A large pliopithecine molar from Germany and some notes on the Pliopithecinae. Folia Primatologica, 52, 156–66. Begun, D. R. & Kordos, L. 1993. Revision of Dryopithecus brancoi Schlosser, 1901, based on the fossil hominoid material from Rudaba´nya. Journal of Human Evolution, 25, 271–85. Bernor, R. L., Fahlbusch, V., Andrews, P., De Bruijn, H., Fortelius, M., Ro¨gl, F., Steininger, F. F. & Werdelin, L. 1996. The Evolution of Western Eurasian Neogene Mammal Faunas: A Chronologic, Systematic, Biogeographic, and Paleoenvironmental Synthesis. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds,), 449–69. New York, Columbia University Press. Bernor, R. L. & Franzen, J. L. 1997. The equids (Mammalia, Perissodactyla) from the Late Miocene (Early Turolian) of Dorn-Du ¨ rkheim 1 (Germany, Rheinhessen). In Die Sa¨ugetiere aus dem Turolium von Dorn-Du¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 117–86. Bouwens, P. & Bruijn, H. de 1986. The Xying squirrels Hylopetes and Petinomys and their fossil record. Proc. Koninkl. Nederl. Akad. Wet., Ser. B, 89, 113–23. Bruijn, H. de 1995. The vertebrate locality Maramena (Macedonia, Greece) at the Turolian-Ruscinian boundary (Neogene) 8. Sciuridae, Petauristidae and Eomyidae (Rodentia, Mammalia). Mu ¨ nchner Geowiss. Abh., A28, 87–102. Cerden ˜ o, E. 1997. Rhinocerotidae from the Turolian site of Dorn-Du ¨ rkheim 1 (Germany). In Die Sa¨ugetiere aus dem Turolium von Dorn-Du ¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 187–203. Daxner-Ho¨ck, G. 1980. Rodentia (Mammalia) des Eichkogels bei Mo¨dling (Niedero¨sterreich). 1. Spalacinae und Castoridae. 2. U ¨ bersicht u¨ber die gesamte Nagetierfauna. Ann. Naturhist. Mus. Wien, 83, 135–52. Daxner-Ho¨ck, G. 1996. Faunenisandel in Obermioza¨n und Korrelation der MN – ‘Zonen’ mit der Biozonen des Pannons der Zentralen Paratethys. Beitra¨ge Zur Pala¨ontologie, 21, 1–9. Draxler, I. & Zetter, R. 1993. 5. Palynologische Untersuchungen in Go¨tzendorf. In Die Primatenfundstelle Go¨tzendorf an der Leitha (Obermioza¨n des Wiener Beckens, Niedero¨sterreich). Ro¨gl, F. et al. (eds.). Jb. Geol. B.-A., 136, 514–15. Fahlbusch, V. 1992. The Neogene mammalian faunas of Ertemte and Harr Obo in Inner Mongolia (Nei Mongol), China. 10. Eozapus (Rodentia). Senckenbergiana Lethaea, 72, 199–217. Fejfar, O. 1989. The Neogene VP sites of Czechoslovakia: A contribution to the Neogene terrestric biostratigraphy of Europe, based on rodents. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.). NATO ASI Series, Ser. A, 180, 211–36. Franzen, J. L. 1981. Eine turolische (obermioza¨ne) Waldfauna aus Rheinhessen. Alzeyer Geschichtsbla¨tter, 16 (1981), 192–202. Franzen, J. L. & Storch, G. 1975. Die unterplioza¨ne (turolische) Wirbeltierfauna von Dorn-Du¨rkheim, Rheinhessen (SW-Deutschland). 1. Entdeckung, Geologie,

Late Miocene mammals from Central Europe

Mammalia. Carnivora, Proboscidea, Rodentia. Grabungsergebnisse 1972–1973. Senckenbergiana Lethaea, 56, 233–303. Gaziry, A. W. 1997. Die Mastodonten (Proboscidea, Mammalia) aus Dorn-Du ¨ rkheim 1 (Rheinhessen). In Die Sa¨ugetiere aus dem Turolium von Dorn-Du ¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 73–115. Gra¨f, I. E. 1957. Die Prinzipien der Artbestimmung bei Dinotherium. Palaeontographica, A108, 131–85. Gregor, H. J. 1982. Die jungtertia¨ren Floren Su ¨ ddeutschlands: Pala¨okarpologie, Phytostratigraphie, Pala¨oo¨kologie, Pala¨oklimatologie, 1–278. Stuttgart: Enke. Gue´rin, C. 1980. Les Rhinocer´os (Mammalia, Perissodactyla) du Mioce´ne Terminal au Pleistoce´ne Supe´rieur en Europe Occidentale. Comparison avec les espe`ces actuelles. Docum. Lab. Geol. Lyon, no. 79 (fasc. 1), pp. 1–421. Haupt, O. 1935. Andere Wirbeltiere des Neozoikums. In Oberrheinischer Fossilienkatalog, W. Salomon-Calvi (ed.), 4 (9), 1–103. Heissig, K. 1996. The Stratigraphical Range of Fossil Rhinoceroses in the Late Neogene of Europe and the Eastern Mediterranean. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 339–47. New York, Columbia University Press. Hu ¨ nermann, K. A. 1966. Der Bau des Biber-Pra¨molaren und seine Verwendbarkeit fu ¨r die Systematik der Castoridae (Rodentia, Mammalia). N. Jb. Geol. Pala¨ont. Abh., 125, 227–34. Hutterer, R. 1994. Generic limits among neomyine and soriculine shrews (Mammalia: Soricidae). Abstr. Conf. Neogene and Quaternary Mammals of the Palaearctic, Krako´w, 32. Hutterer, R. 1995. Archaeodesmana Topachevski & Pashkov, the correct name for Dibolia Ru ¨ mke, a genus of fossil water moles (Mammalia: Talpidae). Bonn. Zool. Beitr., 45, 171–2. Jung, W. & Mayr, H. 1980. Neuere Befunde zur Biostratigraphie der Oberen Su¨ßwassermolasse Su¨ddeutschlands und ihre palo¨kologische Deutung. Mitt. Bayer. Staatsslg. Pala¨ont. hist. Geol., 20, 159–73. Kaup, J. J. 1832. Beschreibung dreyer Gattungen urweltlicher Nager des zoologischen Museum zu Darmstadt, welche von den jetzt lebenden Genera verschieden sind. Isis, 9, 992–6. Koenigswald, G. H. R. von 1956. Gebissreste von MenschenaVen aus dem Unterplioza¨n Rheinhessens I-II. Verhandelingen der Koninklijk Nederlandsche Akademie van Wetenschappen Amsterdam, B 59 (4), 318–34. Kovar-Eder, J., Kvacek, Z., Zastawniak, E., Givulescu, R., Hably, L., Mihajlovic, D., Teslenko, J. & Walther, H. 1996. Floristic Trends in the Vegetation of the Paratethys Surrounding Areas During Neogene Time. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 395–413. New York, Columbia University Press. Kretzoi, M. 1975. New ramapithecines and Pliopithecus from the lower Pliocene of Rudaba´nya in north-eastern Hungary. Nature, 257, 578–81. Kuss, S. 1965. Revision der europa¨ischen Amphicyoninae (Canidae, Carnivora, Mamm.) ausschließlich der voroberstampischen Formen. Sitzungsberichte der Heidelberger Akademie der Wissenschaften, Jahrgang 1965 (1), 1–168.

187

Miocene mammalian successions

188

Mayr, H. & Fahlbusch, V. 1975. Eine Unterplioza¨ne Kleinsa¨ugerfauna aus der Oberen Su¨ßwasser-Molasse Bayerns. Mitt. Bayer. Staatssamml. Pala¨ont. Hist. Geol., 15, 91–111. Meller, B. 1989. Eine Blatt-Flora aus den obermioza¨ nen Dinotheriensanden (Vallesium) von Sprendlingen (Rheinhessen). Documenta Naturae, 54, 1–109. Mein, P. 1986. Chronological succession of hominoids in the European Neogene. In Primate Evolution, Else, J. G. & Lee, P. C. (eds.), 59–79. Cambridge, Cambridge University Press. Mein, P. 1989. Updating of MN zones. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), 73–90. New York, Plenum. Meyer, H. von 1838. Mittheilungen, an Professor Bronn gerichtet. N. Jb. Miner. Geognos. Geol. Petrefakt., 1838, 413–18. Morlo, M. 1997. Die Raubtiere (Mammalia, Carnivora) aus dem Turolium von Dorn-Du¨rkheim 1 (Rheinhessen). Teil 1: Mustelida, Hyaenidae, Percrocutidae, Felidae. In Die Sa¨ugetiere aus dem Turolium von Dorn-Du ¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 11–47. Mottl, M. 1957. Bericht u ¨ ber die neuen MenschenaVenfunde aus O ¨ sterreich, von Sankt Steven im Levanttal, Ka¨rnten. Carinthica 2. Mitteilungen des Naturwissenschaftlichen Vereins in Ka¨rnten, 67, 39–84. Pohlig, H. 1895. Paidopithex rhenanus, le singe anthropomorphe du Plioce`ne rhe´nan. Proce`s-Verbaux de la Societe´ Belge de Ge´ologie, 9, 149–51. Rabeder, G. 1989. Die Sa¨ugetiere des Pontien in O ¨ sterreich und Ungarn. Chronostratigraphie und Neostratotypen. Neogen der Westlichen (‘Zentrale’) Paratethys, 8, Pontien, 821–36. Zagreb and Beograd, Jugoslawische Akademie der Wissenschaften und Kuenste/Serbische Akademie der Wissenschaften und Kuenste. Reumer, J. W. F. 1984. Ruscinian and early Pleistocene Soricidae (Insectivora, Mammalia) from Tegelen (The Netherlands) and Hungary. Scripta Geol., 73, 1–173. Ro¨gl, F. & Daxner-Ho¨ck, G. 1996. Later Miocene Paratethys Correlations. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 47–55. New York, Columbia University Press. Ro¨gl, F. & Steininger, F. F. 1984. Neogene Paratethys, Mediterranean and Indo-paciWc Seaways. Implications for the paleobiogeography of marine and terrestrial biotas. In Fossils and Climate, P. Brenchley (ed.), 171–200. New York, John Wiley & Sons. Ro¨gl, F., Zapfe, H., Bernor, R. L., Brzobohaty, R. L., Daxner-Ho¨ck, G. , Draxler, I., Fejfar, O., Gaudant, J., Herrmann, P., Rabeder, G., Schultz, O. & Zetter, R. 1993. Die Primatenfundstelle Go¨tzendorf an der Leitha (Obermioza¨n des Wiener Beckens, Niedero¨sterreich). Jahrbuch der Geologischen Bundesanstalt Wien, 136, 503–26. Roth, C. & Morlo, M. 1997. Die Raubtiere (Mammalia, Carnivora) aus dem Turolium von Dorn-Du ¨ rkheim 1. Teil 2: Ursidae. In Die Sa¨ugetiere aus dem Turolium von Dorn-Du ¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 49–71. Rzebik-Kowalska, B. 1994. Pliocene and Quaternary Insectivora (Mammalia) of

Late Miocene mammals from Central Europe

Poland. Acta Zool. Cracov., 37, 77–136. Schaefer, H. 1966. Die pontische Sa¨ugetierfauna von Charmoille (Jura bernois). I. Einleitung. Rodentia. Verhandlungen der Naturforschenden Gesellschaft Basel, 77 (1), 87–96. Schlosser, M. 1902. Beitra¨ge zur Kenntniss der Sa¨ugethierreste aus den su¨ddeutschen Bohnerzen. Geologische und Pala¨ontologische Abhandlungen, N.F. 5 (3), 115–258. Stefanescu, G. 1995. Dinotherium gigantissimum. Annuaire du Muse´e de Ge´ologie et de Pale´ontologie de Bucarest, Jahrgang 1994, 127–99. Stirton, R. A. 1935. A review of the Tertiary beavers. Univ. California Publ. Geol. Sci., 23, 391–458. Storch, G. 1978. Die turolische Wirbeltierfauna von Dorn-Du¨ rkheim, Rheinhessen (SW-Deutschland). 2. Mammalia: Insectivora. Senckenbergiana Lethaea, 58, 421–49. Storch, G. 1995. The Neogene mammalian faunas of Ertemte and Harr Obo in Inner Mongolia (Nei Mongol), China. 11. Soricidae (Insectivora). Senckenbergiana Lethaea, 75, 221–51. Storch, G. & Qiu Zhuding 1991. Insectivores (Mammalia: Erinaceidae, Soricidae, Talpidae) from the Lufeng Hominoid locality, Late Miocene of China. Geobios, 24, 601–21. Storch, G. & Qiu Zhuding 1996. Miocene/Pliocene insectivores from China and their relationship to European insectivores. Acta Zool. Cracov., 39, 507–12. Swisher, C. C. III 1996. New 40Ar/39Ar Dates and Their Contribution Toward a Revised Chronology for the Late Miocene of Europe and West Asia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 64–77. New York, Columbia University Press. Thenius, E. 1954. Vo ¨ sendorf – ein Lebensbild aus dem Pannon des Wiener Beckens. Mitteilungen der Geologischen Gesellschaft in Wien, 46, 1–109. Thenius, E. 1982. Ein MenschenaVe (Primates, Pongidae) aus dem Pannon (Jung-Mioza¨n) von Niedero¨sterreich. Folia Primatologica, 39, 187–200. Tobien, H. 1980a. Taxonomic Status of some Cenozoic Mammalian Local Faunas from the Mainz Basin. Mainzer Geowissenschaftliche Mitteilungen, 9, 203–35. Tobien, H. 1980b. Die jungtertia¨re Fossilgrabungssta¨tte Ho¨wenegg im Hegau (Su ¨ dwestdeutschland). Ein Statusbericht. Carolinea, 44, 9–34. Tobien, H. 1983. Bemerkungen zur Taphonomie der spa¨ttertia¨ren Sa¨ugerfauna aus den Dinotheriensanden Rheinhessens (Bundesrepublik Deutschland). Weltenburger Akademie (Erwin Rutte – Festschrift), 191–200. Tobien, H. 1988. Contributions a` l’e´tude du gisement mioce`ne supe´rieur de Montredon (Herault). Les grands mammife`res. 7- Les proboscidiens Deinotheriidae. Palaeovertebrata Montpellier, Me´moire extraordinaire, 1988, 135–75. Van der Made, J. 1997. The fossil pig from the Late Miocene of Dorn-Du¨rkheim 1 in Germany. In Die Sa¨ugetiere aus dem Turolium von Dorn-Du ¨ rkheim 1 (Rheinhessen, Deutschland), J. L. Franzen (ed.). Courier Forschungs-Institut Senckenberg, 197, 205–30. Weerd, A. van de 1976. Rodent faunas of the Mio-Pliocene continental sediments of the Teruel-Alfambra region, Spain. Utrecht Micropaleont. Bull. Spec. Publ., 2, 1–217.

189

Miocene mammalian successions

190

Werdelin, L. 1996. Carnivores, Exclusive of Hyaenidae, from the Later Miocene of Europe and Western Asia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 271–89. New York, Columbia University Press. Werner, J. 1994. Beitra¨ge zur Biostratigraphie der Unteren Su ¨ ßwasser-Molasse Su¨ddeutschlands – Rodentia und Lagomorpha (Mammalia) aus den Fundstellen der Ulmer Gegend. Stuttgarter Beitr. Naturk., Ser. B, 200, 1–263. Woodburne, M. O., Theobald, G., Bernor, R. L., Swisher, C. C. III, Ko¨nig, H. & Tobien, H. 1996. Advances in the Geology and Stratigraphy at Ho¨wenegg, Southwestern Germany. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), 106–23. New York, Columbia University Press. Zapfe, H. 1989. Pongidenza¨hne (Primates) aus dem Pontien von Go¨tzendorf, Niedero¨sterreich. Sitzungsberichte der O ¨ sterreichischen Akademie der Wissenschaften, 197, 423–50.

10 An overview on the Italian Miocene land mammal faunas Lorenzo Rook, Laura Abbazzi and Burkart Engesser

Introduction There is a small number of Miocene localities with land mammals in Italy (Fig. 10.1). Almost all the known Italian Miocene land mammal faunas display endemic characteristics, testifying to a complex paleogeographic history for the ‘central’ Mediterranean during the Late Miocene. The record of land vertebrates of Early and Middle Miocene age on the Italian mainland is made by a few isolated Wndings in marine sediments. A faunule of small mammals was collected in the 1970s near Oschiri (Northeastern Sardinia). The assemblage includes a soricid: Crocidosorex antiquus (Pomel, 1853); two talpids: Geotrypus oschiriensis Ru¨mke 1974, and Nuragha schreuderae Ru ¨ mke 1974; three glirids: Myomimus sp., Microdryomys aV. koenigswaldi de Bruijn 1966, and Glis major de Bruijn 1974; and three ctenodactylids: Sardomys dawsonae de Bruijn 1974, Sardomys antoniettae de Bruijn 1974 and Pireddamys rayi de Bruijn 1974. In addition some amphibians, reptiles and terrestrial gastropods were found (De Bruijn & Ru ¨mke, 1974; Esu & Kotsakis, 1985). The fauna is oligotypic, rather unbalanced and with gigantism in rodents, so bearing characters of insular endemisms. The fauna has European aYnities but includes ctenodactylids, representing African immigrants (Jaeger, 1977; Wang, 1994). Although the dating of this fauna was diYcult, the age is interpreted as Middle or Late Agenian (MN 1 or MN 2). Isolated Wndings in marine sediments, mainly known from old literature, testify to the occurrence of land mammals from the continental areas surrounding the Early and Middle Miocene Tethys in the ‘Mediterranean’ area (Kotsakis et al., 1997). Among these Wndings, of particular interest is a third lower molar of a small-sized mastodon found near the village of Burgio (Agrigento, Sicily), in Lower Burdigalian coastal calcarenites (Checchia Rispoli, 1814). It is about of the same size of the smallest specimens of Gomphotherium from Gebel Zelten (Libya) and its occurrence, as well as the nature of embedding sediments, suggests that the Agrigento area was part of the North Africa shelf during the Early Miocene (Rook et al., 1995). Orleanian mammal age is not represented in Italy. However, the Vallesian and the Turolian mammal ages are documented. The pre-Messinian Late Miocene land mammals assemblages still testify to a rather complex paleogeography, being represented by two diVerent paleobioprovinces, one in the Perityrrhenian area (the Tusco-Sardinia paleobioprovince), the other on the Adriatic side of the Apennines (the Abruzzi-Apulia paleobioprovince).

Miocene mammalian successions

192

[Figure 10.1] Location map of the mammal bearing sites mentioned in the text.

The Abruzzi-Apulia paleobioprovince Fossils of a mammal fauna have been recovered from the Lower Tortonian marine calcarenites of Scontrone, in the Abruzzi National Park (Fig. 10.1). The identiWed taxa include crocodiles, chelonians, a lutrine carnivore, and some artiodactyls, among which are the highly endemic Hoplitomeryx and a larger-sized Amphimoschus-like cervid (Rustioni et al., 1992; Mazza & Rustioni, 1996). The Scontrone fauna is distinctly endemic, and is indicative of an isolated, long-lasting, Early Miocene (Late Oligocene?) to Late Miocene Abruzzo-Apulia paleobioprovince that had absolutely no relationship with any other contemporaneous Italian paleobioprovince. On the basis of stratigraphical and geological studies, the Scontrone fossil bearing beds are referable to the Lower Tortonian so that they were correlated with the latest Vallesian–earliest Turolian (MN 10–MN 11) by Rustioni et al. (1992) and Mazza & Rustioni (1996). According to Steininger et al. (1996) and Agustı´ (this volume) however, the Lower Tortonian would be better correlated with early Vallesian.

An overview on the Italian Miocene

Several elements of the Scontrone faunal assemblage also occur in the richer endemic fauna from the terre rosse karst Wssure Wllings of the Gargano promontory (Fig. 10.1), the so called ‘Microtia fauna’. The ‘Microtia fauna’ occurs over a vast area of the Gargano pedemountain belt (Freudenthal, 1976). The mammals display highly developed endemic characteristics, and the fossil record reveals evolutionary radiations that indicate the emergent areas formed an archipelago, with only sporadic communications with the mainland. Several lines of systematic and paleogeographic evidence show that the Gargano populations were enriched by several migratory events from the continent (De Giuli et al., 1985, 1986, 1990). The examination of the evolutionary trends displayed by many of the endemic taxa has revealed that they underwent important modiWcations, and this has allowed the identiWcation of at least four consecutive biochronologic phases characterised by diVering assemblages and diVering degrees of taxon evolution (De Giuli et al., 1986, 1987, 1990; Abbazzi et al., 1993).The problems concerning the chronostratigraphy and biochronology of the Neogene deposits of the Gargano region and its vertebrate faunas have recently been thoroughly discussed (Abbazzi et al., 1996) and we refer to this study for an exhaustive discussion. The Gargano Microtia faunas represent Wnally the populations of the last exposed lands of the larger Abruzzi-Apulia paleobioprovince. This paleobioprovince was partially disrupted in the Messinian when part of the Abruzzi area became submerged, but survived in the Gargano area until the Early Pliocene.

The Tusco-Sardinian paleobioprovince The Tusco-Sardinian paleobioprovince is documented by the faunas from Southern Tuscany (Maremma) and Northern Sardinia (Fiume Santo) (Fig. 10.1). The Tusco-Sardinian faunal assemblage includes mainly endemic taxa. The faunas of this paleobioprovince are well known by vertebrate paleontologists and paleoanthropologists for the occurrence of the peculiar endemic hominoid Oreopithecus bambolii, which was Wrst described by Gervais in 1872 and whose phylogenetic relationships have been debated since (Hu ¨ rzeler, 1949, 1960, 1968; Harrison & Rook, 1997; Moya`-Sola` & Ko¨hler, 1996). Fossil vertebrates of Late Miocene age have been recovered from lignite mines at some localities in the Maremma region of Southern Tuscany since the middle of the nineteenth century. Baccinello, Casteani, Monte Bamboli, Montemassi, Ribolla ( = Acquanera) and Serrazzano were all exploited for

193

Miocene mammalian successions

194

lignite excavation (Weithofer, 1888; Hu ¨rzeler & Engesser, 1976; Rook & Cioppi, 1997). In addition, a faunal association with Oreopithecus bambolii and other ‘Maremman’ taxa was recently collected in Miocene sediments at Fiume Santo, in Northern Sardinia (Cordy & Ginesu, 1994; Cordy et al., 1996). The greatest number and richest faunas of the Tusco-Sardinian province are found in the Neogene basin of Baccinello. Detailed geological work was conducted in the Baccinello area by Lorenz (1968) and Benvenuti et al. (1995; 1998; this volume), and more general regional studies have been carried out by Bossio et al. (1978), Damiani et al. (1980) and Pasquare´ et al. (1983). The basin records a succession of Wve mammal associations, ranging from the Early Turolian (MN 11) to the Early Villanyan (MN 16) (Hu ¨ rzeler & Engesser, 1976). Four of these assemblages occur in the Miocene continental succession, and are known as V0, V1, V2 and V3 respectively. The oldest three of these faunas are mainly endemic. The V0 assemblage (the oldest in the basin) includes a murid, Huerzelerimys vireti, which permits a tentative correlation with European sites assigned to the MN 11 unit of the early Turolian land mammal age. The most recent faunal assemblage, V3, is comparable with typical European latest Turolian–earliest Ruscinian faunas (MN 13–MN 14; Hu ¨ rzeler & Engesser, 1976; Engesser, 1989). The V1 faunal assemblage occurs in a lignite layer, and is equivalent to the faunas obtained from the coal mines at Casteani, Ribolla ( = Acquanera), Montemassi, and Serrazzano (Hu¨rzeler & Engesser, 1976; Azzaroli et al., 1986; Rook & Cioppi, 1997) (Table 10.1). The high level of endemism of the assemblage, in conjunction with its relatively low taxonomic diversity, the predominance of specialised bovines, the tendency of some of the rodents to develop hypsodonty and large body size, the absence of non-lutrine carnivores and the peculiar adaptation of the hominoid Oreopithecus are all indicative of an insular environment (Hu¨rzeler & Engesser, 1976; Engesser, 1989; Moya`-Sola` & Ko¨hler, 1997; 1999). The V2 assemblage, in the Baccinello basin, occurs in Xuvial sediments located about 100 m above the V1 lignite. The fauna, like the V1 fauna, still exhibits a high level of endemism, which has made biochronologic correlation with other European sites extremely diYcult. However, the V1 and V2 assemblages can be assigned to the middle or late Turolian (MN 12–MN 13; cfr. Rook et al., 1996). The V2 fauna represents an insular community that was similar to the earlier V1 fauna, but whose composition, when examined in detail, is quite diVerent. The two assemblages share relatively few species – Tyrrhenotragus gracillimus, Anthracoglis marinoi, Paludotona etruria, Oreopithecus bambolii, unidentiWed species of soricid, and several species

An overview on the Italian Miocene

Table 10.1. Faunal composition of diVerent localities from the Tusco-Sardinian paleobioprovince Baccinello V0 Tyrrhenotragus sp. Bovidae indet. Huerzelerimys vireti Anthracoglis marinoi Paludotona cf. etruria Soricidae indet. Chiroptera indet. Baccinello V1

Casteani

Oreopithecus bambolii Tyrrhenotragus aff. gracillimus Maremmia haupti Etruria viallii nomen nudum Bovidae indet.

Oreopithecus bambolii

Ribolla ( = Acquanera), Montemassi, Serrazzano Oreopithecus bambolii

Maremmia haupti

Maremmia haupti

Umbrotherium azzarolii nomen nudum Small anthracoterhid Huerzelerimys oreopitheci Parapodemus sp. I Kowalskia sp. Anthracoglis marinoi Gliridae gen. et sp. nov. Paludotona etruria Soricidae cf. Crocidosorex Tyrrhenolutra helbingi Baccinello V2 Oreopithecus bambolii Tyrrhenotragus gracillimus Maremmia lorenzi Umbrotherium azzarolii nomen nudum Bovidae indet. (different forms) Eumaiochoerus etruscus Paludolutra campanii

Anthracomys majori Parapodemus sp. II Anthracoglis cf. marinoi Paludotona aff. etruria Soricidae indet.

Tyrrhenolutra helbingi Montebamboli Oreopithecus bambolii Tyrrhenotragus gracillimus Maremmia lorenzi

Eumaiochoerus etruscus Paludolutra campanii ?Paludolutra maremmana Indarctos laurillardi Mustela majori Anthracomys majori Anthtracoglis sp.

Fiume Santo Oreopithecus cf. bambolii Tyrrhenotragus gracillimus Maremmia lorenzi Giraffidae indet. (two forms) Bovidae indet. (two forms) Suidae indet.

cf. Indarctos anthracitis cf. Huerzelerimys turoliensis Gliridae indet.

Derived from Weithofer, 1888; Hu¨rzeler & Engesser, 1976; Cordy & Ginesu, 1994; Cordy et al., 1996; Rook et al., 1996; Rook & Cioppi, 1997.

Miocene mammalian successions

whose identiWcations are at present tentative. The key diVerence between the V1 and V2 faunas is that the latter is marked by the arrival of new immigrants (i.e., Parapodemus sp. II, Eumaiochoerus etruscus, and possibly also ‘Indarctos’ laurillardi and ‘Mustela’ majori), as well as the appearance of several new species as a result of in situ evolutionary transformation of the endemic forms (i.e., Anthracomys majori from Huerzelerimys oreopitheci, Paludolutra campanii from Tyrrhenolutra helbingi and Maremmia lorenzi from Maremmia haupti). On the basis of the faunal composition also the Sardinian assemblage from Fiume Santo (Cordy & Ginesu, 1994; Cordy et al., 1996) can be assigned to the same ‘biochron’ as Baccinello V2 and Montebamboli (Table 10.1). The Tusco-Sardinian paleobioprovince disappeared during the Messinian, and the endemic fauna was replaced by a completely diVerent vertebrate assemblage, the V3 fauna, which is composed of taxa common to latest Turolian–earliest Ruscinian European localities. A latest Miocene age for the V3 assemblage is strengthened given the constraint furnished by the geological history of the Baccinello-Cinigiano basin. The entire sedimentary succession containing the V0–V3 faunas was in fact deformed by an intra-Messinian tectonic phase (Bernini et al., 1992; Benvenuti et al., 1995) and was unconformably overlain by a capping layer of marine clays and sands of Early Pliocene age (Sphaeroidinellopsis Zone; Bossio et al., 1991).

The definition of the present day Italian physiography and biogeography During the Messinian, as a consequence of the intense tectonism that aVected the emerging Apennines chain, the Late Miocene paleogeographic scenario drastically and rapidly changed, and the present day physiography of the Italian peninsula came to be established (Boccaletti et al., 1990). This change is evidenced by land mammal localities Messinian and Pliocene in age (known all along the Apennine chain and in Sicily) bearing taxa completely comparable with contemporaneous faunas from the European continent. In addition to the above mentioned Baccinello V3, MN 13 faunal assemblages are found in Ciabot Cagna (Asti, Piedmont), Monticino gypsum quarry (Faenza, Emilia-Romagna), the Casino basin and the Velona basin (Siena, Tuscany) and Gravitelli (Messina, Sicily) (Cavallo et al., 1993; De Giuli et al., 1988; Rook, 1992; Rook & Ghetti, 1997). Although not numerous, these localities yielded Messinian continental fossils mammals clearly tes-

An overview on the Italian Miocene

tifying to the establishment of connections with the European mainland (Tables 10.2, 10.3).

Concluding remarks The Italian mammal faunas from the pre-Messinian Late Miocene reveal the existence of distinct bioprovinces that were isolated since the Early Miocene. In the past it was generally accepted that most of the large mammals (including Oreopithecus) of the Tusco-Sardinian faunal province were of African origin (Hu ¨rzeler & Engesser, 1976; Hu ¨rzeler, 1983; Engesser, 1989). Today the zoogeographical aYnities of the Tusco-Sardinian faunal province indicate that this area mainly had connections with continental Europe during Late Miocene. The faunal aYnities of the Abruzzi-Apulia paleobioprovince are completely diVerent. Here the artiodactyls, which seem to be holdovers of very primitive ruminant stocks that possibly migrated from Asia through eastern Europe, are particularly signiWcant. The data from Scontrone and Gargano indicate that the Late Oligocene–Early Miocene was the most probable time of migration for the Scontrone large mammal fauna, while the micromammals may have arrived in two or three subsequent migratory waves that occurred between the Middle to Late Miocene and the Messinian. By the Messinian the Apennine chain had almost reached its present day setting and the Tusco-Sardinian paleobioprovince was joined to peninsular Italy. At the same time the Abruzzi-Apulia paleobioprovince was disrupted, part of the Abruzzi became submerged and the faunal assemblages of the paleobioprovince survived in restricted areas (Gargano) until the earliest Pliocene.

Acknowledgments We are grateful to the organizers of the workshop ‘The Vallesian’ and to the European Science Foundation for oVering us the opportunity to attend the Network workshops. This work is in part supported by Italian MURST grants; L.R. acknowledges support from L.S.B. Leakey Foundation for the Weld work in southern Tuscany.

197

Mesopithecus sp. indet.

Mesopithecus sp. indet

Metailurus parvulus

Viverridae indet. Thalassyctis hyaenoides

Microstonix major ery. Hexaprotodon suculus ?Parabos sp. Bovidae indet I Bovidae indet II

Zygolophodon borsonii Zygolophodon turicensis Dicerorhinus sp. Diceros cf. pachygnatus

GraviKohfidischtelli

Bovidae indet. I Bovidae indet. II Procapreolus sp. Paracervulus sp. Moschus sp. Cervidae indet. Machair. gr. giganteus Metailurus major Plesiogulo crassa Viverra sp. Hyaenidae indet.

Tapirus cf. arvernensis Hipparion sp. I Hipparion sp. II Korynoch. provincialis

Dicer. cf. megarhinus

Baccinello V3

Thalas. cf. hipparionum Eucyon sp. Mesopithecus pentelicus Prolagus sp.

(?)Euprox elsanus

Korynoch. provincialis Hexaprotodon pantanelli Parabos sp.

Tapirus arvernensis Hipparion sp.

Casino

Prolagus sp.

Cervidae indet

Parabos sp. Bovidae indet.

Korynoch. provincialis

Velona

Prolagus michauxi

Ciabot Cagna

Table 10.2. Faunal composition of Italian Messinian localities with continental vertebrate faunas (for Monticino quarry, see Table 10.3)

Erinaceide gen. sp. nov.

Castor cf. praefiber Muscardinus aff. vireti Kowalskia nestori Celadensia grossetana

Apodemus etruscus

aff. Hypolagus Alilepus sp. Hystrix primigenia Anthracomys lorenzi

Dipoides problematicus Dipoides problematicus

Muridae indet. ?Paraethomys sp.

Galerix cf. socialis

Gerbillidae indet.

Muscardinus sp.

Paraethomys cf. anomalus Castillomys sp.

Miocene mammalian successions

200

Table 10.3. Composition of the Monticino gypsum quarry faunal assemblage cfr. Gomphotheriidae Dicerorhinus cf. megarhinus Hipparion sp. Korynochoerus provincialis cf. Parabos Bovidae indet. Samotragus occidentalis Ruminantia (very small size) Cervidae indet. Eucyon monticinensis Felis ex gr. attica-christoli Plioviverrops faventinus Thalassyctis gr. chaeretis-macrostoma Mellivora benfieldi Mesopithecus pentelicus Orycteropus cf. gaudryi Trischizolagus cf. maritzae Prolagus cf. sorbinii Hystrix primigenia Stephanomys debruijni Paraethomys anomalus Centralomys benericettii Apodemus cf. gudrunae Cricetus cf. barrierei Ruscinomys cf. lasallei Atlantoxerus cf. rhodius Hylopetes sp. Muscardinus sp. Myomimus sp. Galerix sp. aff. depereti Postpalerinaceus sp. Episuriculus aff. gibberodon Soricidae indet. (small size) Megaderma cf. mediterraneum Rhinolophus cf. Kowalskii Rhinolophus sp. Hypposioderos (Syndemostis) cf. vetus Asellia cf. mariatheresae Myotis cf. boyeri

An overview on the Italian Miocene

References Abbazzi, L., Masini, F. & Torre D. 1993. Evolutionary patterns in the Wrst lower molar of the endemic murid Microtia. Quaternary International., 19, 63–70. Abbazzi, L., Benvenuti, M., Boschian, G., Dominici, S., Masini, F., Mezzabotta C., Piccini, L., Rook, L., Valleri, G. & Torre, D. 1996. Revision of the Neogene and Pleistocene of the Gargano region (Apulia, Italy): The marine and continental successions, and the mammal faunal assemblages of an area between Apricena and Poggio Imperiale (Foggia). Memorie della Societa` Geologica Italiana, 51, 383–402. Azzaroli, A., Boccaletti, M., Delson, E., Moratti, G. & Torre D. 1986. Chronological and paleogeographical background to the study of Oreopithecus bambolii. Journal of Human Evolution, 15, 533–40. Benvenuti, M., Bertini, A. & Rook, L. 1995. Facies analysis, vertebrate paleontology and palynology in the Late Miocene Baccinello-Cinigiano basin (Southern Tuscany). Memorie della Societa` Geologica Italiana, 48(1994), 415–23. Benvenuti, M., Papini, M. & Rook L. 1998. Geological map of the eastern Margin of the Baccinello-Cinigiano Basin (Southern Tuscany). ARCA, Digital Cartography, Firenze. Bernini, M., Boccaletti, M., Moratti, G., Papani, G., Sani, F. & Torelli, L. 1992. Episodi compressivi neogenico-quaternari nell’area estensionale tirrenica nord-orientale. Dati in mare e a terra. Memorie della Societa` Geologica Italiana, 45(1990), 577–89. Boccaletti, M., CiaranW, N., Cosentino, D., Deiana, G., Gelati, R., Lentini, F., Massari, F., Moratti, G., Pescatore, T., Ricci Lucchi, F. & Tortorici, L. 1990. Palinspastic restoration and Paleogeographic reconstruction of the peri-Tyrrhenian area during the Neogene. Palaeogeography, Palaeoclimatology, Palaeoecology, 77, 41–50. Bossio, A., Esteban, M., Giannelli, L., Longinelli, A., Mazzanti, R., Mazzei, R., Ricci Lucchi, F. & Salvatorini, G. 1978. Some Aspects of the Upper Miocene in Tuscany. Messinian Seminar n°4, Roma October 9–14 1978, 88 pp., Roma. Bossio, A., Costantini, A., Foresi, L., Mazzei, R., Monteforti, B., Salvatorini, G. & Sandrelli, F. 1991. Notizie preliminari sul Pliocene del bacino del medio Ombrone e della zona di Roccastrada. Atti della Societa` Toscana di Scienze Naturali, Memorie, Serie A, 98, 259–69. Cavallo, O., Sen, S., Rage, J. C. & Gaudant, J. 1993. Verte´bre´s messiniens du Facie`s a conge´ries de Ciabo ` t Cagna, Corneliano d’Alba (Pie´mont, Italie). Rivista Piemontese di Storia Naturale, 14, 3–22. Checcia Rispoli, G. 1914. Sul Mastodon angustidens, Cuvier dei dintorni di Burgio in provincia di Girgenti. Giornale di Scienze Naturali ed Economiche, 30, 285–96. Cordy, J. M. & Ginesu, S. 1994. Fiume Santo (Sassari, Sardeigne, Italie): un nouveau gisement a` Ore´opithe`que (Orepithecidae, Primates, Mammalia). Comptes Rendues de l’Academie des Sciences de Paris, ser. II, 318, 697–704. Cordy, J. M., Ginesu, S., Ozer, A. & Sias, S. 1996. Geomorphological and paleogeographical characteristics of the Oreopithecus site of Fiume Santo (Sassari, northern Sardinia, Italy). GeograWa Fisica e Dinamica Quaternaria, 18, 7–16.

201

Miocene mammalian successions

202

Damiani, A.V., Gandin, A. & Pannuzi, L. 1980. Il bacino neogenico dell’Orcia-Ombrone nel quadro dell’evoluzione paleogeograWca e tettonica della Toscana meridionale. Memorie della Societa` Geologica Italiana, 21, 281–7. De Bruijn, H. & Rumke, M. C. G. 1974. On a peculiar mammalian association from the miocene of Oschiri (Sardinia). I and II. Proceedings. Koninklijke Nederlandse Akademie van Wetenschappen, s.B, 77, 44–79. De Giuli, C., Masini, F. & Torre, D. 1985. EVetto arcipelago: un esempio nelle faune fossili del Gargano. Bollettino della Societa` Paleontologica Italiana, 24, 191–3. De Giuli, C., Masini, F. & Torre, D. 1988. The mammal fauna of the Monticino quarry. In Guide Book of the Workshop ‘Continental Faunas at the Mio-Pliocene Boundary’, De Giuli, C. & Vai, G. B. (eds.), pp. 65–9. Faenza. De Giuli, C., Masini, F., Torre, D. & Valleri, G. 1986. Mammal migrations events in the emerged areas of the Apulian platform during the Neogene. Giornale di Geologia, serie 3, 48(1/2), 145–62. De Giuli, C., Masini, F. & Torre, D. 1990. Island endemism in the eastern Mediterranean mammalian paleofaunas: radiation patterns in the Gargano palaeo-archipelago. In Biogeographic Aspects of Insularity. Atti dei Convegni Lincei, 85, 247–62. De Giuli, C., Masini, F. & Valleri, G. 1987. Paleogeographic evolution of the Adriatic area since Oligocene to Pleistocene. Rivista Italiana di Paleontologia e StratigraWa, 93, 109–26. Engesser, B. 1989. The Late Tertiary small mammals of the Maremma region (Tuscany, Italy). II Part: Muridae and Cricetidae (Rodentia, Mammalia). Bollettino della Societa` Paleontologica Italiana, 29, 227–52. Esu, D. & Kotsakis, T. 1985. Les vertebres et les mollusques continentaux du Tertiaire de la Sardeigne: paleobiogeographie et biostratigraphie. Geologica. Romana, 22(1983), 177–206. Freudenthal, M. 1976. Rodent stratigraphy of some Miocene Wssure Wllings in Gargano (Prov. Foggia, Italy). Scripta Geologica, 37, 1–23. Gervais, P. 1872. Sur un singe fossile, d’espe`ce non encore de´crite, qui a e´te´ de´couverte au Monte Bamboli. Comptes Hebdomadaires Se´ances Academie des Sciences de Paris, 74, 1217–23. Harrison, T. & Rook, L. 1997. Enigmatic anthropoid or misunderstood ape: the phylogenetic status of Oreopithecus bambolii reconsidered. In Function, Phylogeny and Fossils: Miocene Hominoid Origins and Adaptations, Begun, D. R., Ward, C. W. & Rose, M. D. (eds.), pp. 327–62. Plenum Press, New York. Hu ¨ rzeler, J. 1949. Neubeschreibung von Oreopithecus bambolii Gervais. Schweitzerische Pala¨ontologishe Abhandlungen, 66, 1–20. Hu ¨ rzeler, J. 1960. Der Bedentung des Oreopithecus fu ¨ r die Stammesgeschichte des Menschen. Triangel, 5, 163–74. Hu ¨ rzeler, J. 1968. Questions et reXe´xions sur l’histoire des Anthropomorphes. Annales de Pale´ ontologie (Vertebre´s), 54(2), 195–233. Hu ¨ rzeler, J. 1983. Un alce´laphide´ aberrant (Bivide´, Mammalia) des ‘Lignites de Grosseto’ en Toscane. C. R. Acad. Sci. Paris, se´r. 2, 295, 697–701. Hu ¨ rzeler, J. & Engesser, B. 1976. Les faunes de mammife`res ne´oge`nes du Bassin de Baccinello (Grosseto, Italie). Comptes Rendues de l’Academie des Sciences de Paris, ser. II, 283, 333–6.

An overview on the Italian Miocene

Jaeger, J. J. 1977. Les rongeurs du Mioce`ne moyen et supe´rieur du Maghreb. Paleovertebrata, 8, 3–186. Kotsakis, T., Barisone, G. & Rook, L. 1997. Mammalian biochronology in an insular domain: the Italian Tertiary faunas. Me´m. Trav. E.P.H.E. Inst. Montpellier, 21, 431–41. Lorenz, H. G., 1968. Stratigraphisches und mikropala¨ontologisches Untersuchungen des Braunkohlengebietes von Baccinello (Grosseto, Italien). Rivista Italiana di Paleontologia e StratigraWa, 74, 147–270. Mazza, P. & Rustioni, M. 1996. The Turolian fossil artiodactyls from Scontrone (Abruzzi, Central Italy). Bollettino della Societa` Paleontologica Italiana, 35, 93–106. Moya`-Sola`, S. & Ko¨hler, M. 1996. The phylogenetic relationships of Oreopithecus bambolii Gervais, 1872. Comptes Rendues de l’Academie des Sciences de Paris, ser. II, 324, 141–8. Moya`-Sola`, S. & Ko¨hler, M. 1997. Ape-like or hominid-like? The positional behaviour of Oreopithecus bambolii reconsidered. Proc. Natl. Acad. Sci. USA, 94, 11747–50. Moya`-Sola`, S., Ko ¨ hler, M. & Rook, L. 1999. Evidence of hominid-like precision grip capabilities in the hand of the European Miocene ape Oreopithecus. Proc. Natl. Acad. Sci. USA, 96, 313–17. Pasquare´, G., Chiesa, S., Vezzoli, L. & Zanchi, A. 1983. Evoluzione paleogeograWca e strutturale di parte della Toscana meridionale a partire dal Miocene superiore. Memorie della Societa` Geologica Italiana, 25, 145–57. Rook, L. 1992. Italian Messinian localities with vertebrate faunas. Paleontologia y Evolucio´, 24–25, 141–7. Rook, L. & Cioppi, E. 1997. A small fauna with Oreopithecus from the old lignite mine of Serrazzano in Val di Cecina (Late Miocene, Italy). Folia Primatologica, 68, 362–5. Rook, L. & Ghetti, P. 1997. Il bacino neogenico della Velona (Toscana, Itallia): stratigraWa e primi ritrovamenti di vertebrati fossili. Bollettino della Societa` Geologica Italiana, 116(2), 335–46. Rook, L., Harrison, T. & Engesser, B. 1996. The taxonomic status and biochronological implications of new Wnds of Oreopithecus from Baccinello (Tuscany, Italy). Journal of Human Evolution, 39, 3–27. Rook, L., Torre, D., Ficcarelli, G., Kotsakis, T., Masini, F., Mazza, P. & Sirotti, A. 1995. Preliminary observations on the biogeography of the ‘central’ Mediterranean during late Miocene. International Conference on Biotic and Climatic EVects of the Messinian Event on the Circum-Mediterranean, January 14–18 1995. University of Garyounis-Bengazi Libya. Rustioni, M., Mazza, P., Azzaroli, A., Boscagli, G., Cozzini, F., Di Vito, E., Masseti, M. & Pisano, A. 1992. Miocene vertebrate remains from Scontrone, National Park of Abruzzi, central Italy. Rendiconti Lincei: Scienze Fisiche e Naturali, ser. 9, 3, 227–37. Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Miocene and Pliocene) marine-continental chronologic correlations of European Mammal Units. In The Evolution of Western Eurasian Mammal Faunas, Bernor, R. L., Falhbush, V. & Mittman, H.-W. (eds.), pp. 7–46. Columbia University Press, New York.

203

Miocene mammalian successions

204

Wang, B. 1994. The Ctenodactyloidea of Asia. In Rodent and Lagomorph Families of Asian Origins and DiversiWcations, eds. Tomyda, Y. et al., National Sciences Museum Monographs, 8, 35–47, Tokyo. Weithofer, K. A. 1888. Alcune osservazioni sulla fauna delle ligniti di Casteani e di Montebamboli (Toscana). Bollettino del Comitato Geologico Italiano, ser II, 11/12, 361–8.

11 The Miocene large mammal succession in Greece Louis de Bonis and George D. Koufos

Introduction The Miocene Greek large fossil mammal localities are not documented for the beginning of the Miocene (MN 1 and MN 2). The MN 3 zone could be represented by the locality Kalimeriani but this locality could also be dated to MN 4. This latter zone is represented by Aliveri and the limit MN 4–MN 5 by Antonios. The biozone MN 5 is well known through the localities of Thymiana (island of Chios) with small and large mammals. A sole locality (Chrysavgi) has been found from MN 6 to MN 8. The late Miocene (MN 9 to MN 13) is very well known with several localities. The two ages of mammals, Vallesian and Turolian, have clearly diVerent faunas. Actually new excavations seem to be necessary for a better knowledge of early and middle Miocene faunas and to describe more precisely the biostratigraphy of the late Miocene. Some Miocene Greek fossil large mammal localities (continental Greece and Aegean archipelago) have been known for the past century but others have been found recently. They cover a large part of the Miocene but those of the latest Miocene (Turolian) are more abundant and better known. The fossils coming from these localities allow quite good dating of the Greek continental deposits. The aim of the present article is to classify the large fossil mammal localities in a chronological order using the available biochronological data as well as other processes of dating. Faunal lists of these localities are given together with some other information. The Greek Miocene series of fossil mammal localities is incomplete. The biozones MN 1, MN 2, MN 6 and maybe MN 3 have never been found in Greece.

Biozones MN 3–MN 4 Until now, a fossil large mammal locality has never been found in Greece in the Wrst Miocene biozones, MN 1 and MN 2. One mandible of Brachyodus onoideus has been referred from the locality ‘Kalimeriani’ in Evia (Melentis, 1966a). This species has been found elsewhere in MN 3 and MN 4. More material is needed to make the dating more precise.

Miocene mammalian successions

206

Biozone MN 4 The oldest well known large fossil mammal locality is ‘Aliveri’, Evia island. The Aliveri fauna includes a rich micromammalian fauna but only few macromammalian remains. Two carnivorans, Euboictis aliverensis and Palaeogale sp., have been referred from this site (Schmidt-Kittler, 1983) but the dating to the end of MN 4 is founded only on micromammals (de Bruijn et al., 1992).

Biozones MN 4–MN 5 Remains of vertebrates have been found recently in Chalkidiki in the locality of ‘Antonios’ (Koufos & Syrides, 1997). One species of Proboscidian, two species of Pseuaelurus, Bunolistriodon, Sanitherium, Eotragus, Dorcatherium, Palaeomeryx and several micromammals allow us to date this locality to late MN 4–early MN 5.

Biozone MN 5 The only known macromammalian locality from this biozone is that of Chios. The middle Miocene localities of Chios are located on the northeastern part of the island close to the village of Thymiana. The Wrst locality was found in 1940 by Professor E. Paraskevaidis who described the Wrst collected fossil remains and gave some information on the age of the fauna (Paraskevaidis, 1940, 1977). Later on, during the 1960s, a team from the University of Athens and the University of Mainz worked in the same area and collected some material, including micro- and macromammals (Melentis & Tobien, 1967; Tobien, 1968, 1977). The material collected during this expedition includes a complete skull associated with the mandible of Choerolophodon chioticus as well as some isolated teeth of various artiodactyls whose identiWcations are questionable (Tobien, 1980; Lehman & Tobien, 1993–95). The micromammals are not yet described and only a preliminary list has been given (Tobien, 1968). A new Hellenic–French Weld campaign in the Neogene of Chios began in 1991 and was continued in 1993. This campaign allowed recovery of both small and large fossil mammals. The outcrop has also been sampled for paleomagnetic studies. Three fossiliferous horizons were found. The lower one, named ‘Thymiana A’ (THA) and the upper one, named ‘Thymiana C’ (THC), contain rich micromammalian faunas, while the intermediate one,

The Miocene large mammal succession in Greece

named ‘Thymiana B’ (THB), contains macromammals (Kondopoulou et al., 1993) and corresponds to the level worked by Paraskevaidis. The determination of the collected material together with the old faunal data completes our knowledge on the fauna and on the dating of the sites. The age of the Chios mammalian fauna has been discussed after the Wrst dating to late Miocene (Paraskevaidis, 1940). It has been referred to middle–late Miocene (Melentis & Tobien, 1967), to Aragonian–early Sarmatian (Besenecker, 1973), to middle Astaracian (Tobien, 1980), to middle Astaracian MN 7 (Benda & Meulenkamp, 1990) or to the end of MN 5 (Steininger et al., 1990; Mein, 1990; de Bruijn et al., 1992). The newly recovered fossils allow conWrmation of the dating to MN 5. The sanitheres are abundant and have many similarities with the Leoben material (Bonis et al., 1997b). The latter is dated to MN 5 (de Bruijn et al., 1992) and the same age is possible for the Thymiana fauna. The giraYd Georgiomeryx georgalasi belongs to a primitive group and has close aYnities with Zarafa, a giraYd of the Jebel Zelten dated to MN 4, but is more evolved and it seems to indicate a younger age (Bonis et al., 1997a). The bovids are close to some specimens of Bielometcheskaia, dated to the end of MN 6 (de Bruijn et al., 1992) but a little smaller and probably older (Bonis et al., in prep.). The viverrid Lophocyon paraskevaidisi indicates also a middle Miocene age (Koufos et al., 1995). The micromammals determined from both levels (Bonis et al., 1997a; Sen et al., in prep) suggest a younger age than that of Aliveri (MN 4) and they are quite similar to that of Komotini (de Bruijn et al., 1992) whose dating is MN 5. The magnetostratigraphy of a 125 m thick section has provided ten successive polarity zones which are correlated to the intervals of Chrons C5Bn-1n - C5Cr. The fossiliferous levels are included in a long reverse episode into the Chron C5Br corresponding to the time interval 15.2 to 16.0 a. So, taking into account that the macromammalian fauna is the intermediate one, its age could be estimated to 15.5 Ma (Kondopoulou et al., 1993; Bonis et al., 1997a; Sen et al., in prep.).

Biozones MN 6–MN 8 The macromammalian faunas of these zones were almost unknown until now. Recently, a new micromammalian locality named ‘Chrysavgi’ (CHR), found in the Mygdonia basin (Thessaloniki) allowed the dating of old material of macromammals unearthed from this locality. Some isolated teeth of a rhinocerotid were referred to Diceros pachygnathus (Psarianos, 1958) or to Dicerorhinus orientalis (Dimopoulos, 1972). Both authors proposed a late

207

Miocene mammalian successions

208

Miocene age. The micromammalian fauna (Koliadimou, 1996) includes several species which indicate that the locality belongs to the end of the middle Miocene (MN 7–8).

Biozones MN 9–MN 10 These biozones correspond to the Vallesian age of mammals. The locality ‘Kastellios Hill’ is situated in central Crete, 1 km north from the village of Kastelliana. The fauna consists mainly of micromammals found in several distinct levels (de Bruijn et al., 1971; de Bruijn & Zachariasse, 1979). Among the few macromammalian remains there is a large Hipparion. From the illustrations it could belong to H. primigenium, a Vallesian species. Moreover, de Bruijn & Zachariasse (1979) refer to the presence of an artiodactyl indet. in the K3 level while the presence of Dorcatherium sp., Mustelidae indet. and Bovidae indet. is referred from another site in the eastern Xank of the hill (de Bruijn et al., 1971). However, the dating of the locality to Vallesian in founded on the study of the rodents and of the foraminifers. According to de Bruijn et al. (1992), the locality of Kastellios could be older than the Vallesian localities of the Axios valley. Some isolated remains of Dorcatherium found near the village of Melambes (Rhetymnon, central Crete) would indicate a late Miocene age (MN 9–MN 11) for the lower molassic Formation of Crete (Bonneau & Ginsburg, 1974). Most of the Greek Vallesian macromammalian sites are in Macedonia in the basin of the lower Axios valley, in the Nea Messimvria Formation (Bonis et al., 1992b). The ‘Ravin de la Pluie’ (RPL) was found in 1973 near the village of Nea Messimvria, about 25 km northwest from Thessaloniki. It is well known because of the presence of the hominoid primate Ouranopithecus macedoniensis. The fauna (see Table 11.1) is quite rich and allows a good appreciaton of the geological age although it is diYcult to compare some of the taxa with fossils of central and western Europe. The large Hipparion primigenium is characteristic of Vallesian localities; the RPL form is more evolved than the typical form of Eppelsheim (MN 9) indicating a younger age, MN 10 (Koufos, 1986, 1990). In the early Vallesian localities (Eppelsheim, Nombrevilla, Can Llobateres) only one species of Hipparion is known while there are two species in the late Vallesian localities (Masia del Barbo, Montredon). The presence of a second Hipparion in the RPL fauna supports a late Vallesian age (Koufos, 1990). The giraYd Decennatherium? has aYnities with the Spanish Vallesian species. For the bovids, Mesembriacerus is more primitive than the ovibovine from the Turolian and Samo-

The Miocene large mammal succession in Greece

tragus praecursor is also more primitive and smaller than S. crassicornis from Samos (Bouvrain & Bonis, 1984, 1985). The hyaenid Adcrocuta eximia is represented by a sub-species more primitive than the typical Turolian one (Bonis & Koufos, 1981). The rodent Progonomys cathalai also conWrms the dating (Bonis & Melentis, 1975). The locality ‘Ravin des Zouaves 1’ (RZ1) is very close to RPL and corresponds to the same level. Mesembriacerus and Samotragus are present as in RPL and there is also another bovid, Ouzocerus gracilis. The locality ‘Xirochori 1’ (XIR) found in 1989 near the village of Xirochori (XIR), 35 km northwest from Thessaloniki, is situated 2 km from RPL. The fauna contains Ouranopithecus macedoniensis (Bonis et al., 1990a; Bonis & Koufos, 1993) and some mammalian remains which do not allow accurate dating. Nevertheless, taking into account the presence of Ouzocerus sp. and Ouranopithecus as well as the place of this locality in the Nea Messimvria Formation, we can assume a late Vallesian age for XIR with a high probability. The locality ‘Pentalophos 1’ (PNT) is in the vicinity of the village of Pentalophos, about 15 km northwest from Thessaloniki. It lies in the Nea Messimvria Formation but quite far from the other localities. It was discovered in 1985 and several species have been described. This is the most puzzling locality of Macedonia and its dating is not easy because some species are not found elsewhere. The rhinoceroses and mastodonts cannot give us indications on the precise dating. Two species of Hipparion could be the same as in RPL but their study is in progress. The giraYd Decennatherium? macedoniae has similarities with the Vallesian Spanish species D. pachecoi, but it is diVerent. It is smaller and it looks more primitive than Decennatherium? sp. from RPL (Geraads, 1989) and it could indicate an older age for the locality. It shares also some similarities with ‘Samotherium’ pamiri described from the Vallesian of the Sinap Tepe in Turkey. Among the bovids (Bouvrain, 1997), Protoryx is present with a species whose characteristics are intermediate between P. solignaci from the Beglia formation in Tunisia (MN 7–8) and P. laticeps or P. crassicornis from Samos as well as P. carolinae from Pikermi. Ouzoceros is also present but with a species diVerent from O. gracilis found in RZ1, the relationships between both species being unclear. The presence of the large hyaenid Dinocrocuta gigantea also favours a Vallesian age because these large hyaenids which are present in Vallesian sites such as Bou HaniWa are replaced by Adcrocuta in the late Vallesian and the Turolian (Koufos, 1995b). The aardvark found in PNT is Orycteropus pottieri, whose type-species comes from the Vallesian of the Sinap Tepe (Bonis et al., 1994). So, despite some problems, the age of Pentalophos can be estimated as Vallesian and probably a little older than that of Ravin de la Pluie.

209

Zone

MN 3 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 4 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 7–8 MN10 MN10

Sites

Kalimeriani Aliveri Aliveri Antonios Antonios Antonios Antonios Antonios Antonios Antonios Antonios Antonios Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Thymiana B (THB) Chrysavgi Ravin de la Pluie (RPL) Ravin de la Pluie (RPL)

MN 4 MN 4 MN 4 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 5 MN 7–8 MN 10 MN 10

Zone

38.00 38.30 38.30 40.30 40.30 40.30 40.30 40.30 40.30 40.30 40.30 40.30 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 40.30 41.00 41.00

Lat 24.00 24.00 24.00 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 26.00 26.00 26.00 26.00 26.00 26.00 26.00 26.00 26.00 26.00 23.00 23.00 23.00

Long Artiodactyla Carnivora Carnivora Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Carnivora Carnivora Proboscidea Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Proboscidea Proboscidea Carnivora Carnivora Perissodactyla Artiodactyla Artiodactyla

Order Anthracotheriidae Indet. Viverridae Bovidae Tragulidae Tragulidae Suidae Sanitheriidae Palaeomerycidae Felidae Felidae Gomphotheriidae Bovidae Cervidae Tragulidae Giraffidae Suidae Sanitheriidae Gomphotheriidae Deinotheriidae Viverridae indet. Rhinocerotidae Bovidae Bovidae

Family

Table 11.1. List of the known Miocene large mammal localities of Greece

Brachyodus Palaeogale Euboictis Eotragus Dorcatherium Dorcatherium Bunolistriodon Sanitherium Palaeomeryx Pseudailurus Pseudailurus Gomphoterium Tethytragus ? Lagomeryx Dorcatherium Georgiomeryx Listriodon Sanitherium Choerolophodon Deinotherium Lophocyon gen. gen. Mesembriacerus Samotragus

Genus onoideus sp. aliverensis sp. cf. peneckei sp. lockarti slagintweiti cf. kaupi quadridentatus Cf. loreteti sp. cf. koheri sp. sp. georgalasi sp. schlagintweiti chioticus sp. paraskevaidisi sp. sp. melentisi praecursor

Species

sub-species

Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin de la Pluie (RPL) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Ravin des Zouaves 1 (RZ1) Xirochori 1 (XIR)

MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN10 MN 10 MN10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10

MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Carnivora Carnivora Proboscidea Proboscidea Proboscidea Rodentia Rodentia Rodentia Insectivora Primates Artiodactyla Artiodactyla Artiodactyla Carnivora Carnivora Perissodactyla Perissodactyla Proboscidea Artiodactyla

Bovidae Bovidae Giraffidae Giraffidae Giraffidae Giraffidae Equidae Equidae Rhinocerotidae Hyaenidae Hyaenidae Gomphotheriidae Gomphotheriidae Deinotheriidae Muridae Sciuridae Gliridae Erinaceidae Hominidae Bovidae Bovidae Bovidae Hyaenidae Hyaenidae Equidae Equidae Gomphotheriidae Bovidae

Prostrepsiceros ?Palaeoryx Palaeotragus Palaeotragus Helladotherium Bohlinia Hipparion Hipparion indet Adcrocuta Protictitherium Choerolophodon Tetralophodon Deinotherium Progonomys Spermophilinus indet Palerinaceus Ouranopithecus Mesembriacerus Ouzocerus Samotragus Adcrocuta Thalassictis Hipparion Hipparion Choerolophodon Ouzocerus vallesiensis sp. cf. coelophyrs cf. couenii sp. cf. attica primigenium macedonicum sp. eximia leptorhyncha ff. gaillardi pentelici sp. sp. cathalai sp. sp. sp. macedoniensis Melentisi gracilis Praecursor eximia wongi cf. primigenium cf. macedonicum pentelici sp.

Zone

MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10

Sites

Xirochori 1 (XIR) Xirochori 1 (XIR) Xirochori 1 (XIR) Xirochori 1 (XIR) Xirochori 1 (XIR) Xirochori 1 (XIR) Xirochori 1 (XIR) Kastellios Kastellios Kastellios Kastellios Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT)

Table 11.1 (cont.).

MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10

Zone

41.00 41.00 41.00 41.00 41.00 41.00 41.00 35.O 35.O 35.O 35.O 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00

Lat 23.00 23.00 23.00 23.00 23.00 23.00 23.00 25.30 25.30 25.30 25.30 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00

Long Artiodactyla Artiodactyla Perissodactyla Proboscidae Carnivora Carnivora Primates Perissodactyla Artiodactyla Artiodactyla Carnivora Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Carnivora Carnivora Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla

Order

Rhinocerotidae Gomphotheriidae Hyaenidae Hyaenidae Hominidae Equidae Bovidae Tragulidae Mustelidae Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Giraffidae Hyaenidae Hyaenidae Equidae Equidae Rhinocerotidae Rhinocerotidae Chalicotheriidae

Bovidae

Family Samotragus Bovidae indet. Choerolophodon Protictitherium Adcrocuta Ouranopithecus Hipparion indet. Dorcatherium indet. Ouzocerus Helladorcas ? Gazella Boselaphini Protoryx ? Palaeotragus ? Decennatherium Dinocrocuta Proticthitherium Hipparion Hipparion Aceratherium Ceratotherium Ancylotherium

Genus sub-species

praecursor ?Palaeoryx sp. sp. pentelici cf. crassum sp. macedoniensis sp. sp. sp. sp. pentalophosi geraadsi sp. indet. sp. coelophrys macedoniae gigantea cf. crassum cf. primigenium cf. macedonicum kiliasi neumayri sp.

Species

Pentalophos 1 (PNT) Pentalophos 1 (PNT) Pentalophos 1 (PNT) Diavata Diavata Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 1 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Nikiti 2 (NKT) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO)

MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 10 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

MN 10 MN 10 MN 10 MN 10 MN 10 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

41.00 41.00 41.00 41.00 41.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 41.00 41.00 23.00 23.00 23.00 23.00 23.00 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.00 23.00 Proboscidea Proboscidea Tubulidentata Perissodactyla Carnivora Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Carnivora Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Probboscidea Carnivora Artiodactyla Artiodactyla

Gomphotheriidae Gomphotheriidae Orycteropidae Equidae Hyaenidae Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Giraffidae Giraffidae Giraffidae Suidae Equidae Equidae Rhinocerotidae Hyaenidae Hominidae Bovidae Bovidae Bovidae Bovidae Giraffidae Equidae Gomphotheriidae Hyaenidae Bovidae Bovidae

Choerolophodon Tetralophodon Orycteropus Hipparion Dinocrocuta Tragoportax Prostrepsiceros Oioceros ? Gazella indet. Helladotherium Bohlinia Bohlinia Palaeotragus Microstonyx Hipparion Hipparion indet. indet. Ouranopithecus Nisidorcas cf. Tragoportax Oioceros Ouzocerus Helladotherium Hipparion Choerolophodon indet. Palaeoras Tragoportax pentelici sp. pottieri sp. salonicae gaudry houtumschindleri syridesi aff. atropatenes sp. sp. duvernoyi attica nikitiae cf. rouenii major sp. (large) cf. macedonicum sp. sp. macedoniensis planicornis sp. sp. sp. duvernoyi sp. (small) pentelici sp. zouavei rugosifrons

Zone

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

Sites

Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Ravin des Zouaves 5 (RZO) Vathylakkos 1 (VLO) Vathylakkos 1 (VLO) Vathylakkos 1 (VLO)

Table 11.1 (cont.).

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 12 MN 12 MN 12

Zone

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00

Lat 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00

Long Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Carnivora Carnivora Carnivora Carnivora Carnivora Primates Proboscidea Proboscidea Rodentia Artiodactyla Artiodactyla Perissodactyla

Order Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Giraffidae Suidae Suidae Equidae Equidae Equidae Chalicotheriidae Rhinocerotidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Felidae Cercopithecidae Gomphotheriidae Mammutidae Gliridae Bovidae Bovidae Equidae

Family Nisidorcas Prostrepsiceros Prostrepsiceros Gazella Gazella Helladotherium Palaeotragus Microstonyx Propotamochoerus Hipparion Hipparion Hipparion Indet. Ceratotherium Adcrocuta Chasmaporthetes Ictitherium Plioviverrops Machairodus Mesopithecus Choerolophodon Zygolophodon ? Valerimys Palaeoreas Nisidorcas Hipparion

Genus planicornis zitteli rotundicornis pilgrimi sp. duvernoyi rouenii major cf. hysudricus proboscideum dietrichi macedonicum sp. neumayri eximia bonisi viverrinum orbignyi sp. delsoni pentelici turicensis sp. lindermayeri planicornis dietrichi

Species

eximia

erymanthius

sub-species

Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 2 (VTK) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT)

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Carnivora Carnivora Proboscidea Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Perissodactyla

Bovidae Bovidae Bovidae Bovidae Suidae Equidae Equidae Rhinocerotidae Hyaenidae Hyaenidae Gomphotheriidae Cercopithecidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Tragulidae Giraffidae Giraffidae Suidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Felidae Mustelidae Equidae

Tragoportax Gazella Nisidorcas Prostrepsiceros Microstonyx Hipparion Hipparion indet. Hyaenotherium Plioviverrops Choerolophodon Mesopithecus Gazella Nisidorcas Prostrepsiceros Tragoportax Palaeoreas Protoryx Dorcatherium Samotherium Bohlinia Microstonyx Adcrocuta Hyaenotherium Ictitherium Plioviverrops Plioviverrops Machairodus Plesiogulo Hipparion rugosifrons pilgrimi planicornis zitteli major dietrichi macedonicum sp. wongi orbignyi pentelici cf. pentelicus pilgrimi planicornis zitteli rugosifrons lindermayeri sp. puyhauberti boissieri attica major eximia wongi viverrinum orbignyi cf. guerini sp. crassa dietrichi erymanthius

Zone

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 9 MN 9 MN 9 MN 9

Sites

Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Vathylakkos 3 (VAT) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Prochoma (PXM) Pyrgos Vassilisis Pyrgos Vassilisis Pyrgos Vassilisis Pyrgos Vassilisis

Table 11.1 (cont.).

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

Zone

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 38.00 38.00 38.00 38.00

Lat 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 23.00 24.00 24.00 24.00 24.00

Long Perissodactyla Perissodactyla Perissodactyla Perissodactyla Proboscidea Primates Rodentia Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Carnivora Carnivora Proboscidea Artiodactyla Artiodactyla Artiodactyla Artiodactyla

Order Equidae Rhinocerotidae Rhinocerotidae Chalicotheriidae Gomphotheriidae Cercopithecidae Muridae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Suidae Equidae Equidae Chalicotheriidae Viverridae Hyaenidae Gomphotheriidae Bovidae Bovidae Bovidae Giraffidae

Family Hipparion Ceratotherium indet. Macrotherium Choerolophodon Mesopithecus Parapodemus Tragoportax Nisidorcas Prostrepsiceros Gazella Protoryx Palaeoreas Helladotherium Microstonxy Hipparion Hipparion Chalicotherium Plioviverrops Ictitherium Choerolophodon Tragoportax Gazella Gazella Helladotherium

Genus macedonicum neumayri sp. macedonicum pentelici sp. schaubi rugosifrons planicornis zitteli pilgrimi sp. sp. duvernoyi major dietrichi macedonicum goldfussi orbignyi viverrinum pentelici amalthea deperdita ? cf. gaudry ? duvernoyi

Species

erymanthius

sub-species

Pyrgos Vassilisis Pyrgos Vassilisis Pyrgos Vassilisis Pyrgos Vassilisis Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1)

MN 9 MN 9 MN 9 MN 9 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

38.00 38.00 38.00 38.00 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 24.00 24.00 24.00 24.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 Perissodactyla Perissodactyla Proboscidea Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla

Equidae Rhinocerotidae Gomphotheriidae ? Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Cervidae Giraffidae Giraffidae Giraffidae Giraffidae Suidae Suidae Equidae Equidae Equidae Rhinocerotidae Rhinocerotidae Rhinocerotidae

Hipparion Dicerorhinus? Choerolophodon Graecopithecus Prostrepsiceros Oioceros Protoryx Protragelaphus Palaeoryx Pseudotragus Tragoreas Criotherium Tragoprotax Gazella Gazella Sporadotragus Pliocervus Palaeotragus Palaeotragus Samotherium Helladotherium Microstonyx Propotamochoerus Hipparion Hipparion Hipparion Ceratotherium Dicerorhinus Aceratherium mediterraneum orientalis pentelici freybergi zitteli wegneri laticeps sp. pallasi capricornis oryxoides argalioides rugosifrons pilgrimi sp. parvidens pentelici rouenii coelophrys boissieri duvernoyi major cf. hysudricus proboscideum matthewi dietrichi neumayri pikermiensis sp.

Zone

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

Sites

Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Samos (A-1) Tanagra Tanagra Tanagra Tanagra

Table 11.1 (cont.).

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

Zone

37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 37.30 38.3O 38.3O 38.3O 38.3O

Lat 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 27.00 23.30 23.30 23.30 23.30

Long Perissodactyla Perissodactyla Proboscidea Proboscidea Tubulidentata Hyracoidea Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Artiodactyla Artiodactyla Artiodactyla Artiodactyla

Order Rhinocerotidae Chalicotheriidae Gomphotheriidae Deinotheriidae Orycteropodidae Procaviidae Ursidae Ursidae Mustelidae Mustelidae Mustelidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Felidae Felidae Felidae Felidae Bovidae Bovidae Bovidae Bovidae

Family Chilotherium Ancyclotherium Choerolophodon Deinotherium Orycteropus Pliohyrax Indarctos Ursavus Promeles Parataxidea Promephitis Plioviverrops Ictitherium Hyaenotherium Hyaenictitherium Lycyaena Belbus Adcrocuta Felis Metailurus Metailurus Machairodus Gazella Gazella Tragoreas Prostrepsiceros

Genus persiae pentelicum pentelici giganteum gaudryi graecus atticus sp. palaeattica maraghana larteti orbignyi viverrinum wongi hyaenoides chaeretis beaumonti eximia attica parvulus major giganteus deperdita sp. oryxoides aff. houtumschindleri

Species sub-species

Tanagra Tanagra Tanagra Chalkoutsi Chalkoutsi Triada Triada Triada Triada Kerassia Kerassia Alifaka Alifaka Alifaka Alifaka Alifaka Achladi Achladi Achladi Achladi Achladi Rhodes Rhodes Rhodes Rhodes Servia Servia Servia Servia Servia

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

MN 12 MN 12 MN 12 MN 13 MN 13 MN 12 MN 12 MN 12 MN 12 MN 13 MN 13 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 13 MN 13 MN 13 MN 13 MN 13 38.3O 38.3O 38.3O 38.30 38.30 39.O 39.O 39.O 39.O 39.00 39.00 39.30 39.30 39.30 39.30 39.30 39.00 39.00 39.00 39.00 39.00 36.00 36.00 36.00 36.00 40.00 40.00 40.00 40.00 40.00 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 22.30 22.30 22.30 22.30 22.30 23.3O 23.3O 23.3O 23.3O 23.3O 28.00 28.00 28.00 28.00 22.00 22.00 22.00 22.00 22.00 Artiodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Artiodactyla Perissodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Rodentia Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Artiodactyla Perissodactyla Perissodactyla Proboscidea Proboscidea

Bovidae Equidae Rhinocerotidae Equidae Equidae Bovidae Bovidae Suidae Equidae Suidae Equidae Bovidae Bovidae Giraffidae Equidae Hystricidae Bovidae Bovidae Suidae Equidae Equidae Bovidae Bovidae Giraffidae Equidae Suidae Equidae Thinocerotidae Gomphotheriidae Deinotheriidae

indet. Hipparion indet. Hipparion Hipparion Tragoportax Palaeoryx Propotamochoerus ? Hipparion Microstonyx Hipparion Tragoportax ?Palaeoryx Helladotherium Hipparion Hystrix Tragoportax Palaeoryx Propotamochoerus ? Hipparion Hipparion Palaeoryx ?Palaeoryx Helladotherium Hipparion Microsonyx Hipparion indet. Choerolophodon Deinotherium sp. mediterraneum sp. sp. 1 sp. 2 amalthea pallasi sp. mediterraneum major sp. amalthea sp. duvernoyi mediteraneum primigenia amalthea cf. pallasi sp. mediterraneum cf. brachypus pallasi ‘aff. stutzeli’ duvernoyi dietrichi sp. sp. sp. pentelici sp. erymanthius

Zone

MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11 MN 11

Sites

Ano Metochi Ano Metochi Ano Metochi Ano Metochi Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL) Halmyropotamos (HAL)

Table 11.1 (cont.).

MN 13 MN 13 MN 13 MN 13 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

Zone

41.00 41.00 41.00 41.00 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30 38.30

Lat 23.30 23.30 23.30 23.30 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O 24.O

Long Artiodactyla Artiodactyla Artiodactyla Perissodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Proboscidea Proboscidea Carnivora Carnivora Carnivora Carnivora Carnivora Hyracoidea

Order Bovidae Bovidae Giraffidae Equidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Cervidae Suidae Equidae Equidae Rhinocerotidae Chalicotheriidae Deinotheriidae Gomphotheriidae incert. sed. Hyaenidae Felidae Felidae Felidae Procaviidae

Family Prostrepsiceros Gazella Helladotherium Hipparion Protragelaphus Prostrepsiceros Prostrepsiceros Palaeoreas Tragoportax Palaeoryx Gazella Helladotherium Pliocervus Microstonyx Hipparion Hipparion Dicerorhinus Ancylotherium Deinotherium Choerolophodon Simocyon Adcrocuta Metailurus Metailurus Machairodus Pliohyrax

Genus woodwardi sp. cf. duvernoyi sp. skouzesi rotundicornis sp. lindermayeri amalthea pallasi sp. duvernoyi sp. major sp. (large) mediterraneum pikermiensis pentelicum giganteum pentelici primigenius eximia major parvulus giganteus graecus

Species sub-species

Halmyropotamos (HAL) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK)

MN 11 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

38.30 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 24.O 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 Rodentia Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Perissodactyla Proboscidea Proboscidea Proboscidea Proboscidea Hyracoidea Primates Carnivora

Hystricidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Cervidae Tragulidae Giraffidae Giraffidae Giraffidae Suidae Equidae Equidae Rhinocerotidae Rhinocerotidae Rhinocerotidae Chalicotheriidae Gomphotheriidae Gomphotheriidae Mammutidae Deinotheriidae Procaviidae Cercopithecidae Ursidae Hystrix Prostrepsiceros Oioceros Ptotoryx Protragelaphus Palaeoryx Palaeoreas Tragoportax Tragoportax Gazella Sporadotragus Pliocervus Dorcatherium Helladotherium Palaeotragus Bohlinia Microstonyx Hipparion Hipparion Ceratotherium Dicerorhinus Aceratherium Ancylotherium Choerolophodon Tetralophodon Zygolophodon Deinotherium Pliohyrax Mesopithecus Indarctos primigenia rotundicornis rothi carolinae skouzesi pallasi lindermayeri amalthea gaudryi capricornis parvidens pentelici sp. duvernoyi rouenii attica major mediterraneum brachypus neumayri pikermiensis sp. pentelicum pentelici atticus turicensis giganteum graecus pentelicus atticus erymanthus

Zone

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

Sites

Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Pikermi-MNHN(PIK) Chomateres Chomateres Chomateres Chomateres Chomateres

Table 11.1 (cont.).

MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12 MN 12

Zone

38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00 38.00

Lat 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00

Long Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Rodentia Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla

Order inc. sed. Mustelidae Mustelidae Mustelidae Mustelidae Mustelidae Mustelidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Hyaenidae Felidae Felidae Felidae Felidae Felidae Felidae Hysticidae Bovidae Bovidae Cervidae Suidae Equidae

Family Simocyon Sinictis Martes ? Plesiogulo Promeles Promephitis ? Enhydriodon Plioviverrops Ictitherium Hyaenictitherium Hyaenotherium Lycyaena Hyaenictis Adcrocuta Felis Felis Metailurus Metailurus Machairodus Paramachairodus Hystrix Tragoportax Gazella Pliocervus Microstonyx Hipparion

Genus primigenius pentelici woodwardi sp. palaeattica larteti laticeps orbignyi viverrinum hyaenoides wongii chaeretis graeca eximia attica sp. parvulus major giganteus orientalis primigenia gaudryi sp. pentelici major mediterraneum

Species sub-species

Chomateres Chomateres Chomateres Chomateres Chomateres Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 1 (DTK) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT) Dytiko 2 (DIT)

MN 12 MN 12 MN 12 MN 12 MN 12 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13

MN 12 MN 12 MN 12 MN 12 MN 12 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13

38.00 38.00 38.00 38.00 38.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 24.00 24.00 24.00 24.00 24.00 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 Perissodactyla Perissodactyla Proboscidea Primates Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Carnivora Perissodactyla Perissodactyla Perissodactyla Perissodactyla Proboscidea Tubulidentata Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Perissodactyla Perissodactyla

Chalicotheriidae Thinocerotidae Gomphotheriidae Cercopithecidae Cercopithecidae Bovine Bovidae Bovidae Bovidae Bovidae Cervidae Suidae Giraffidae Hyaenidae Equidae Equidae Equidae Rhinocerotidae Gomphotheriidae Orycteropidae Cercopithecidae Bovidae Bovidae Bovidae Bovidae Tragulidae Giraffidae Giraffidae Equidae Equidae

Chalicotherium Aceratherium Choerolophodon Mesopithecus Mesopithecus Gazella Palaeoreas Tragoportax Protragelaphus indet Pliocervus Microstonyx Bohlinia Chasmaporthetes Hipparion Hipparion Hipparion Ceratotherium Choerolophodon Orycteropus Mesopithecus Gazella Palaeoreas Tragoportax indet Dorcatherium Bohlinia Palaeotragus Hipparion Hipparion goldfussi sp. pentelici pentelicus pentelicus deperdita lindermayeri gaudryi theodori sp. pentelicus major attica bonisi mediterraneum matthewi periafricanum neumayri pentelici gaudryi aff. pentelicus deperdita lindermayeri gaudryi sp. puyhauberti attica rouenii mediterraneum matthewi major

pentelicus microdon

Zone

MN 13 MN 13

MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 11 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13

Sites

Dytiko 2 (DIT) Dytiko 2 (DIT)

Dytiko 2 (DIT) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Dytiko (DKO) Dytiko 3 (DKO) Dytiko 3 (DKO) Serres Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena

Table 11.1 (cont.).

MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14

MN 13 MN 13

Zone Long

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.30 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 22.9 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30

41.00 22.9 41.00 22.9

Lat

Primates Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Carnivora Perissodactyla Perissodactyla Perissodactyla Proboscidea Primates Rodentia Rodentia Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla Artiodactyla

Proboscidea Primates

Order

Cercopithecidae Bovidae Bovidae Bovidae Bovidae Bovidae Tragulidae Hyaenidae Equidae Equidae Chalicotheriidae Gomphotheriidae Cercopithecidae Hystricidae Castoridae Bovidae Bovidae Bovidae Bovidae Bovidae Bovidae Giraffidae Cervidae

Gomphotheriidae Cercopithecidae

Family

Mesopithecus Protragelaphus Tragoportax Hispanodorcas indet indet Dorcatherium Protictitherium Hipparion Hipparion Macrotherium Choerolophodon Mesopithecus Hystrix Steneofiber Tragoportax Tragoportax Ouzocerus Norbertia Gazella Boselaphini Samotherium Pliocervus

Choerolophodon Mesopithecus

Genus pentelici cf. monspessulanus ? aff. pentelicus theodori gaudryi orientalis sp. 1 sp. 2 puyhauberti crassum mediterraneum matthewi macedonicum pentelici aff. pentelicus primigenia jaegeri gaudryi cf. amalthea aff. gracilis hellenica sp. indet. cf. boissieri graecus

Species sub-species

Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena Maramena

MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13 MN 13

MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14 MN 14

41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 41.00 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 23.30 Artiodactyla Perissodactyla Perissodactyla Perissodactyla Proboscidea Carnivora Carnivora Carnivora Carnivora Carnivora Carnivora Primates

Suidae Equidae Equidae Rhinocerotidae Gomphotheriidae Mustelidae Mustelidae Mustelidae Mustelidae Viverridae Hyaenidae Cercopithecidae

Korynochoerus Hipparion Hipparion indet. Choerolophodon Martes Promeles Lutra Promephitis indet. Chasmaporthetes Mesopithecus palaeochoerus sp. (large) sp. (small) sp. pentelici lefkonensis macedonicus affinis sp. sp. sp. cf. pentelicus

Miocene mammalian successions

226

A few fossils found close to the village of Diavata, 10 km west from Thessaloniki, have been mentioned by Andrews (1918). The material includes part of the maxilla of a large hyaenid and a maxilla with some bones of a hipparonine horse. The locality, following the information of the collector in a letter accompanying the fossils, is situated in the red beds of the Nea Messimvria Formation where all the known localities are Vallesian (Koufos, 1985). The hyaenid from ‘Diavata’ is large and it has been described as Hyaena salonicae Andrews 1918. It was transferred later to Dinocrocuta (Howell & Petter, 1985) which could characterize the Vallesian. The study of the Hipparion remains indicates that they belong to large-sized species which could be Vallesian/Turolian (Koufos, 1985). So we can conclude that the locality of ‘Diavata’ probably must be attributed to the Vallesian. The locality ‘Nikiti 1’ (NKT) is situated in Chalkidiki peninsula, about 120 km from Thessaloniki. It was found in 1990 and in the intervening years abundant material has been unearthed. The initially determined material and its relationships suggested a Vallesian/Turolian age (Koufos et al., 1991). We can be more precise with the dating of the locality with the species determined today (Table 11.1). The presence of the hominoid Ouranopithecus macedoniensis (Koufos, 1993, 1995a) suggests a Vallesian age. However, the bulk of the fauna is diVerent from that of the known Vallesian ones from Macedonia (Greece). The bovids found in NKT are diVerent than those from RPL. Prostrepsiceros of NKT is completely diVerent than that of RPL, while Tragoportax (absent in RPL) is closer to that from the Spanish Valllesian localities and more primitive than the specimens from Pikermi (Kostopoulos & Koufos, 1996). The hipparions found in NKT belong to two species, one large-sized and the other small-sized. The latter has close similarities with H. macedonicum known from the late Vallesian–early Turolian of the lower Axios valley. The study of the suid found in NKT indicated that it belongs to the early Turolian forms of Microstonyx major (Kostopoulos, in press). Taking into account all the above mentioned data NKT can be dated to the limit late Vallesian–early Turolian (MN 10–MN 11).

Biozones MN 11–MN 13 These biozones correspond to the Turolian. Localities of this age are the most numerous in Greece and some of them have been known for the past century, especially Pikermi with the excavations of Gaudry. There is a signiWcant change between the Vallesian and Turolian faunas, since most of the Vallesian species of bovids disappeared.

The Miocene large mammal succession in Greece

The locality ‘Nikiti 2’ (NIK) is situated near NKT and in a slightly higher stratigraphic level. This stratigraphic position indicates an age not very diVerent from that of NKT. The fauna of NIK (Table 11.1) is still under study and the results are provisional. However, the presence of Nisidorcas planicornis which is known in the early Turolian (MN 11) localities of the lower Axios valley suggests a similar dating. Early Turolian mammalian localities (MN 11 and MN 12) are known in the area of Vathylakkos (= ‘Vatilu ¨ k’), Macedonia, Greece. The Wrst fossiliferous localities were discovered during the First World War by Arambourg and the fossils were published later (Arambourg & Piveteau, 1929). These localities are into the greyish layers of the Vathylakkos Formation which overlies unconformably the reddish sediments of the Nea Messembria Formation, and they are ‘Ravin des Zouaves 5’ (RZO) east of Vathylakkos at the base of the Formation, ‘Vathylakkos 1, 2, 3’ (VLO, VTK and VAT) close to the village and ‘Prochoma 1’, west of Vathylakkos near the village of Prochoma. ‘Vathylakkos 4’ (remains of one individual of Tragoportax) and ‘Vathylakkos 5’ (one metapodial of Hipparion) are not very signiWcant. ‘Vathylakkos 3’ (VAT) is Arambourg ’s locality called ‘ravin de Vatilu ¨ k’ and so we have added into the list some taxa found by Arambourg and stored in the Muse´um national d’Histoire naturelle, Paris. The older level (RZO) is characterized by the presence of the small Hipparion macedonicum, found also in the Vallesian, with H. dietrichi and H. proboscideum, both species existing in the Samos fauna. Two other species could indicate that, in this set, RZO could be at the base: Palaeoreas zouavei, found only in this locality (Bouvrain, 1980), and especially the primate Mesopithecus delsoni. The latter diVers from the other Macedonian specimens of Mesopithecus by its large size and some morphological characteristics which seem to be more primitive (Bonis et al., 1990b). So the locality RZO could be dated to MN 11. Among the bovids, Nisidorcas planicornis seems to be characteristic of this faunal set (Bouvrain, 1979); Gazella pilgrimi is also present in Samos but not in Pikermi (Bouvrain, 1996) as is the case for Tragoportax rugosifrons. Two genera of suid are present. Propotamochoerus cf. hysudricus is similar to a specimen from Samos; Microstonyx major erymanthius is a little smaller than the sub-species of Pikermi and could be considered as a diVerent one, a little older. But on the other hand the rodent Parapodemus schaubi identiWed in the locality of VAT (P. Mein, pers. com.) is characteristic of MN 12. So if RZO is in MN 11, the other localities of the Vathylakkos Formation could be dated to the latest MN 11 or the early MN 12. The island of Samos is one of the oldest known fossiliferous localities in Greece. It contains several fossil vertebrate sites which have been excavated by many people. Fossils from Samos are housed in several institutions

227

Miocene mammalian successions

228

throughout the world but it is almost impossible to know their precise origins. Among the Samos localities we have chosen Quarry 1 and Quarry A which would correspond to a single site in the ‘main bone beds’ (Solounias, 1981). Nevertheless, the presence in the faunal list of 13 species of bovids is quite unusual and probably indicates that this is a composite fauna. This list shares with the Macedonian sites of the Vathylakkos area several species which are absent in Pikermi. There are the giraYd Samotherium boissieri and the bovids Tragoportax rugosifrons and Prostrepsiceros zitteli. The suid Propotamochoerus cf. hysudricus is similar in both areas, while Hipparion proboscideum and H. dietrichi exist also in both areas but H. macedonicum is lacking in Samos (Koufos & Melentis, 1984). On the other hand, the Samos fauna shares two bovids with Pikermi, Palaeoryx pallasi and Sporadotragus parvidens. For other genera the species are diVerent, except for some taxa which do not characterize a single biostratigraphical level (giraYds or most of the carnivores). So it seems, following the listing of the fauna, that these Samos localities are older than Pikermi and younger than the Vathylakkos sites. But we have noted above that this list contains too many bovids and the number would be near 20 if we take into account the whole fauna of the ‘main bone beds’. Radiometric dating and paleomagnetic studies have been made in Samos (Van Couvering & Miller, 1971; Van Couvering, 1972; Weidman et al., 1984; Sen & Valet, 1986; Swisher, 1996) but the problem remains to know where the fossil collections have come from. There are certainly several sites and several fossiliferous levels in Samos. Despite the large amount of bones actually stored in institutions, the most signiWcant work now would be to excavate again in Samos and to note carefully the stratigraphic position of the fossils. Halmyropotamos is situated in the Evia island and it is also an old excavated locality whose material is housed in the laboratory of Geology and Palaeontology of Athens (Melentis, 1967). The fauna does not diVer signiWcantly from that of Pikermi and we can remark on the occurrence of Hipparion mediterraneum. Pikermi, 20 km from Athens, is one of the Wrst fossil mammal localities found in Greece (Gaudry & Lartet, 1856; Gaudry, 1862–7). The area has been excavated by several people and insofar as we do not know the exact locations of these excavations, we preferred to base our list principally on the Gaudry specimens housed in Paris. The fauna diVers from that of Samos by diVerent species of Oioceros (O. rothi in place of O. wegneri), Protoryx (P. carolinae in place of P. laticeps and crassicornis), Tragoportax (T. amalthea and gaudryi in place of T. rugosifrons) and Gazella (G. capricornis in place of G. pilgrimi). Pikermi is the type locality of Microstonyx major erymanthius; and we saw above that the specimens found in the ‘Vathylakkos group’ of

The Miocene large mammal succession in Greece

the lower Axios valley are sligthly smaller (Bonis & Bouvrain, 1996). Hipparion mediterraneum is found with H. brachypus, a species unknown in the MN 11 Greek localities. ‘Chomateres’ is situated near Gaudry’s Pikermi locality. The excavations are in progress and little material has been published (Marinos & Symeonidis, 1974; Symeonidis, 1973; Zapfe, 1991). The age is probably not very diVerent from that of Pikermi. The sites of ‘Dytiko 1, 2 and 3’ (DTK, DIT and DKO) are the younger localities of the late Miocene of the lower Axios valley. The bovids diVer from those of Pikermi by another gazelle, Gazella deperdita whose type specimen comes from the Luberon, France (Bouvrain, 1996), Protragelaphus theodori in place of the less evolved Protragelaphus skouzesi and the absence of Prostrepsiceros and especially Nisidorcas. The suid Microstonyx major major is also similar to the type sub-species of the Luberon (Bonis & Bouvrain, 1996). These localities are also characterized by the appearance of Hipparion periafricanum which is unknown in the Pikermi fauna. The primates are a little confusing. A species close to Mesopithecus pentelicus may be present together with probably another species close to M. monspessulanus. Maybe, in the latest Miocene, both species coexisted contemporaneously before the latter replaced the former in the early Pliocene. The localities of the ‘Dytiko group’ are younger than the localities of the Vathylakkos group and younger than the classical locality of Pikermi, and they could belong to MN 13. The locality of ‘Alifakas’ is situated about 16 km southwest to Larissa, Thessaly, central Greece, and the known material is poor (Melentis & Schneider, 1966). The authors considered that the ‘Alifakas’ fauna can be correlated with the Pikermi fauna and a ‘Pontian’ age is suggested for it. However the material is very fragmentary and the determinations doubtful; e.g. the bovids Tragoportax amlthea and ?Palaeoryx sp. were determined from a maxilla and from a mandible, respectively. Helladotherium has a geological range covering Vallesian and Turolian of Europe. H. mediterraneum characterizes middle–late Turolian. The remains of Hystrix primigenia are the best determinable material and show great similarities with the Pikermi, Samos and Dytiko (Axios valley) material of H. primigenia (Bonis et al., 1992a). The latter species cover the whole Turolian and thus the presence of Hystrix cannot give a more detailed age than a Turolian one. Taking into account all the above mentioned data, a Turolian age is possible for the Alifakas fauna. The locality of ‘Chalkoutsi’ is situated north of Athens and not very far from Pikermi and Tanagra. The species Hipparion mediterraneum and Hipparion koenigswaldi have been recognized from some maxillary and mandibular remains (Koumantakis, 1971). The latter species was recog-

229

Miocene mammalian successions

230

nized by two teeth of the upper jaw which have a larger size than the others and may belong to H. brachypus. Otherwise there are two species, one Turolian and the other Vallesian. The confusion is even greater because Paraskevaidis (1977) refers to the presence of a third hipparion, H. matthewi. Thus the only age that we can infer for this faunule is late Miocene. The localities of ‘Kerassia’ (Van der Made & Moya`-Sola`, 1989; Theodorou, 1992),‘Achladi’ and ‘Triada’, Evia, have yielded some fossils which characterize the late Miocene. The co-occurrence of H. mediterraneum and H. brachypus in the former could indicate MN 12 but the material needs to be reappraised. The locality of ‘Pyrgos Vassilissis’ is situated northwest of Athens and it was found in 1944. The Wrst determined fauna (Freyberg, 1951) includes the species: ‘Mesopithecus pentelicus’ (= Graecopithecus freybergi Von Koenigswald 1972), Mastodon pentelici, Gazella deperdita, Gazella cf. gaudryi, Tragoportax amalthaea, Helladotherium duvernoyi. Paraskevaidis (1977) determined some material from this locality and gave the following faunal list: Hipparion mediterraneum, Dicerorhinus orientalis, Camelopardalis (= Bohlinia) attica, Gazella sp., Adcrocuta eximia. Most of these determinations are based on very fragmentary material and they are somewhat doubtful. For example there is no horncore to identify the bovids and the names are given from pieces of jaws or isolated teeth whose sizes are close to some species from Pikermi. This is the same for the hominoid primate which is identiWed Mesopithecus, because Mesopithecus exists in the Pikermi fauna. So a general late Miocene age can be only proposed for this fauna. The locality of ‘Servia’ western Macedonia has yielded the following fossil mammals: Hipparion sp., Choerolophodon pentelici, Dinotherium sp., Microstonyx sp. and Rhinocerotidae ind. (Paraskevaidis, 1977). This fauna is very poor for a precise age determination and only a late Miocene age can be proposed. The fauna of ‘Tanagra’ was found near Theva (Beotia), north of Athens during the construction of a tunnel. The collected material is poor (Mitzopoulos, 1961). The fauna is considered similar to Pikermi, Samos and Vathylakkos, and a ‘Pontian’ age is suggested for it. The Hipparion may belong to the species H. mediterraneum whose stratigraphic range is about MN 12. The determination of the bovids is questionable since it is based on mandibular remains, except Prostrepsiceros aV. houtumschindleri from which there are pieces of the horncores. The latter species is known from Halmyropotamos (early Turolian; Bouvrain, 1982) and Maragha (late Vallesian to middle Turolian;Bernor et al., 1979). Thus to give a precise age for the Tanagramaterial is not possible and only a late Miocene age can be proposed. An old collection of large mammals is known from the island of Rhodes

The Miocene large mammal succession in Greece

(Aegean Sea, Greece); it is referred to by Boni (1943). This fauna is very poor and the material not enough for certain determinations. e.g. the two bovids were determined by two maxillae, and hipparion by a sole maxilla with P2/-M1/. Thus it is not possible to suggest a more precise age than late Miocene. The locality of ‘Ano Metochi’ is situated in the Strymon basin, eastern Macedonia, Greece. The known material from this locality is scarce and fragmentary. Some material was determined by Professor J. Melentis and the species Hipparion mediterraneum, Prostrepsiceros woodwardi and Gazella cf. gaudryi were recognized (Armour-Brown et al., 1977). One of us (G. K.) had determined some remains from Ano Metochi as Helladotherium cf. duvernoyi, Hipparion sp. and Gazella sp., and proposed a late Miocene age (Karistineos, 1984). Sondaar & de Bruijn (1979), basing their conclusions on the relative proportions and indices of some metapodials from Ano Metochi, have suggested a late Turolian age for this locality. However, the material is very scarce for a detailed dating and it is better to consider the age of Ano Metochi as late Miocene. Remains of SteneoWber have been described from the basin of Serres (Melentis, 1966b) and dated to late Miocene. The locality of ‘Maramena’, northern Macedonia, has been recently studied (Schmidt-Kittler, 1995). It is quite diYcult to get a precise idea of the dating through the large mammals but the micromammals allow us to date the locality near the limit of late Miocene–early Pliocene.

Conclusion When compared with some other European countries, the Miocene Greek large mammalian faunas are poorly documented in some biozones. Nothing is known for MN 1 to MN 2 and maybe MN 3. We know only two species of carnivores and maybe Brachyodus from MN 4, but the new locality of Antonios will allow a better knowledge of the limit MN 4–MN 5. New excavations on the island of Chios allow a better knowledge of MN 5. MN 6–8 are as yet very poorly documented. No locality can be certainly attributed to the early Vallesian (MN 9) but the data are far better for the late Vallesian (MN 10), especially in Macedonia. The Turolian (MN 11–13) is certainly the best period for fossil mammalian localities of Greece. Nevertheless it would be necessary to excavate some old localities because the origins of the material found during the nineteenth century and the Wrst half of the twentieth century are not always well known. Such work has begun and in the next few years the knowledge of Miocene Greek fossil faunas will certainly be increasing.

231

Miocene mammalian successions

232

Acknowledgements We thank the European Science Foundation for inviting us to the workshop in San Feliu de Guixol, and our colleague J. Agustı´ for organizing the workshop. Most of the Weld campaigns for excavating in Greece were allowed by fundings of the Fondation Singer-Polignac, the Leakey Foundation for Anthropological Research, the University of Poitiers and the Aristotle University of Thessaloniki. This work is also a part of the program ‘Pale´oenvironnement–Evolution des Hominide´s’ of the French CNRS. We thank very much all the people who excavated with us over the years and those who have worked to prepare the fossils. We thank particularly G. Bouvrain for great help in the listing of the faunas. The manuscript has been prepared by G. Florent. We are greatly indebted to P. Andrews for his careful reading of the manuscript.

References Andrews, C. W. 1918. Note on some fossil mammals from Salonica and Imbros. Geological Magazine, London, 5 (6), 540–3. Arambourg, C. & Piveteau, J. 1929. Les verte´bre´s du Pontien de Salonique. Annales de Pale´ontologie, 18, 59–138. Armour-Brown, A., Bruijn, H. de, Maniati, C., Siatos, G. & Niesen, P. 1979. The geology of the Neogene sediments north of Serrai and the use of rodent faunas for biostratigraphic control. Proceedings of the VI° Colloquium on the Geology of Aegean Region, 2, 615–22. Benda, L. & Meulenkamp, J. E. 1990. Biostratigraphic correlations in the eastern Mediterranean Neogene. 9. Integrated biostratigraphic and chronostratigraphic scales. News in Stratigraphy, 23, 1–10. Bernor, R. L., Tobien, H. & Van Couvering, J. 1979. The mammalian biostratigraphy of Maragheh. Annales Ge´ologiques des Pays Helle´niques, hors se´rie 1, 91–9. Beseneker, H. 1973. Neogen und Quarta¨r der Insel Chios (Agais). Diss. Thesis, p. 196. Boni, A. 1943. Fauna ad Hipparion a Rodi. Paleontographia Italica, 41, 23–36. Bonis, L. de & Bouvrain, G. 1996. Suidae du Mioce`ne supe´rieur de Gre`ce. Bulletin Muse´um National Histoire Naturelle, 4 e se´r., 18, C (1), 107–32. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. 1990a. New hominid skull material from the late Miocene of Macedonia in northern Greece. Nature, 345, 712–14. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. D. 1990b. New remains of Mesopithecus (Primates, Cercopithecoidea) from the late Miocene of Macedonia (Greece) with the description of a new species. Journal of Vertebrate Paleontology, 10 (4), 473–83. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. D. 1992a. A skull of Hystrix primigenia from the late Miocene of Macedonia. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen, 2, 75–87. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. D. 1992b. Diversity and

The Miocene large mammal succession in Greece

paleoecology of Greek late Miocene mammalian faunas. Palaeogeography Palaeoclimatology Palaeoecology, 91, 99–121. Bonis, L. de, Bouvrain, G., Geraads, D., Koufos G. & Sen, S. 1994. The Wrst aardvarks (Mammalia) from the late Miocene of Macedonia, Greece. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen, 194, 2–3, 343–60. Bonis, L. de & Koufos, G. D. 1981. A new hyaenid (Carnivora, Mammalia) in the Vallesian (late Miocene) of Northern Greece. ScientiWc Annals of the Faculty of Physics and Mathematics, University of Thessaloniki, 21, 79–94. Bonis, L. & Koufos, G. 1993. The face and the mandible of Ouranopithecus macedoniensis: description of new specimens and comparisons. Journal of Human Evolution, 24, 469–91. Bonis, L. de, Koufos, G. D. & Sen, S. 1997a. A giraYd skull and mandible from the middle Miocene of the island of Chios (Aegean Sea, Greece). Palaeontology, 40 (1), 121–33. Bonis, L. de, Koufos, G. D. & Sen, S. 1997b. The sanitheres (Mammalia, Suoidea) from the middle Miocene of Chios island, Aegean sea, Greece. Revue de Pale´obiologie, Gene`ve, 16 (1), 259–70. Bonis, L. de, Koufos, G. D. & Sen, S. (in press). Ruminants (Bovidae, Tragulidae) from the middle Miocene (MN 5) of Chios island (Greece). Bonis, L. de & Melentis, J. 1975. Premie`re de´couverte de Muride´s (Mammalia, Rodentia) dans le Mioce`ne de la re´gion de Thessalonique (Gre`ce). Pre´cision sur l’aˆge des dryopithecine´s de Mace´doine. Comptes Rendus Acade´mie des Sciences, Paris, 28, D, 1233–6. Bonneau, M. & Ginsburg, L. 1974. De´couverte de Dorcatherium puyhauberti Arambourg et Piveteau (Mammalia) dans les facie`s continentaux de base de la molasse tertiaire de Cre`te (Gre`ce): Conse´quences stratigraphiques et tectoniques. Compte Rendu Sommaire de la Socie´te´ Ge´ologique de France, Paris, 16, (1), 11–12. Bouvrain, G. 1979. Un nouveau genre de Bovide´ de la Wn du Mioce`ne. Bulletin de la Socie´te´ Ge´ologique de France, Paris, 21, 4, 507–11. Bouvrain, G. 1980. Le genre Palaeoreas (Mammalia, Artiodactyla, Bovidae): syste´matique et extension ge´ographique. Pala¨ontologische Zeitschrift., 54, 1–2, 55–65. Bouvrain, G. 1982. Re´vision du genre Prostrepsiceros. Pala¨ontologische Zeitschrift, 56 (1–2), 113–24. Bouvrain, G. 1996. Les gazelles du Mioce`ne supe´rieur de Mace´doine, Gre`ce. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen, 199 (1), 111–32. Bouvrain, G. 1997. Les bovide´s du Mioce`ne supe´rieur de Pentalophos (Mace´doine, Gre`ce). Mu ¨ nchner Gewissenschaften Abhandlungen, (A), 34, 5–22. Bouvrain, G. & Bonis, L. de 1984. Le genre Mesembriacerus (Bovidae, Artiodactyla, Mammalia): un ovibovine´ primitif du Valle´sien (Mioce`ne supe´rieur) de Mace´doine (Gre`ce). Palaeovertebrata, 14 (4), 201–23. Bouvrain, G. & Bonis, L. de 1985. Le genre Samotragus (Artiodactyla, Bovidae) une antilope du Mioce`ne supe´rieur de Gre`ce. Annales de Pale´ontologie, 71 (4), 257–94. Bruijn, H. de, Daams, R., Daxner-Ho¨ck, G., Fahlbusch, V., Ginsburg, L., Mein, P. & Morales, J. 1992. Report of the RCMNS working group on fossil mammals, Reisenberg 1990. Newsletter in Stratigraphy, 20 (213), 65–118.

233

Miocene mammalian successions

234

Bruijn, H. de, Sondaar, P. & Zachariasse, E. W. J. 1971. Mammalia and Foraminifera from the Neogene of Kastellios hill (Crete): correlations of continental and marine biozones. Koninkllijke Nederlands Akademie van Wetenschappen, Amsterdam, B 74, 1–22. Bruijn, H. de & Zachariasse, W. J. 1979. The correlation of marine and continental biozones of Kastellios hill reconsidered. Annales Ge´ologiques des Pays Helle´niques, 1, 219–26. Dimopoulos, G. C. 1972. Dicerorhinus orientalis aus dem obermioza¨n (Pont) des Beckens von Langadas (Mazedonien,Griechenland). Folia Biochimica et Biologica Graeca, 9, 47–60. Freyberg, G. B. v. 1951. Die Pikermifauna von Tour la Reine. Annales Ge´ologiques des Pays Helle´niques, 1, 3, 7–10. Gaudry, A. 1862–7. Animaux fossiles et ge´ologie de l’Attique, Savy, Paris, 472 pp. Gaudry, A. & Lartet, E. 1856. Sur les re´sultats de recherches pale´ontologiques entreprises dans l’Attique sous les auspices de l’Acade´mie. Comptes Rendus de l’Acade´mie des Sciences, Paris, 43, 271–4. Geraads, D. 1989. Un nouveau GiraYde´ du Mioce`ne supe´rieur de Mace´doine (Gre`ce). Bulletin du Muse´um National d’Histoire Naturelle, 4e se´r., 11, C, 4, 189–99. Howell, F. C. & Petter, C. 1985. Comparative observation on some middle and upper Miocene hyaenids. Genera: Percrocuta Kretzoı¨, Allohyaena Kretzoi (Mammalia, Carnivora, Hyaenidae). Geobios, Lyon, 18, 419–76. Karistineos, N. 1984. Paleogeographic evolution of the Serres basin. Lithostratigraphy, Biostratigraphy and Tectonics. Ph. D. Thesis, University of Thessaloniki, 230 pp. Koliadimou, K. 1996. Palaeontological and biostratigraphical study of the Neogene/Quaternary micromammals of Mygdonia basin (Macedonia, Greece). Ph. D. Thesis, University of Thessaloniki, 465 pp. Kondopoulou, D., Bonis, L. de, Koufos, G. D. & Sen, S. 1993. Paleomagnetic data and biostratigraphy of the middle Miocene vertebrate locality of Thymiana (Chios Island, Greece). Proceedings of the 2nd Congress of the Geophysical Society of Greece, 2, 626–35. Kostopoulos, D. S. (in press). Microstonyx major (Suidae, Artiodactyla) from the late Miocene Locality of ‘Nikiti-1’, Macedonia, Greece; some remarks about the species. Proceedings of the 7th Congress of the Geological Society of Greece, Thessaloniki 1994. Bulletin of the Geological Society of Greece. Kostopoulos, D. & Koufos, G. D. 1996. Late Miocene bovids (Mammalia, Artiodactyla) from the locality ‘Nikiti 1’ (NKT) Macedonia, Greece. Annales de Pale´ontologie, 81, 251–300. Koufos, G. D. 1985. Hipparion sp. (Equidae, Perissodactyla) from Diavata (Thessaloniki, Northern Greece). Bulletin of the British Museum of Natural History, London, Geology, 38, 5, 335–45. Koufos, G. D. 1986. Study of the Vallesian hipparions of the Lower Axios Valley (Macedonia, Greece). Geobios, Lyon, 19, fasc. 1, 61–79. Koufos, G. D. 1990. The hipparions of the lower Axios valley (Macedonia, Greece). Implications about the Neogene stratigraphy and the evolution of Hipparions. In European Neogene Mammal Chronology, Lindsay, E., Fahlbusch, V. & Mein, P. (eds.), pp. 321–38, Plenum Press, New York. Koufos, G. D. 1993. A mandible of Ouranopithecus macedoniensis from the late

The Miocene large mammal succession in Greece

Miocene of Macedonia (Greece). American Journal of Physical Anthropology, 91, 225–34. Koufos, G. D. 1995a. The Wrst female maxilla of the hominoid Ouranopithecus macedoniensis from the late Miocene of Macedonia (Greece). Journal of Human Evolution, 29, 385–99. Koufos, G. D. 1995b. The late Miocene percrocutas (Carnivora, Mammalia) of Macedonia, Greece. Palaeovertebrata, 24 (1–2), 67–84. Koufos, G. D., Bonis, L. de & Sen, S. 1995. Lophocyon paraskevaidisi, a new viverrid (Carnivora, Mammalia) from the middle Miocene of Chios island, Greece. Geobios, 28 (4), 511–23. Koufos, G. D. & Melentis, J. K. 1984. The late Miocene (Turolian) mammalian fauna of Samos Island (Greece). ScientiWc Annals of Faculty Sciences University, Thessaloniki, 24, 47–78. Koufos, G. D. & Syrides, G. E. 1997. A new early/middle Miocene mammal locality from Macedonia, Greece. Comptes Rendus Acade´mie des Sciences Paris, 325, 511–16. Koufos, G. D., Syrides, G. E., Koliadimou, K. K. & Kostopoulos, D. S. 1991. Un nouveau gisement de verte´bre´s avec hominoı¨de dans le Mioce`ne supe´rieur de Mace´doine (Gre`ce). Comptes Rendus de l’Acade´mie des Sciences, Paris, III, 313, 691–6. Koumantakis, J. 1971. Pontian Formations at Chalkoutsi (Atica). Annales Ge´ologiques des Pays Helle´niques, 23, 274–84. Lehman, V. & Tobien, H. 1993–95. Artiodactyle fossilen (Mammalia) aus dem Mioza¨n von Thymiana, Chios. Annales Ge´ologiques des Pays Helle´niques, 36, 403–14. Marinos, G. & Symeonidis, N. 1974. Neue Funde aus Pikermi, Attika und eine allgemeine geologische u¨bersicht dieses pala¨ontologischen Raumes. Annales Ge´ologiques des Pays Helle´niques, 26, 1–27. Mein, P. 1990. Updating of MN zones. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 21–338, Plenum Press, New York. Melentis, J. K. 1966a. Stu ¨ dien u ¨ ber fossile Vertebraten Griechenlands 14. Der erste Nachweis von Brachyodus onoideus (Mammalia, Anthracotheriidae) aus Griechenlands und die Datierung der Fundschichten. Annales Ge´ologiques des Pays Helle´niques, 17, 221–35. Melentis, J. K. 1966b. Stu¨dien u ¨ ber fossile Vertebraten Griechenlands 18. SteneoWber jaegeri aus Ligniten von Serrae und die Datierung der Fundschichten. Annales Ge´ologiques des Pays Helle´niques, 17, 289–97. Melentis, J. K. 1967. Die pikermifauna von Halmyropotamos (Eu ¨ boa/Griechenland). Practika, Akademie Athen, 41, 261–6. Melentis, J. K. & Schneider, H. 1966. Eine neue Pikermifauna in der ortschaft Alifaka in Thessalien (Griechenland). Annales Ge´ologiques des Pays Helle´niques, 17, 267–88. Melentis, J. & Tobien, H., 1967. Pala¨ontologische Ausgrabungen auf der Insel Chios (eine vorla¨uWge Mitteilung). Praktika Akademie Athen, 42, 147–52. Mitzopoulos, M. K. 1961. Die Hipparionfauna von Tanagra bei Theben. Annales Ge´ologiques des Pays Helle´niques, 12, 301–14. Paraskevaidis, E. 1940. Eine obermioca¨ne Fauna von Chios. Neues Jahrbuch fu ¨r Mineralogie, Geologie und Pala¨ontologie, 83 (B), 363–442.

235

Miocene mammalian successions

236

Paraskevaidis, E. 1977. Sa¨ugetierreste aus Griechenland. Procceedings of the VI° Colloquium on the Geology of the Aegean Region, 3, 1143–54. Psarianos, P. 1958. Neue Rhinocerotidenfunde aus dem Tertia¨r und Quarta¨r von Mazedonien (Griechenland). Proceedings Akademie Athens, 33, 1–10. Schmidt-Kittler, N. 1983. The Mammals from the Lower Miocene of Aliveri (Island of Evia, Greece) III. A new species of Sivanasua Pilgrim, 1931, Feliformia, Carnivora) and the phylogenetic position of this genus. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Amsterdam, B, 86, 3, 301–18. Schmidt-Kittler, N. 1995. The vertebrate locality Maramena (Macedonia, Greece) at the Turolian-Ruscinian boundary. Mu ¨ nchner Gewissenschaften, A 28. Sen, S., Bonis, L. de & Koufos, G. D., in prep. Micromammals of the Middle Miocene (MN 5) of Chios island (Greece). Sen, S. & Valet, J. P. 1986. Magnetostratigraphy of late Miocene continental deposits in Samos, Greece. Earth and Planetary Science Letters, 80, 167–74. Solounias, N. 1981. The Turolian fauna from the island of Samos, Greece. Contributions to Vertebrate Evolution, 6, 1–232. Sondaar, P. Y. & Bruijn, de H. 1979. Hipparion a useful tool for biostratigraphic zonation. Annales Ge´ologiques des Pays Helle´niques, hors se´r., 1, 1123–6. Steininger, F., Bernor, R. & Fahlbusch, V. 1990. European Noegene marine/continental chronologic correlations. In European Neogene Mammal Chronology, Lindsay, E., Fahlbusch, V. & Mein, P. (eds.), pp. 15–46, Plenum Press, New York. Swisher III, C. C. 1996. New 40Ar-39Ar dates and their contribution toward a revised chronology for the late Miocene of Europe and west Asia. In The Evolution of western Eurasian Neogene Mammal Faunas, Bernor, R. L., Falhbusch, V. & Mittman, H. W. (eds.), pp. 64–77. Columbia University Press, New York. Symeonidis, N. 1973. Chalicotherium goldfussi (Perissodactyla, Mammalia) aus dem Altplioza¨n von Pikermi (Griechenland). Annales Ge´ologiques des Pays Helle´niques, 25, 301–7. Theodorou, G. E. 1992. Comparative study of Hipparion from Argolis, Attica and Euboa (Southern Greece). Xth Regional Committee of the Mediterranean Neogene, abstracts, Journal of Stratigraphy, 76, 127–8. Tobien, H. 1968. Pala¨ontologische Ausgrabungen nach jugtertia¨ren Wirbeltieren auf der Insel Chios (Griechenland) und bei Maragheh (NW Iran). Jahrbuch Vereinig ‘Freunde Univ. Mainz’, 7, 51–8. Tobien, H. 1977. Die mittelmioza¨nen Wirbeltierfundstellen su¨dlich Thymiana (Insel Chios, Agais, Griechenland). Annales Ge´ologiques des Pays Helle´niques, 28, 489–94. Tobien, H. 1980. A note on the skull and mandible of a new choerolophodonte mastodont (Proboscidea, Mammalia) from the middle Miocene of Chios (Aegean sea, Greece). In Aspects of Vertebrate History: Essays in honor of Edwin Harris Colbert, Jacobs, L. L. (ed.), pp. 299–307, FlagstaV, Arizona. Van Couvering, J. A. 1972. Radiometric calibration of the European Neogene. In Calibration of Hominoid Evolution, Bishop, W. W. & Miller, J. A. (eds.), pp. 247–71, Scottish Academic Press, Edinburgh. Van Couvering, J. A. & Miller, J. A. 1971. Late marine and non marine time scale in Europe. Nature, 230, 559–63.

The Miocene large mammal succession in Greece

Van Der Made, J. & Moya`-Sola`, S. 1989. European Suinae (Artiodactyla) from the late Miocene onwards. Boletino della Societa Paleontologica Italiana, 28, 329–39. Weidman, M., Solounias, N., Drake, R. E. & Curtis, G. H. 1984. Neogene stratigraphy of the eastern basin, Samos island, Greece. Geobios, 17, 477–90. Zapfe, H. 1991. Mesopithecus pentelicus Wagner aus dem Turolian von Pikermi bei Athen. Odontologie und osteologie. Neue Denk-Schriften des Naturhistorisches Museum, Wien, 5, 13–203.

237

12 Chronology and mammal faunas of the Miocene Sinap Formation, Turkey Juha Pekka Lunkka, Mikael Fortelius, John Kappelman and Sevket Sen

Introduction The Sinap Formation of Central Anatolia, Turkey, contains fossiliferous sediments that represent the time period from about 15 Ma to about 2.5 Ma and documents an important interval of the terrestrial Neogene large land mammal evolution. Although the middle and late Miocene mammalian faunas from the Sinap Formation have been known for over 50 years, it is not until recently that a geological and chronological framework has emerged for these sediments. This framework is critical for producing a comprehensive view of the mammalian succession that can in turn be compared with results from localities in other regions. The fossiliferous Miocene sediments of the Sinap Formation are located some 55 km northwest of Ankara, Central Anatolia (Fig. 12.1). The Formation is named after Sinap Tepe, a prominent butte or ‘tepe’ in the area. The tectonic setting for this part of Central Anatolia is dominated by small fault-bounded basins (termed intermontane basins of various sizes: cf. Lu ¨ ttig & SteVens, 1976; Erol, 1981) where folding events that are synchronous and pre- and postdate the Miocene have further complicated the structural geology of the area (Lunkka et al., 1998). These features are most likely related to the extensional neotectonics of Anatolia and the major North Anatolian transform fault (Angelier et al., 1981; Inci, 1991) that is located to the north of this region. The Miocene sediments in the study area rest unconformably upon suites of pre-Miocene tectono-sedimentary melange and forearc basin-Wll rocks that are related to the closure of the Neo-Tethyan Ocean (S¸engo¨r & Ylimaz, 1981; Koc¸yigˇit, 1991; see Fig. 12.1). Terrestrial Miocene strata consists mainly of volcaniclastic sediments that are relatively well-exposed along the Xanks of several prominent buttes in the general region around the town of Kazan (see Fig. 12.1). The Miocene sequence is conformable on both sides of the major fault zones (see Fig. 12.1), but because these sediments are often covered by recent alluvium, the total throw of the faults is not known. Biochronological and magnetostratigraphical results, however, as well as observations on the structural geology of the area, provide a Wrm basis for cross-faultzone correlations of the sediments. Previous work carried out on the lithostratigraphy of the Miocene sequence in the area has been very limited due partly to the complexity of the sedimentary sequence that resulted from tectonic movements and volcanic

Chronology and mammal faunas

[Figure 12.1] Location map of the general geology of the Ankara area modified after Koc¸yigˇit (1991) and a detailed map of the Sinap Tepe – Beycedere – Kavakdere area NW of Kazan where most of the fossil localities occur. The location of the fossil localities at Igˇbek (Loc. 49) and C¸obanpinar (Loc. 42) are marked in the general geology map.

Miocene mammalian successions

240

activity during the Neogene as mentioned above. Ozansoy (1957, 1965) and ¨ ngu later O ¨ r (1976) established a stratigraphical scheme for the Miocene sequence. Ozansoy (1957, 1965) named three members of the Sinap Formation and recognized a conformable contact between the lower and the middle members, and an erosional unconformity between the upper and the middle members. O ¨ ngu ¨ r (1976) named an additional older unit, the Pazar Formation, that is separated from the overlying Sinap Formation by a volcanic unit (basaltic Xow). Thereafter work on the Miocene sediments was carried out mainly by the Sinap Project Members (see Sen, 1991; Kappelman et al., 1996). This work has considerably extended the knowledge of the litho- and magnetostratigraphy of the sediments and paleoenvironmental and faunal succession of the area during the Miocene. The Sinap Formation now oVers a stratigraphically and paleoenvironmentally resolved sequence of land mammal faunas spanning almost 10 million years, including a densely sampled interval from about 11 to 10 million years. This densely sampled part contains both the middle / late Miocene boundary that is identiWed by the appearance of hipparionine horses and is the time leading up to one of the major turnover phases in the Neogene history of Eurasian land mammals, the ‘Vallesian crisis’ (Agustı´ & Moya`-Sola`, 1990; Agustı´, this volume). Major changes in the mammal fauna are observed both in western and central Europe (hereafter West, as deWned in Fortelius et al., 1996) and in eastern Europe and western Asia (hereafter East). The changes, which outside the Spanish type area have primarily been documented from isolated localities dated primarily or exclusively by land mammal biochronology, had a diVerent ecological content in West and East. In West, brachydont herbivores and small, chieXy arboreal carnivores decreased drastically in diversity, accounting for most of a decrease by some 75% of the large mammal species richness. In East no marked change in the ecological structure of the mammal fauna was found, but large mammal species richness was more than doubled. The change in West appears to have been relatively sudden while the process in East was more prolonged, continuing (indeed, accelerating) past the MN 9 / MN 10 boundary horizon of the classical Vallesian crisis into MN 11 (Fortelius et al., 1996). The mammal faunas of the Sinap Formation oVer the possibility to investigate the details of the changes following the middle/late Miocene boundary in East and their temporal relationship to the Vallesian crisis of West. The descriptive work on the Sinap material is still in progress, however, and only a preliminary overview is given here. For this reason we have conWned the present study to genus-level, basic richness and turnover analysis of a set of ‘major’ Sinap localities in the approximate interval 15–6

Chronology and mammal faunas

Ma. More complete results, including taphonomic and ecomorphological analyses will be presented in a forthcoming monograph volume (Fortelius et al., in prep.). In this article we will establish a more deWned chronology for the Sinap Formation and discuss the sedimentary and faunal events that occurred in the area during late middle and late Miocene including the Vallesian with evidence obtained from combined geological, magnetostratigraphical, and paleontological studies.

General lithostratigraphy and inter-areal correlations of the Miocene sequence The general lithostratigraphy of the study area around Kazan (see Fig. 12.1) is shown in Fig. 12.2. The oldest Miocene sediments in the area are the current-bedded volcaniclastics that occur below the basalt Xow and thus ¨ ngu ¨ r (1976). Above the basalt Xow, dated belong to the Pazar Formation of O at 15.2 ± 0.3 Ma by whole rock K/Ar radiometric analysis (Kappelman et al., 1996), is a thick accumulation of alluvial volcaniclastics that forms the Sinap Formation. The best outcrops of conformable section occur to the south of the Fault Zone (FZ 1; see Fig. 12.1) and this is also where most of the fossil localities have been found. Based on lithological characteristics, the Sinap Formation can be divided into Yellidoruk (dominated by pebble to cobble conglomerate units), Lower Sinap (characterized by thick mudstone units), Middle Sinap (includes cycles of upward Wning units from sandstone to mudstone) and Upper Sinap Members (consists of pebble to boulder conglomerate units). Of these four members, the Lower, Middle and the Upper Sinap Members correspond more or less to those already recognized by Ozansoy (1957, 1965), but the total thickness of the conformable sequence is much greater than recognized by Ozansoy (1957, 1965) and approaches 200 m. The contact between the Middle and Upper Sinap Members is erosional and is unconformable further south of Sinap Tepe. It is, therefore, assumed that the coarse conglomerate units that compose the Upper Sinap Member represent post-Miocene accumulation that took place after an episode of faulting and this member is not discussed here. North of the Fault Zone 2 (FZ 2) the conformable Sinap sequence above the basalt Xow (15.2 ± 0.3 Ma) is more than 450 m thick. In this area the Sinap Formation can be grouped into Beycedere (270 m thick) and Kavakdere (180 m thick) Members. There seems, however, to be a perhaps

241

Miocene mammalian successions

242

[Figure 12.2] Lithostratigraphy of the Sinap Formation in the study area. South of the fault zone 1 (FZ 1 in Fig. 12.1) the Sinap Formation consists of the Yellidoruk, Lower Sinap and Upper Sinap Members, while in the north of FZ 1 sediments are divided into Beycedere, Kavakdere and C¸alta Members. Note that the Upper Sinap Member defined by Ozansoy (1957, 1965) and O¨ngu¨r (1976) is considered to represent post-Miocene sediments and is not discussed in this paper.

considerable hiatus between these two members that is manifested by a Xat iron surface of siliciWed sediments. Rare and indeterminate fossil vertebrates have been discovered in the Beycedere, but fossil plants are rather common and sometimes occur in thin lenses. Although fossil vertebrates are also uncommon in the lower portion of the Kavakdere Member, several localities in the upper part of this Member contain rich accumulations.

Chronology and mammal faunas

Inter-areal correlation of sediments on both sides of the fault zones and the fossil-rich locality at Igˇbek located to the east of Kazan (see Fig. 12.1) is based on observations from the structural geology, biochronology and magnetostratigraphy as shown in Fig. 12.3. This article concentrates only on those members of the Sinap Formation that are important to the faunal events that are discussed. A full account of the areal geology will be discussed in Lunkka & Ekart (in prep.).

Sedimentation and sedimentary environments The major portion of the Sinap Formation sediments displays a suite of facies and facies architecture that indicate terrestrial deposition by channelized and non-channelized alluvial systems, whereas the lower part of the Kavakdere sediments are of lacustrine origin (Kappelman et al., 1996). The main facies and architectural elements of each lithostratigraphical member (listed in Table 12.1) clearly indicate that the landscape during the time of deposition was varied and dynamic. Sedimentological history of the area witnessed sometimes drastic changes as a result of tectonic movements. Graben formation and strike-slip faulting events changed the land surface gradient several times during the late middle and late Miocene and these events in turn lead to signiWcant lateral and vertical facies changes in sediments. The basal part of the sequence (i.e., the Lower Sinap Member) is dominated by thick mudstone units (see Fig. 12.3) with moderately well developed paleosols at distinct horizons. Mudstone certainly represents overbank Wnes that were laid down in a well-drained Xood plain environment. Only a few minor coarser units occur in the lowest member and they all represent channel Wll gravels and sands (maximum thickness of channel Wlls is 3 m and channel width is less than 8 m). Corrected paleocurrent measurements from current induced structures in the channel gravels and sands indicate paleoXow from the northeast. The Yellidoruk Member is characterized by gravity Xow and Xuidal Xow pebble to cobble conglomerate units (see Table 12.1 and Fig. 12.3) that have a clast composition that is similar to the sandstone and conglomerate units in the Lower Sinap Member. Litho- and magnetostratigraphical as well as provenance studies indicate that, during the period when the Sinap Tepe area experienced a prolonged period of Xood plain deposition (i.e., during the deposition of the Lower Sinap Member), tectonic movements further north of the study area initiated alluvial fan deposition. The Yellidoruk Member represents a rather proximal facies of an alluvial fan in a

243

Miocene mammalian successions

[Figure 12.3] Age of the Sinap Formation sediments based on magneto- and lithostratigraphical results obtained from the study area. The main fossil localities discussed in the text are also indicated on the logs along the various lithostratigraphical members and their environmental settings.

Chronology and mammal faunas

situation where streams conWned by narrow valleys emerged onto a trunk river valley. In the Sinap Tepe area the Lower Sinap Member passes gradually into the Middle Sinap Member. This next member in the succession is characterized by numerous laterally extensive horizons of upward Wning channel Wlls. These channelized Xows pass laterally and vertically into non-channelized, upward Wning cycles of sheet Xood deposits. These Xows in turn pass conformably into a sequence that is composed of thick mudstone beds that include relatively uncommon channel Wll gravel and sand units. The Middle Sinap Member represents a facies change from the small braided channels on the lower fan segment to Xood deposits located in a distal fan that was close to the adjacent Xood plain. These distal alluvial fan sediments show increased interWngering with Xood plain sediments in the down current direction. Paleocurrent measurements and clast and sandstone lithology indicate that the provenance area for the lower fan and distal fan sediments was located in the northwest, and it is suspected that the alluvial fan sedimentation was again caused by tectonic events that took place north or northwest of the area. However, the upper portion of the Middle Sinap Member sediments belongs to the Xood plain depositional regime where sands and gravels in channels show paleoXow from the northeast, most likely indicating orientation of Xow in the trunk river perpendicular to the distal alluvial fan sediment input. The Kavakdere area is located northwest of the fault zones (see Fig. 12.1) and seems to have experienced a slightly diVerent kind of sedimentation history compared to the Sinap Tepe area. In the Kavakdere region the Xuviatile Beycedere Member is characterized by thick, slightly pedogenically altered mudstone beds. The overlying sediments of the Kavakdere Member appear to be partially lacustrine in origin at their base as indicated by thick mudstone facies that include frequent limestone beds and freshwater gastropods (e.g., Lymnaea). Sediments coarsen towards the top of the Kavakdere Member, thus indicating the inWlling of the lake basin. The Xuvial sequence at Igˇbek (see Loc. 49 on Fig. 12.1) is part of a similar but probably much larger river system than that encountered in the Sinap Tepe–Kavakdere area. The Igˇbek sequence located to the east of Kazan can be correlated to those sediments described above by means of the fossil mammalian chronology and paleomagnetic reversal stratigraphy. Fossil localities are not evenly distributed in the Sinap sequence. The densest occurrence of fossils (e.g., fossil localities 108, 72, 91, 114, 8B, 8A, 12) was found in the boundary of the distal alluvial fan and the distal Xood plain subenvironments (see Fig. 12.3). It is assumed that in such a setting deposition was relatively continuous and thus more favorable for burial

245

Miocene mammalian successions

246

Table 12.1. Facies geometry, main facies and lower contacts of the architectural elements in diVerent members of the Sinap Formation and their paleoenvironmental interpretations

Member

Facies geometry and main facies

Lower contact

Environment (small fluvial – flood plain with moderately developed paleosols fluvial channel deposits and levee sands debris flow in a proximal alluvial fan migrating small stream deposits in an alluvial fan

Lower Sinap

sheets – Fm (channels – Gm, Sm upwards fining) (sheets that thin distally – Sm)

conformable (erosional)

Yellidoruk

wedge-shaped sheets, lenticular shaped lobes – Gms (broad channels – Gm, Gp, Gt) (sheets – Sm, Fm) channels (Gm, Gp) (wedge- and tabular sheets – Sm, Fm)

non-erosional (erosional) (sharp, conformable)

Middle Sinap

Beycedere

erosional (sharp, conformable)

tabular and wedge-shaped sheets, upwards fining cycles – Gm, Gp, Sh, Sp, Sr, Sd, Fm

sharp (occasionaly loaded)

thick sheets – Fm (channels – Gp, Sp, wedge-shaped sheets – Sm)

conformable (sharp, erosional

thick sheets – Fm

conformable

(channels and bars, upwards fining cycles – Gm, Sp, St, Sr, Fm, wedge-shaped sheets – Sm sheets – Fl and lake marl)

(conformable)

small fluvial channel and bar deposits in mid-fan environment with sand and mud deposition between channels fluvial channel deposits in mid-fan pass laterally and vertically into non-channelized sheet flood deposits in a distal fan environment fluvial – flood plain with moderately developed paleosols, minor tributary channels and bars fluvial – flood plain with moderately developed paleosols lacustrine laminated mud and lake marl

Chronology and mammal faunas

Table 12.1 (cont.).

247

Member

Facies geometry and main facies

Lower contact

Environment

Kavakdere

thick sheets – Fm

conformable

(broad channels – Gm, Gp, Sp, Sr and thin sheets – Sm) broad channels and bars – Gm, Gp, Sp, St, Sr, Sh

(sharp, erosional)

lacustrine mud including lake marls (limestone) fluvial channels filling the lake basin

Igˇbek

(sheets and wedge shaped sheets – Fm and Sm)

erosional

fluvial channel and bar deposits

sharp, conformable

overbank sands and fines

Facies codes are modified after Miall (1978). Gms = massive, matrix supported gravel; Gm = massive or crudely bedded gravel; Gp = planar cross-bedded gravel; Gt = trough cross-bedded gravel; Sp = planar cross-bedded sand; St = trough cross-bedded gravel; Sr = small ripple-bedding; Sh = plane-bedded sand; Sm = massive sand; Fm = massive fines (clay and silt).

and preservation of the skeletal material. In the Xood plain environment, fossil accumulations (e.g., fossil localities 65, 64, 4, 94, 26, 34) were frequently discovered but not at as dense intervals as in the previous environmental setting. The virtual absence of fossil accumulations in the Yellidoruk Member and channel-gravel dominated parts of the Igˇbek section are most probably a result of the destructive nature of these high energy environments.

Chronology Attempts to produce an absolutely dated chronology for the Sinap Formation have in the past been complicated by both the isolated outcroppings of sediments and the fact that many of these exposures are fault-bounded. Sen (1991) demonstrated that the sediments of the Sinap Formation preserve a stable primary magnetic signal that makes these sediments useful for paleomagentic reversal stratigraphy, and these Wndings have been the basis of all later work. Much of the focus of the Sinap Project aimed to tie together the exposed outcroppings by direct stratigraphic mapping in order to produce sections of suYcient thickness so that paleomagentic reversal stratigraphy could be usedto date the Formationand correlatebetweenisolated areas.

Miocene mammalian successions

248

Preliminary results from the paleomagnetic reversal stratigraphy for sediments located around Sinap Tepe and Kavakdere are presented in Kappelman et al. (1996). These results have now been enlarged upon with the inclusion of additional samples collected from conformable geologic sections newly described in both regions. In addition, samples from Igˇbek (discussed above) are also now available for analysis (Feseha, 1996). The new work followed the same laboratory protocols as Kappelman et al. (1996) and will be treated in full by Kappelman et al. (in prep.). The total thickness of conformable sediments from the three lower members of the Sinap Formation now stands at 195 m with additional sections added to both the bottom and top of the 110 m described up to 1996. The most important aspect of this new work is that the Sinap Tepe section now provides a much better correlation with the long normal of Chron C5n of the geomagnetic reversal time scale (GRTS). Early results documented nearly 100 m of normally magnetized sediments with a single site of reversed polarity located at the top of this interval (Kappelman et al., 1996: Wgs. 6.7 & 6.10), and this interval was tentatively correlated with the younger portion of Chron C5n. Additional sampling of about 70 m of overlying sediments and 20 m of underlying sediments revealed several additional intervals of predominantly normal polarity along with four shorter intervals of reversed polarity. It now appears that the majority of Chron C5n is documented in the sediments of Sinap Tepe, along with C5r.2r to C5r.1r at the base of the section, and C4Ar.2n at the top of the section (see Fig 12.3). The majority of the fossil localities occur in the middle portion of C5n and provide a dense sampling of this interval of time. New sampling in Kavakdere extended the base of this section 110 m down to the siliciWed Xat irons noted above (Kappelman et al., 1996: Wg. 6.9). This new section is also predominantly normally magnetized, but several intervals of reversely magnetized sediments occur in its upper portion. It is important to note that the basal 115 m of this section thickness of 295 m total is normally magnetized, with only two sites near the top of this interval being reversely magnetized. We interpret this normal interval as representing Chron 5, and suspect that the two short duration events of reversed polarity may document some of the short duration reversed events (or ‘cryptochrons’ C5n.2n- 1 to C5n.2n-3 of Cande & Kent, 1995) that occur within this interval. The overlying sediments represent continuous section with no obvious breaks in sedimentation, and we interpret the Kavakdere section as continuing up to Chron C3Br.2r. The fossil localities are found near the top of this interval (see Fig. 12.3). The Igˇbek section is located about 40 km to the east of Kazan and presents a relatively short and isolated section of about 95 m in thickness.

Chronology and mammal faunas

Primary remnant magnetism was isolated in these sediments and permitted the construction of a paleomagnetic reversal stratigraphy. The results from Feseha (1996) are summarized here. The basal 45 m of this section are normally magnetized, while the upper 45 m are of predominantly reversed polarity with several intervals of normal polarity. Clearly, because this section is not very thick, several correlations with the GRTS are possible. The correlation that we favor ties the lowest long interval of normal polarity with C5n, and the upper interval of reversed polarity to Chron C4Ar. A series of channels at about 70 m in the section has an erosional base that cuts into and erodes the underlying sediments. Sediments directly below the base of this channel are reversely magnetized (see the brief reversed polarity interval at about 70 m) while those in and above the channel are of normal polarity. The presence of this erosional channel leads us to suspect that a greater duration of reversed polarity was once recorded here, and provides the support for a correlation that ties this reversed interval to Chron C4r.2r. The fossils from L.49 are found at the top of the section and occur in normally magnetized sediments which we interpret to be the base of Chron C4An. This is not, however, a unique correlation, but it seems unlikely that L.49 is any younger than Chron C4An. It could, however, be correlated with Chron C4Ar.1n. The radioisotopic date of 15.2 ± 0.3 Ma for the basalt Xow located at the base of the Sinap Formation is some 270 m below the contact between the Kavakdere and Beycedere Members. These dates provide a lower limit for the overlying paleomagnetic reversal stratigraphy of Kavakdere. The interval of time from 15 to about 12 Ma is one marked by numerous normal and reversed polarity events (C5Bn to C5An) of nearly equal duration that are followed by the relatively long duration reversed interval of C5r and normal interval of C5n. All of the sections from the Sinap Formation are marked by a long interval of normally magnetized sediments at their base, and we take this interval to represent Chron C5n. If sedimentation rates for the lowest sediments of the Kavakdere Member are similar to those of the Beycedere Member, then 270 m of the Beycedere Member would represent about three million years of time, which, when added to the hypothesized basal age of nearly 11 Ma for the Kavakdere Member, would total about 14 Ma for the base of the Beycedere Member. As noted above, these two members are separated by siliciWed Xat iron sediments, and we suspect that the formation of these Xat irons represents an unconformity of unknown duration. These calculations, along with the characteristic patterns of normal and reversed polarity from the GRTS, provide additional support for the correlation presented in Fig. 12.3. Additional support for the correlation is provided by the fauna as discussed below.

249

Miocene mammalian successions

250

Mammal succession

Materials and methods The Sinap collection is stored at the Museum of Anatolian Civilizations, Ankara. Most of the ranges reported here (Table 12.2) derive from the collection made during seven Weld seasons (1989–95) by the Sinap Project. The identiWcations in Table 12.2 are from the ongoing descriptive work by the following project members: small mammals (except Schizogalerix): Sevket Sen; Schizogalerix: Lena Sela¨nne; Primates: John Kappelman; Carnivora: Lars Werdelin and Suvi Viranta; Equidae: Raymond L. Bernor; Rhinocerotidae: Mikael Fortelius, Kurt Heissig, Gercek Sarac, and Sevket Sen; Suoidea: Jan van der Made and Mikael Fortelius; Ruminantia: Alan Gentry; Proboscidea: William J. Sanders; Tubulidentata: Mikael Fortelius, Sirpa Nummela, and Sevket Sen. Uncertain identiWcations are shown in Table 12.2 but were not used for the calculations. For the primates, rhinoceroses and suoids, some material collected prior to the Sinap Project has also been included. In these cases all uncertainties about provenance have been interpreted such that range extensions are minimized. Since the work is still in progress, the data are given and used here at the genus level. Lack of species-level data has prevented ecomorphological analysis of the faunas. A monograph publication is in progress and will provide a more detailed analysis (Fortelius et al., in prep.). The stratigraphic relationships of the Sinap localities are given in Table 12.2. The relationships between Sinap Project localities and previously known localities follow unpublishednotes by one of us (S. S.) and will be given elsewhere (Fortelius et al., in prep.). Locality 51, which is not in a measured section, was interpolated by minimizing the ranges of the taxa found at that locality. Locality 26 here includes several of the Upper Kavakdere localities that occur at the same level (26, 27, 32, 33, 61, 70). Locality 24 is situated close to the strike-slip fault southwest of the study area (see Fig. 12.1). The sediments at locality 24 represent hydrothermally altered mudstone most probably related to the tectonic activity that took place somewhat after the emplacement of the basaltic Xow around 15.2 Ma. Locality 42 is given a tentative MN 13 biochronologic correlation and an approximate age of 6 Ma. The methods used to quantify faunal dynamics mainly follow standard procedures (e.g., Fortelius et al., 1996). Since the small mammals known from the Sinap Formation have a highly clumped temporal distribution, they were excluded from most comparisons involving the entire sequence. Except for the Upper Kavakdere (‘locality 26’) we have not clumped localities together but have instead treated each locality as an ‘interval’ in the

Chronology and mammal faunas

sequence. Sampling completeness for a given interval was evaluated with a calculation of the completeness index (CI1) which is the ratio of recorded diversity (the sum of Wrst and last occurrences of taxa, taxa known only from the interval, and taxa known before, during and after the interval) to maximum diversity (the sum of recorded diversity, plus those taxa known before and after the interval but not during the interval (otherwise known as Lazarus taxa)) (Krause & Maas, 1990; Barry et al., 1995). This value is expressed as an index and sometimes as a percentage. Raw specimen counts at each interval were also used as an independent means for evaluating the eVects of sampling bias. Richness was studied as the raw recorded diversity and as maximum diversity with Lazarus taxa included, both at the genus level only. Turnover was studied with Lazarus taxa included. To assess the temporal inXuence on turnover statistics, a crude ‘duration’ for each locality was calculated as half of the interval between it and the preceding locality plus half of the interval between it and the succeeding locality. Entries, exits, and relative turnover (the sum of entries plus exits divided by richness) were studied both as such and normalized for this ‘duration’.

Sampling It is well known that the completeness index CI1 is a rather coarse estimator of actual completeness (Barry et al., 1995) and this statistic further suVers because it is not possible to calculate conWdence intervals. With these cautions in mind for the interpretations that follow, it is the case that CI1 Xuctuates in close harmony with specimen counts throughout the sequence as would be expected (Fig. 12.4a) and this tracking tentatively suggests that the Xuctuations describe actual variations in sampling. The completeness index CI1 Xuctuates from slightly over 10% to nearly 90% for all mammals (Table 12.3), with most values falling between 30% to 60%. If one uses the 70% upper cut-oV limit that is suggested by Maas et al. (1995), the Sinap localities are rather poorly sampled for completeness. By this criterion the only well sampled localities for all mammals are locality 49 (CI1 = 84.2%) and locality 26 (a lumped locality: CI1 = 86.7%). The index is indeWnable for the last locality in the series, locality 42, but this is also a well-sampled locality as indicated by its high specimen count (Table 12.3). If small mammals are excluded from the calculation, the results are quite similar with locality 49 (CI1 = 88.9%) and locality 26 (CI1 = 92.9%) joined by locality 12 (CI1 = 76.2%), but more localities shift into the 60–70% range. As noted above, the CI1 completeness index is only a coarse estimator of completeness. Barry et al. (1995) recommend the use of a second completeness index, CI2, that is the ratio of the number of taxa known before, during,

251

;

;

15?

Age (Ma)

Schizogalerix Byzantinia Myocricetodon Democricetodon Cricetulodon Peridyromys Myomimus Progonomys Pliospalax Spermophilinus Atlantoxerus Proochotona Ankarapithecus Hemicyon Sansanosmilus Pseudaelurus Miomachairodus Thalassictis Protictitherium Ictitherium Hyaenotherium ?Belbus Dinocrocuta Indarctos Anchiterium ‘Hipparion’

24

Locality

11.0

65

;

;

;

;

;

; ; ; ;

;

10.9

4

10.9

64

;

; ;

;

10.9

94

;

;

10.8

108

;

?

10.7

72

;

;

10.7

91

;

10.7

114

;

;

10.7

8B

Table 12.2. Occurrence and ranges of mammal genera in the Sinap sequence

;

; ; ;

; ;

;

; ;

; ;

;

; ;

; ; ;

;

;

10.4

84

; ;

;

10.6

12

; ; ;

10.7

8A

;

10.4

7

;

;

10.4

51

;

10.4

1

;

9.0

49 7.6

34

;

7.3

26

;

; ; ;

;

6?

42

15?

; ; ;

Age (Ma)

Brachypotherium Hoploaceratherium Begertherium Acerorhinus Chilotherium Stephanorhinus Ceratotherium Taucanamo Schizochoerus Kubanochoerus Bunolistriodon Listriodon Propotamochoerus Hippopotamodon Microstonyx Pecora indet. sp. Cervidae indet. Palaeotragus Bohlinia Decennatherium Helladotherium Giraffidae sp. Hypsodontus Turcocerus ?Protoryx Tragoportax

; ; ;

;

* * ;

*

24

Locality

Table 12.2 (cont.).

;

11.0

65

;

?

*

*

10.9

64

;

?

10.9

4

?

10.9

94

;

10.8

108

;

?

;

;

10.7

72

;

10.7

91

;

10.7

114

?

10.7

8B

*

*

10.7

8A

;

;

;

* ;

10.6

12 10.4

84 10.4

7

;

;

10.4

51

; ;

*?

*?

*

* ;

10.4

1

;

; ; ?

;

;

;

9.0

49

;

;

7.6

34

;

;

;

;

; ; ; ;

7.3

26

;

;

;

;

;

;

6?

42

15?

Age (Ma)

?

11.0

65

;

;

;

10.9

64

;

;

;

;

;

10.9

94

?

10.9

4

? ;

10.8

108

; ;

10.7

72

; ; ;

10.7

91 10.7

114

;

;

10.7

8B 10.7

8A

; ;

; ; ;

10.6

12 10.4

84

Lazarus taxa: Lazarus taxa (larger):

0 0

1 1

0 0

4 2 5 3 6 3 7 4

11 8

15 12

15 12

12 12

12 5

18 18

x = positive id. ? = tentative id. o = previous find (interpreted to minimize range if exact locality is not known). Loc. 42 ( = Cobanpinar) is given an ‘MN 13’ date on the basis of its fauna only. Loc. 51 is interpolated based on a rhino shared with Loc 1 and possibly with Loc 12, and to minimize range of Bohlinia.

Tethytragus Prostrapsiceros Sinapodorcas Palaeoreas Gazella ?Criotherium Protoryx Pachytragus Pseudotragus Bovidae smaller sp. Bovidae small sp. Bovidae sp. smaller than Gazella Bovidae sp. size of Gazella Deinotherium Gomphotherium ; Choerolophodon Orycteropus

24

Locality

Table 12.2 (cont.).

17 16

10.4

7

13 12

;

;

10.4

51

11 10

;

10.4

1

3 2

; ;

;

;

; ; ;

;

9.0

49

10 9

;

;

;

7.6

34

2 1

; ;

; ;

7.3

26

0 0

;

6?

42

Chronology and mammal faunas

255

[Figure 12.4] Sampling completeness in the Sinap sequence. (a) Specimen count and completeness index (CI1) are closely correlated. (b) Raw taxonomic richness (‘Raw’: the number of genera actually found at each data point ( = locality) is closely correlated with completeness index (CI1). When Lazarus taxa are added, the maximum richness curve is somewhat less sensitive to sampling variation at the individual localities.

and after the interval (NBDA) to the sum of NBDA plus the number of ‘range through’ (NRTT) or Lazarus taxa, and this index is arguably more powerful because one can calculate conWdence intervals (see Barry et al., 1995). The CI2 indices and 95% conWdence limits for the two most complete localities (all mammals) in the Sinap section are as follows: Locality 49: CI2 [10/(10 + 2)] = 83.3%; CL1 = 50.4%, CLu = 97.8% Locality 26: CI2 [6/(6 + 2)] = 75.0%; CL1 = 32.5%, CLu = 96.4%

The CI2 indices for both localities are lower than their CI1 indices, and the conWdence limits for these localities easily encompass the full range of CI1 indices for the other localities. When these results are compared with those from other studies, it should be kept in mind that the Sinap ‘intervals’ are, with the sole exception of locality 26, individual fossil localities, while the

Large mammals only Richness (incl. Lazarus taxa) Lazarus taxa Entries Exits Relative turnover Completeness Total specimen counts

0 2 5 5 4 1 1.29 0.75 80.0 153 335

0

138

8

7

14

8

0 4 12 5 7 1 1.46 0.55 73.3

11

0

13

94

108 72

91 114 8B 8A 12 84 7

51

3 1 4 0.56 75.0 288

9

5 1 4 0.42 70.6

12

3 2 1 0.43 70.0 210

7

6 2 1 0.30 62.5

10

4 4 0 0.40 71.4 175

10

7 4 0 0.31 65.0

13

8 4 1 0.36 63.6 135

14

11 4 1 0.29 60.7

17

12 1 0 0.07 53.8 56

16 12 1 2 0.19 57.1

12 1 0 0.07 55.6 116

12 6 3 0.38 66.7

24

15

15 1 0 0/.06 54.5

15 1 0 0/06 53.1

14

18

17

5 7 3 0.48 80.8 502

21

12 7 3 0.36 70.0

28

18 0 0 0.00 50.0

18

18 1 7 0.31 59.1

26

16 0 1 0.06 52.9 66

18

17 0 1 0/05 52.8

19

12 1 1 0.11 60.0 67

18

13 1 1 0.11 59.4

19

10.82 10.79 10.70 10.53 10.48 10.44 10.40 10.37 10.35 10.31 10.19 10.14a 10.1

4

13

15.2

Age All mammals Richness (incl. Lazarus taxa) Lazarus taxa: Entries Exits Relative turnover Completeness

64

14

24

Locality

Table 12.3. Sampling, richness, and turnover statistics of mammals from the Sinap sequence 49

10 0 1 0.06 63.0 46

17

11 0 1 0/06 62.1

18

2 2 5 0.39 90.0 1062

18

3 2 5 0.37 86.4

19

10.12a 9.0

1

9 1 3 0.29 60.9 23

14

10 1 3 0.27 60.0

15

7.60

34 42

12

11 1 0 3 3 6 0.64 93.3 490 515

14

2 0 3 3 6 0.60 88.2

15

7.30a 6?

26

Chronology and mammal faunas

Siwalik intervals of Barry et al. (1995) are summations of taxa from several individual fossil localities all occurring within a deWned 0.5 Ma duration. If a comparison is carried out with the well-sampled 0.5 Ma interval at Sinap dating from 10.75 Ma to 10.25 Ma (taxa from localities 94, 108, 72, 91, 114, 8B, 8A, and 12), the CI2 is now at 100% but because of small sample size (NBDA = 4, NRT = 0), the conWdence limits remain quite wide (55% to 100%). Richness oVers another useful comparison, and raw richness (without Lazarus taxa) Xuctuates in close harmony with completeness (Fig. 12.4b). The inclusion of Lazarus taxa in the calculation of maximum richness decouples this correlation to some extent, as would be expected (Fig. 12.4b).

Taxonomic richness The oldest locality in the Sinap sequence, locality 24 (= Yno¨nu ¨ I), has a moderate raw generic richness for all mammals (14 genera) at 15.2 Ma, and this value is virtually identical to the richness found at the Wrst post-hiatus locality (locality 64) at 10.82 Ma (Fig. 12.5a). Raw richness then descends over an interval of increasingly poorly sampled localities until it rises as a sharp spike at the cluster of rich localities around 10.3 Ma (8B, 8A, 12 and 84). After this spike the raw richness remains low until the well-sampled locality 49 at approximately 9 Ma, after which it seems to begin a slow decline but this possible trend is represented by relatively few localities. When Lazarus taxa are added for the calculation of maximum richness, the spike at around 10.3 Ma is even more pronounced but otherwise the pattern is repeated at a higher level and with somewhat less Xuctuation. When small mammals are excluded from the analysis, the diversity spike is much more subdued, especially when Lazarus taxa are included (Fig. 12.5b). Locality 12 still has the highest taxon richness, and the declining trend during the rest of the sequence is still indicated.

Entries, exits and relative turnover Because of the clumped occurrences of small mammals, turnover statistics were only studied for large mammals (Fig. 12.6). There is much Xuctuation, and the majority of it is probably related to sampling eVects, with the minor turnover peak at locality 12 most likely reXecting the intense sampling at this locality. Normalizing the turnover statistics for ‘duration’ of the data points (as described under Materials and Methods above) is shown in Fig. 12.7. It is likely that the major exit peak at locality 8A (10.4 Ma) is due to a combination of the low specimen count (n = 35) and raw richness (n = 4) at this locality, while the intense excavations at locality 12 are again probably

257

Miocene mammalian successions

258

[Figure 12.5] Taxonomic richness in the Sinap sequence. The lower curve of raw richness shows the number of genera actually found at each data point ( = locality) while the upper curve is maximum richness and has Lazarus taxa added. (a) All mammals. (b) Large mammals only.

responsible for the entry peak at 10.3 Ma. It is conceivable that the simplistic notion of ‘duration’ applied here overcompensates for sampling interval, and that the true turnover trend lies somewhere between the versions shown by Figs. 12.6 and 12.7.

Discussion and conclusions As a whole the sedimentary sequence in the study area is rather complex. Tectonic activity during the Neogene in Central Anatolia resulted in a sedimentological regime that was characterized by local changes aVecting deposition in diVerent ways in each individual basin. A consequence of this pattern is that each basin has to be analysed individually and that there is little hope of inter-basin correlation based solely on changes in lithology. In

Chronology and mammal faunas

259

[Figure 12.6] Large mammal taxonomic richness and turnover in the Sinap sequence calculated from a matrix with Lazarus taxa included. The effect of sampling is probably responsible for some of the spikes.

[Figure 12.7] Large mammal turnover in the Sinap sequence as normalized for ‘duration’ of data points as explained in the text. The straight line is the least squares regression line of the turnover value (see the vertical axis on the right side of the plot) on the locality sequence, and has no quantitative significance beyond indicating a general trend of declining turnover values.

other words, while the evolution of each basin may be described and the changes in lithology related to changes in the individual fossil assemblages, correlations between even adjacent basins have to rely not only on lithostratigraphy but also on magnetostratigraphy and biochronology. The sedimentary sequence represented by the Lower and Middle Sinap Members in the Sinap Tepe area, and the Beycedere, Kavakdere and C ¸ alta Members north of the study area is highly complete. It is assumed that the Lower Sinap Member is at least in part comparable to the upper portion of

Miocene mammalian successions

260

the Beycedere Member although thereafter the individual basins on both sides of the fault zones were only weakly interconnected. The sediments in the Sinap Tepe area (Lower and Middle Sinap Members) represent more or less continuous sedimentation in a relatively low energy environment (Xood plain and distal alluvial fan) for approximately one million years while Xuvial and lacustrine sediments in the Beycedere and Kavakdere area (the upper part of the Beycedere Member and Kavakdere Member) cover more than four million years of time. Regarding the occurrence of the fossil accumulations, it is interesting to note that the densest fossil accumulations have been discovered in the lower part of the Middle Sinap Member. For a time period of about 300 kyr this area was in the depositional regime where the distal alluvial fan accumulation interacted with the more distal Xood plain deposition. Conditions in this environmental setting appear to have been the most suitable for the preservation of bone material. The situation for bone preservation in the proximal alluvial fan environment (the Yellidoruk Member) and in the area close to the trunk river (Igˇbek section and the upper part of the Middle Sinap Member) was apparently not so favorable. In these areas erosional events appear to have been highly energetic and must have prevented the accumulation of fossil-rich sites. As described above, local tectonic factors largely determined the mode and pattern of sedimentation during the time when the Sinap Formation sediments were laid down. However, in order to indicate wider-scale environmental changes during the time period represented by the Sinap Formation, we are undertaking isotopic analysis of carbonate nodules collected from paleosol horizons in various levels of the Sinap Formation sediments (Lunkka & Ekart, in prep.). Similarly, taphonomic and paleoecological results concerning the mammal fauna are also in progress. The results from ecomorphological studies appear to be relatively robust with respect to sampling and taphonomic biases (see Fortelius et al., 1996) and may correct some the deWciencies that have come to light from the analyses of completeness statistics presented here. The genus richness curves (Fig. 12.5) are clearly strongly inXuenced by sampling but even so there appears to have been little or no net change in diversity during the 10 Ma period sampled by these sediments. There is weak evidence of a minor diversity peak around 10.3 Ma in that better sampled, later localities have lower richness but this diVerence could also be due to intense sampling eVorts at the 10.3 Ma level. It may prove to be the case that the middle / late Miocene boundary and earliest Late Miocene might genuinely have been a time of low taxonomic richness. As the range

Chronology and mammal faunas

chart (Table 12.2) shows, typical Late Miocene ungulates were appearing during this interval and a classic ‘open woodland’ ungulate fauna with genera like Acerorhinus, Chilotherium, Ceratotherium, Hippopotamodon / Microstonyx, Bohlinia, Tragoportax, Prostrepsiceros and Palaeoreas seems to have been in place by the well sampled interval around 10.3 Ma. Only two cervid specimens have been recovered from the Middle Sinap Formation and all suoids are rare, in strong contrast to the typical Vallesian faunas of western and central Europe (West). The murid Progonomys, often taken to indicate the beginning of MN 10 in West appears at 10.4 Ma, nearly a million years earlier than recent calibrations of the MN 9 / 10 boundary (Steininger et al., 1996; Krijgsman et al., 1996). Thus there is a general impression of precocious development of the Sinap mammal fauna compared with that of West. Interestingly, there is no indication at all at Sinap of the explosive diversity increase seen for the East bloc in the regional compilation of Fortelius et al. (1996). If anything, richness seems to have decreased from about 10.3 Ma onwards. In terms of its diversity history, Sinap seems to be something of a hybrid between East and West, showing neither the dramatic diversity crash at 9.5 Ma nor the massive diversity increase continuing until at least 8 Ma. This pattern lends some support to the suggestion that the exceptionally high Late Miocene diversity of East may be partly a result of lumping vast areas with regionally diVerent faunas. On the other hand, it is obvious that the Turolian localities of Upper Kavakdere or C ¸ obanpinar are not comparable to super-rich occurrences like Pikermi or Samos, and that the lack of a diversity increase at Sinap may simply reXect the lack of super-normal sampling. The suggestion of decreasing turnover in the data from Sinap provides limited support for the emergence during the Vallesian of a stable, wideranging open woodland mammal fauna with similarity to that of western Eurasia from MN 9 to MN 12 (Fortelius et al., 1996). It must be left for species-level studies to answer the question as to whether or not the Xuctuations hinted at in the turnover values can be regarded as signiWcant. The occurrence at Sinap of the hominoid Ankarapithecus within the diversity spike at 10.3 Ma probably reXects the intense sampling of this interval. The absence of hominoid primates from the well-sampled and rich locality 49 and later localities may indicate that they, like their western relatives, had become extinct by 9.0 Ma. There is no indication of a process comparable to the Vallesian crisis, however, nor any evidence that anything similar to the ‘supersaturated’ forest faunas of pre-crisis West ever occurred at Sinap.

261

Miocene mammalian successions

262

Acknowledgements The Sinap Project (1989–95) enjoyed the eVorts of a large number of colleagues and friends, and the results summarized here represent the collective work of the entire project. We wish to thank Professor Dr Berna Alpagut with whom we initiated this project, and I. Temizsoy, Director, Museum of Anatolian Civilizations, who held the excavation permits and greatly facilitated our eVorts in the Weld. We would also like to express our gratitude to Aileen Davis, JeV Crabaugh, Douglas Ekart, and Phil Gibbard who took part in the geological Weldwork. Paleomagnetic Weld lab work was assisted by A. Duncan, M. Feseha, W. Gose, P. Hake, D. Johnson, and P. StubbleWeld, and mammal identiWcations by Lena Sela¨nne, Lars Werdelin, Suvi Viranta, Raymond L. Bernor, Kurt Heissig, Gercek Sarac, Jan van der Made, Alan Gentry, William J. Sanders, and Sirpa Nummela. Thanks are also due to John Barry and Robert Scott for their comments on an earlier version of this manuscript. The Weld and laboratory work were funded by the following grants: US National Science Foundation (EAR 9304302), the L.S.B. Leakey Foundation, and the Academy of Finland (Project 34080). The research was supported by the general Directorate of Antiquities, T.C. Ministry of Culture and Tourism, and we would like to thank them for their seven years of support. We would also like to thank the European Science Foundation and J. Agustı´ for inviting us to participate in the Sant Feliu de Guixols conference and to contribute to this volume.

References Agustı´, J. & Moya`-Sola`, S. 1990. Mammal extinctions in the Vallesian (Upper Miocene). In Extinctions Events in Earth History, KauVman, E. G. & Walliser, O. H. (eds.). Lecture Notes in Earth Sciences, 30, 425–32. Springer-Verlag, Berlin-New York. Angelier, R., Dumont, J. J. F., Karamandersei, H., Poisson, A., Sinsek, S. & Uys, S. 1981. Analyses of fault mechanisms and expansion of southwestern Anatolia since the Late Miocene. Tectonophysics, 75, T1–T9. Barry, J. C., Morgan, M. E., Flynn, L. J., Pilbeam, D., Jacobs, L. L., Lindsay, E. H., Raza, S.M. & Solounias, N. 1995. Patterns of faunal turnover and diversity in the Neogene Siwaliks of Northern Pakistan. Palaeogeography, Palaeoclimatology, Palaeoecology, 115, 209–26. Cande, S. C. & Kent, D. V. 1995. Revised calibration of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 100, B4, 6093–5. Erol, O. 1981. Neotectonic and geomorphological evolution of Turkey. Zeitschrift fu ¨r Geomorphologie, Neue Folge, Supplement Band 40, 193–211. Feseha, M. 1996. Magnetostratigraphy of the Fossil-Bearing Igˇbek Section of the Miocene Sinap Formation. M.A. Thesis, The University of Texas at Austin.

Chronology and mammal faunas

Fortelius, M., Werdelin, L., Andrews, P., Bernor, R. L., Gentry, A., Humphrey, L., Mittmann, H.-W. & Viranta, S. 1996. Provinciality, diversity, turnover and paleoecology in land mammal faunas of the later Miocene of Western Eurasia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-V. (eds.), pp. 414–48. Columbia University Press, New York. Fortelius, M., Kappelman, J., Sen, S. & Bernor, R. L. (eds.). The Miocene Sinap Formation of Central Turkey. Columbia University Press. (In preparation.) Inci, U. 1991. Miocene alluvial fan-alkaline playa lignite-trona bearing deposits from an inverted basin in Anatolia: sedimentology and tectonic controls of deposition. Sedimentary Geology, 71, 73–97. Kappelman, J., Sen, S., Fortelius, M., Duncan, A., Alpagut, B., Crabaugh, J., Gentry, A., Lunkka, J. P., McDowell, F., Solounias, N., Viranta, S. & Werdelin, L. 1996. Chronology and Biostratigraphy of the Miocene Sinap Formation of Central Turkey. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittman, H.-W. (eds.), pp. 78–95. Columbia University Press, New York. Koc¸yigˇit, A. 1991. An example of an accretionary basin from northern Central Anatolia; its implications for the history of subduction of Neo-Tethys in Turkey. Geological Society of America Bulletin, 103, 22–36. Krause, D. W. & Maas, M. C. 1990. The biogeographic origins of late Paleocene–early Eocene mammalian immigrants to the western interior of North America. Geological Society of America Special Paper 243, 71–105. Krijgsman, W., Garce´s, M., Langereis, C. G., Daams, R., Van Dam, J., Van der Meulen, A. J., Agustı´, J. & Cabrera, L. 1996. A new chronology for the middle to late Miocene continental record in Spain. Earth and Planetary Science Letters, 142, 367–80. Lunkka, J. P. & Ekart, D. Sedimentology of the Sinap Formation. Columbia University Press. (In preparation.) Lunkka, J. P., Kappelman, J., Ekart, D. & Sen, S. 1998. The Pliocene vertebrate locality of C ¸ alta, Ankara, Turkey. 1. Sedimentation and Lithostratigraphy, Sen, S. (ed.). Geodiversitas, 20(3), 329–38. Lu ¨ ttig, G. & SteVens, P. 1976. Explanatory Notes for the Paleogeographic Atlas of Turkey from the Oligocene to the Pleistocene. Bundesanstalt fu ¨ r Geowissenschaften und RohstoVe. Hannover, 64 pp. Maas, M. C., Anthony, M. R. L., Gingerich, P. D., Gunnell, G. F. & Krause, D. W. 1995. Mammalian generic diversity and turnover in the late Paleocene and early Eocene of the Bighorn and Crazy Mountains basins, Wyoming and Montana (U. S. A.). Palaeogeography, Palaeoclimatology, Palaeoecology, 115, 181, 207. Miall, A. D. 1978. Lithofacies types and vertical proWle models in braided river deposits: a summary. In Fluvial Sedimentology, Miall, A. D. (ed.), pp. 597–604. Canadian Society of Petroleum Geology Memoir. ¨ ngu O ¨ r, T. 1976. Kizilcahamam, Camlidere, Celtikci ve Kazan dolayinin jeoloji durumu ve jeotermal enerji olanakrari. Unpubl. rept., MTA, Ankara. Ozansoy, F. 1957. Faunes de Mammife`res du Tertiaire de Turquie et leurs re´visions stratigraphiques. Bull. Min. Res. Expl. Inst. Turkey (Foreign Ed.), 49, 29–48. Ozansoy, F. 1965. E´tude des gisements continentaux et de mammife`res du Ce´nozoique de Turquie. Mem. Soc. Ge´ol. France (N.S.), 44, 1–92.

263

Miocene mammalian successions

264

Sen, S. 1991. Stratigraphie, faunes de mammife`res et magne´tostratigraphie du Ne´oge`ne de Sinap Tepe, province d’Ankara, Turquie. Bulletin du Museum National d’Histoire Naturelle, Paris, 4e se´ries, 12, 243–77. S¸engo¨r, A. M. C. & Yilmaz, Y. 1981. Tethyan evolution of Turkey: A plate tectonic approach. Tectonophysics, 75, 181–241. Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Miocene-Pliocene) marine-continental chronologic correlations of European mammal units. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 7–46. Columbia University Press, New York.

13 The Late Miocene small mammal succession in Ukraine Valentin A. Nesin and Vadim A. Topachevsky

Introduction Taxonomic analysis of widespread groups of small-sized mammals in Eurasia (with special reference to Cricetidae, Arvicolidae, Muridae, Spalacidae and Dipodidae of rodents, and Palaeolaginae, Ochotonidae and Leporidae of lagomorphs, as well as some insectivores, from Ukrainian localities) has characterized known mammalian complexes and allocated several new ones from Middle Sarmatian up to Early Pleistocene (Table 13.1). This period of time encompasses about 10 million years. An important feature of most localities is the fact that continental deposits are enclosed within marine layers of the regional stages of Sarmatian, Meotian and Pontian. Their Wxed position in the conventional stage scheme is a reliable basis for their correlation with the continental scale MN zones. It is a common practice in Eastern Europe to arrange the periodicity of development of faunas by means of faunistic complexes. Faunas of the Late Miocene are the least investigated, and about 10 Late Miocene sites have been analysed here. This has resulted in us recognizing six Vallesian– Turolian complexes based on small mammal distributions.

Grytsivsky mammalian complex Fossil remains of the fauna are found in karst deposits near the villages of Grytsiv and Klymentovychi in the Khmelnytsky region (see Fig. 13.1). The sediments are precisely dated as Novomoskovsky layers of Bessarabian. That corresponds to a bottom horizon of Middle Sarmatian (Topachevsky et al., 1996). About 100 species of fossil vertebrates are found here. The systematic composition of small-sized mammals is submitted as follows: Insectivora: Schizogalerix sp., Lanthanotherium sp., Amphechinus sp., Domninoides sp., Proscapanus sp., Urotrichini gen et sp.1, Urotrichini gen et sp.2, Plesiodimylus sp., Dinosorex sp., Crusafontina sp., Episoriculus sp.; Lagomorpha: Amphilagus sarmaticus I. Topachevski, 1991; Rodentia: Miopetaurista sp., Blackia sp., Forsythia sp., Sciurotamias (= Spermophilinus) sp., Monosaulax sp., Palaeomys sp., SteneoWber (?) sp., Eomyops sp., Keramidomys sp., Glis vallesiensis Agustı´, 1981, Muscardinus topachevskii sp.nov., Myoglis ucrainicus sp.nov., Paraglirulus cf. werenfelsi Engesser, 1972, Miodyromys griciviensis sp. nov., Anomalomys sp., Lophocricetus sp.,

Miocene mammalian successions

266

[Figure 13.1] Miocene localities of small mammal faunas in Ukraine and adjacent areas.

Cricetulodon complicidens Topachevski & Scorik, 1992, Sarmatomys podolicus Topachevski & Scorik, 1988. Obviously, pre-Vallesian forms dominate the list, mainly of Astaracian origin. The dominant elements are last representatives of Dimylidae, Lanthanotherium among the Erinaceidae and Dinosorex among the Heterosoricinae (which is close to ancient D. zapfei). Among the Palaeolaginae there is Amphilagus sarmaticus. Also among the Rodentia this level records the last Miopetaurista and Eomyops, the last Myoglis, Paraglirulus and Miodyromys among Gliridae, as well as the last Microtocricetini – Sarmatomys podolicus. A Vallesian age is estimated for this assemblage because of the occurrence of Crusafontina and Episoriculus among the Soricidae, the Wrst Lophocricetus and the Wrst Cricetulodon among the Cricetidae. At the same time, the Grytsiv fauna is characterized by the absence of the Muridae among the Rodentia and the Ochotonidae among the Lagomorpha. The Vallesian age of fauna of the locality is precisely conWrmed also by fossil remains of Hipparion cf. primigenium and small-sized Hipparion sp. Thus, the Grytsiv fauna seems to be one of the most ancient Vallesian communities within the MN 9 zone. It is worth noticing that the typical Vallesian locality Can Llobateres, in which Muridae already occur, is later than Grytsiv as well. The palaeomagnetic data do not contradict this conclusion. Most of the mammals from Grytsiv are inhabitants of warm, damp forest ecosystems. As palaeoenviromental indicators, Petauristinae, Gliridae,

Late Miocene small mammal succession

Sciurotamias and some other taxa hold the most promise. Moreover, many representatives of Amphibia and Reptilia from Grytsiv can be regarded as inhabitants of subtropical and even tropical ecosystems (Korotkevich, 1988).

Kalfynsky mammalian complex The typical locality is Kalfa in Moldova (Lungu, 1981). Faunas of this complex are not known from the Ukraine territory. It corresponds to the Middle Sarmatian and coincides with the Vasilievsky horizon of Bessarabian. The mammal fauna of Kalfa is assigned to the Late Vallesian in the continental scale. It probably corresponds to the level of Musia del Barbo, MN 10 zone, of the Tethys area. In contrast to the previous complex, the dominant groups of this fauna are rodents and lagomorphs. At the same time, although Insectivora are considerably reduced in number, they did not change their taxonomic composition. The genera Hispanomys and Anomalomys dominate. Among Muridae the Wrst Progonomys occur. Among Lagomorpha there is the Wrst occurrence of Ochotonidae represented by Proochotona. The Microtocricetini disappear.

Mykhailivsky mammalian complex This complex is timed to the contact zone between Middle and Upper Sarmatian, and more precisely belongs to the upper horizon of Middle Sarmatian. The typical locality is Mykhailivka-1 located near Mykhailivka village of the Mykolaivsky region. The faunal changes demonstrate arid climates. The systematic composition of small-sized mammals is as follows: Insectivora: Galerix sp., Mygalinia sp., Desmana sp., Talpidae gen. et sp., Amblycoptus sp.; Lagomorpha: Amphilagus sp., Proochotona sp.; Rodentia; Stylocricetus sp., Ischymomys quadriradicatus Zazhygin, 1977, Parapodemus cf. lugdunensis Schaub, 1938. Advanced arvicolid-like Cricetidae (Ischymomys) accompanied by the Wrst Cricetini (Stylocricetus) become abundant, with Lagomorphs and rodents the most abundant groups (Fig. 13.2). Latest Eastern European Palaeolaginae (Amphilagus) prevail in the faunal structure. Among Ochotonidae Proochotona is still present. Among Rodentia, Cricetidae (Ischymomys) and the earliest true Cricetini (Stylocricetus) are background groups. A few Wrst representatives of the murid Parapodemus occur. The

267

Miocene mammalian successions

268

[Figure 13.2] Ischimomys ponticus. 1: right upper M 1; 2: right upper M 2; 3: right upper M 3; 4: left lower M 1; 5: left lower M 2; 6: left lower M 3.

numbers of insectivore taxa is low. For the Wrst time the Desmaninae appear and dominate. They are represented by two parallel lines – Mygalinia and Desmana. Fossils of large mammals are not known from Mykhailivka. This complex corresponds to the terminal stage of the MN 10 zone (Uppermost Vallesian of the continental scale).

Beryslavsky mammalian complex An Upper Sarmatian age is estimated for this complex. The most signiWcant localities are presented in Table 13.1. The systematic composition of smallsized mammals is as follows: Insectivora: Amblycoptus sp., Sulimskia sp.; Lagomorpha: Proochotona sp., Prolagus sp., Veterilepus sp.; Rodentia: Eozapus sp., Lophocricetus complicidens Topachevsky & Scorik, 1984,

Late Miocene small mammal succession

Table 13.1. Small mammal faunal complexes and localities from the northern part of Eastern Paratethys

269

Miocene mammalian successions

270

Lophocricetus sarmaticus Topachevsky & Scorik, 1984, Kowalskia progressa Topachevsky & Scorik, 1992, Ischymomys sp., Parapodemus lugdunensis Schaub, 1938, Parapodemus gaudryi (Dames, 1883). At its early stage, the Mykhailivka association is dominated by Ischymomys, while the Palaeolaginae are absent from this level. Among the Muridae, Parapodemus lugdunensis persists from the previous community. The sharp shifts in the communities of small-sized mammals are Wxed at the late stage of the Grebenykivsky subcomplex. Ischymomys from the previous communities are replaced by the most ancient cricetines of the genus Kowalskia as dominating rodents. Muridae (Parapodemus) become the second dominant group, here represented by the species P. gaudryi. The most ancient Dipodidae (Lophocricetinae) and earliest Zapodidae also acquire rather wide representation. Among Lagomorpha, this level records the Wrst occurrence of Leporidae (Veterilepus) as well as the Ochotonidae of the genus Prolagus. The comparison of this complex with similar mammalian faunas of Tethys and Central Paratethys is very complicated owing to their zonal diVerences, but an approximate correlation is possible with the small mammal fauna of KoWdisch and Dorn-Durkheim. The most characteristic localities of the complex are Mykhailivka-2 and Novoelyzavetivka-2, both referred to the MN 11 zone.

Bilkynsky mammalian complex This complex is referred to the Lower and Middle Meotian. The most signiWcant localities are presented in Table 13.1. The systematic composition of small-sized mammals is as follows: Insectivora: Galerix sp., Desmana sp., Talpidae gen. et sp., Miosorex sp., Amblicoptus sp.; Lagomorpha: Prolagus sp., Veterilepus lascarevi (Chomenko, 1914); Rodentia: Petauristinae gen. et sp., Sciurotamias sp., Monosaulax sp., Muscardinus sp. nov., Gliridae gen. et sp., Eozapus sp., Lophocricetus maeoticus Topachevsky & Scorik, 1984, Nannospalax sp., Byzantinia sp., Kowalskia cf. falbuschi Bachmayer & Wilson, 1970, Pseudocricetus antiquus Topachevsky & Scorik, 1992, Pseudocricetus orienteuropaeus Topachevsky & Scorik, 1992, Stylocricetus meoticus Topachevsky & Scorik, 1992, Microtoscoptes sp., Parapodemus gaudryi (Dames, 1883). The basic taxonomic composition of the previous community is in general retained, but there are essential changes among the Cricetidae. Kowalskia is replaced by Pseudocricetus, and this genus and Parapodemus gaudryi among Rodentia and Prolagus among Lagomorpha are characteristic elements of this stage of the Meotian. Distinction between initial and latest

Late Miocene small mammal succession

stages is based on speciWc changes in the genera Pseudocricetus and Lophocricetus. Microtoscoptes appear for the Wrst time in the early stages of the complex. Later stages of the complex are characterized by the presence of Spalacidae. In the Tethys and Central Paratethys zones the above-mentioned communities can be correlated with those of the MN 12 zone.

Cherevychansky mammalian complex This complex corresponds to the end of the Meotian phase. Basic localities are presented in Table 13.1. A typical locality is Novoukrainka-2 in the Odessa region. The systematic composition of small-sized mammals is as follows: Insectivora: Soricidae gen. et sp.; Lagomorpha: Prolagus sp., Veterilepus hungaricus; Rodentia: Gliridae gen. et sp., Nannospalax compositodontus Topachevsky, 1971, Prospalax sp., Pseudocricetus kormosi Schaub, 1930, Microlophiomys vorontsovi Topachevsky & Scorik 1984, Pseudomeriones abbreviatus (Teilhard, 1926), Occitanomys sp., ?Apodemus sp. This complex records a radical reorganization of the murid association. The Parapodemus background changes for that of Apodemus. Some increase in the number of Spalacidae that is chieXy presented by Nannospalax is observed as well. At the same time, remains of more primitive Prospalax are known. At this stratigraphic level the occurrence of the Wrst Pseudomeriones is also Wxed. Besides that, in the early stages of the complex Lophiomyidae appear for the Wrst time, a family that is of African origin. Based on its rodent content, this complex has elements in common with some localities of the Tethys and Central Paratethys areas such as Crevillente 6, Polgardy and probably Baltavar. It can be therefore assigned to the MN 13 zone. Palaeogeographic analysis of the Meotian communities shows that their representatives were basically inhabitants of savannas or foreststeppes diVering from similar recent ecosystems by the abundance of coniferous vegetation. It is already known that Ukraine was the territory of migration of African and Asiatic faunae to Eastern Europe during the MioPliocene. Hence their further study may help to solve many palaeontological problems.

References Korotkevich, E. L. 1988. A history of formation Hipparion fauna of East Europe. Kiev, Naukova dumka, 164 pp.

271

Miocene mammalian successions

272

Lungu, A. N. 1981. The hipparion fauna Middle Sarmatian of Moldavia (Insectivora, Lagomorpha and Rodentia). Kishineu, Shtiinza, 140 pp. Topachevsky, V. A., Nesin, V. A., Topachevsky, I. V. & Semenov, Yu. A. 1996. The most ancient locality of the Middle Sarmatian small mammals fauna (Insectivora, Lagomorpha, Rodentia) in the East Europe. Dopovidi NAN Ukraine, N2, 107–110.

PART III

Palaeoenvironments: non-mammalian evidence

14 Marine invertebrate (chiefly foraminiferal) evidence for the palaeogeography of the Oligocene–Miocene of western Eurasia, and consequences for terrestrial vertebrate migration Robert Wynn Jones

Introduction Global palaeoclimate Xuctuated cyclically during the Oligocene–Holocene, generally (though never irreversibly) deteriorating until the Pleistocene, which was characterised by widespread glaciation. Eurasian palaeogeography was modiWed not only by associated (climatically induced) cyclic Xuctuations in sea-level (glacio-eustasy) but also by mountain-building (tectonism) associated with the convergence between Arabia and Eurasia and the closure of the formerly intervening ocean known, after the daughter of the mythological earth-goddess Gaia, as Tethys (Suess, 1893). While Tethys, also known as the Tethyan Ocean or Seaway, was in existence, marine faunas were able to migrate between the IndoPaciWc and Mediterranean, and when it ceased to exist, terrestrial faunas were able to migrate between Africa and Eurasia.

[Figure 14.1] Stratigraphic correlation. Ages in Ma, magnetostratigraphy, chronostratigraphy, planktonic foraminiferal biostratigraphy (P–N zones) and calcareous nannoplankton biostratigraphy (NP–NN zones) after Berggren et al. (1995). Vertebrate biostratigraphy (MN zones and Agenian–Villafranchian stages) principally after Ro¨gl (1996a,b; in press; and this volume) and Steininger et al. (1996). Larger benthonic foraminiferal biostratigraphy (Indo-Pacific Letter Stages Ta-Tgh) principally after Adams (1970, 1984) and Wonders & Adams (1991). Tb equivalent to SB20 of Cahuzac & Poignant (1997), Tc to SB21, Td to SB22, Tel-4 to SB23, Te5 to SB24, Tf to SB25–26. Indian and Pakistani (litho) stratigraphy after Flynn et al. (1990), Retallack (1991) and Pilbeam et al. (1996) (see also Bossart & Ottiger (1989) and Jones (1997)). Iranian (litho) stratigraphy after Stocklin & Setudehnia (1972), Jones & Racey (1994) and Goff et al. (1995). Paratethyan (litho) stratigraphy modified as appropriate after Jones & Simmons (1996). Regional sequence stratigraphy modified as appropriate after Chepalyga (1985): Sppt = salinity parts per thousand (inferred from palaeontological analyses (curve indicates extent of open marine connection)). Global sequence stratigraphy (coastal onlap curve) modified as appropriate after Haq et al. (1988): numbers refer to original absolute ages assigned to sequence boundaries and maximum flooding surfaces. Regional climatostratigraphy after Zubakov & Borzenkova (1990): numbers refer to superclimathems (SCTs). Regional (Mediterranean) palaeotemperature curve modified as appropriate after Demarcq (1985) (biotic). Global (low-latitude Pacific) palaeotemperature curve after Haq (1991) (isotopic). Calibration best-fit. *1 = Recent and Ancient Caspian Deposits; *2 = Mishan (Middle Fars); *3 = Gachsaran (Lower Fars); U.R.F. = Upper Red Formation; L.R.F. = Lower Red Formation.

Palaeoenvironments: non-mammalian evidence

276

This paper attempts to collate and integrate all the available marine invertebrate (chieXy foraminiferal) and associated evidence pertinent to the stratigraphy and palaeogeography of Western Eurasia over the critical Oligocene–Miocene interval, to describe the palaeogeographic evolution of the area during this time in the form of a series of time-slice maps, and to consider the consequences for terrestrial vertebrate migration.

Stratigraphy The best-Wt stratigraphic correlation that forms the basis for the time-slice identiWcation and palaeogeographic mapping of the present paper is given in Fig. 14.1. Chronostratigraphy (global (Mediterranean) and regional (IndoPaciWc) stages), biostratigraphy (planktonic foraminiferal, calcareous nannoplankton and mammalian zones (and mammalian stages)), magnetostratigraphy, regional (Indian/Pakistani, Iranian and Paratethyan) lithostratigraphy, regional (Paratethyan) and global sequence stratigraphy, climatostratigraphy and palaeotemperature are all shown. For the sake of consistency, I have attempted wherever possible to use biostratigraphy for the purposes of age assignment and stratigraphic correlation. Inconsistencies can easily arise in the injudicious use of, for instance, absolute ages (which can vary according to analytical technique or precision) or stage names (whose calibration can vary from one time-scale to another).

Timing of the closure of Tethys Previous workers have disagreed about the precise timing of the closure of the Tethyan Seaway linking the Indo-PaciWc and Mediterranean. To summarise, Drooger (1979) inferred an end Oligocene, Steininger & Ro¨gl (1979), 1984), Ro¨gl & Steininger (1983, 1984), Steininger et al. (1985a,b), and Ro¨gl (in press, and this volume) (see also Bernor et al., 1987) a Middle Miocene, and Adams et al. (1983, 1990), and Adams et al. (in press) (see also Ali, 1983, and McCall et al., 1994) an Early Miocene age-date. Adams et al. (in press) challenged both Drooger’s and Steininger & Ro¨gl’s conclusions on the basis that they relied on imprecise or unveriWable agedating. Drooger’s conclusion depended on the dating of the separate evolution of Indo-PaciWc and Mediterranean lineages of lepidocyclinid and miogypsinid larger benthonic (bottom-dwelling) foraminifera (Chattian according to his interpretation, at least as young as Burdigalian according to my interpretation). Steininger & Ro¨gl’s conclusion depended in part on dating of certain sediments in the Middle East and Paratethys, which was

Marine invertebrate evidence

not directly veriWable by reference to illustrations of key fossils. It also depended in part on the correct interpretation of the palaeogeographic signiWcance of key fossils, which was not always unequivocal. I believe that the Tethyan Seaway was always open before the end of the Oligocene, and always closed after the end of the Middle Miocene. I also believe that the Tethyan Seaway was sometimes (at times of relative highstand of sea-level) open, and sometimes (at times of relative low-stand) closed during the Early and Middle Miocene. In other words, there were a series of closures (related to glacio-eustasy and tectonism) rather than a single instantaneous event. There is some published palaeontological evidence, though no incontrovertible proof, to support a marine connection from the Indo-PaciWc by way of the Mesopotamian Basin (North-Eastern Iraq) into the Mediterranean (Maritime Levant) in the Middle Miocene (see also Ponikarov, 1967, and Buchbinder, 1996). A marine connection as far as North-Eastern Iraq in the Middle Miocene, Langhian–Serravallian is implied by the records in the Jeribe and Lower– Middle Fars Formations of the planktonic (free-Xoating) foraminifer Orbulina universa (no older than Langhian, global standard planktonic foraminiferal zone N9) and of unspeciWed Serravallian (N13) planktonic foraminifera respectively (Ctyroky et al., 1975; Karim, 1978; Prazak, 1978). While it is acknowledged, as it was forcefully argued by Adams et al. (1983), that the identiWcations of the planktonic foraminifera forming the bases of the age assignments require veriWcation by reference to specimens or photographs, it should be stated that Orbulina universa is readily identiWable in matrix-free material, and as such is unlikely to have been misidentiWed (in thin-section, there is more potential for misidentiWcation). A marine connection as far as the Misis-Andirin area of the Iskenderun Sub-Basin of south-eastern Turkey in the Middle Miocene, Langhian–Serravallian is implied by the occurrences in the Karatas and Kizildere Formations of the planktonic foraminifera Globorotalia (Fohsella) peripheroronda (no younger than Serravallian, global standard planktonic foraminiferal zone N11), Praeorbulina spp. (Langhian, N8–N9), Orbulina spp. (no older than Langhian, N9), Globorotalia menardii (no older than Serravallian, N12), Globigerinoides subquadratus (no younger than Serravallian, N13) and Globorotalia mayeri (no younger than Serravallian, N14) (Gokcen et al., 1991). The probable severance of this connection by the Late Miocene is implied by the records only in isolated outcrops of the planktonic foraminifera Globigerinoides ruber (no older than Tortonian, N15) and G. extremus (no older than Tortonian–Messinian, N17) (Gokcen et al., 1991; see also Ponikarov, 1967). The identiWcation of the planktonic foraminifera forming

277

Palaeoenvironments: non-mammalian evidence

278

the bases of the Middle and Late Miocene age assignments has been veriWed by reference to photographs. A marine connection as far as the Aleppo area of north-western Syria in the Middle Miocene (‘Vindobonian’–‘Helvetian’) is implied by the records in the Terbol Formation of the bivalve molluscs Chlamys albina (also described from the Ottnangian (Burdigalian) of Central Paratethys (Ro¨gl, 1988)), C. submalvinae and Flabellipecten larteti and the echinoids Amphiope bioculata (originally described from the ‘Helvetian’ of France (A. sp. also described from the Dam Formation (late Early–early Middle Miocene) of Saudi Arabia) (A. B. Smith, Natural History Museum, London, personal communication, February 1998), Clypeaster intermedius, C. martini, Echinolampas hemisphaericus (E. spp. also described from the upper parts of the Asmari Formation (Early–early Middle Miocene) of South-West Iran) (Adams, 1967; T. D. Adams, 1969), Scutella lusitanica (Langhian of Portugal) and S. rotundaeformis (Langhian of the Maritime Alps) (S. spp. also described from the upper parts of the Asmari Formation (Early–early Middle Miocene) of South-West Iran) (T. D. Adams, 1969; Dubertret et al., 1963; see also Wolfart, 1967). A further connection into Jordan is implied by the record in the lower part of the Usdom Group of the echinoids Clypeaster and Echinolampas (Bender, 1974). The identiWcations of the fossils forming the bases of the age assignments require veriWcation. There is also some unpublished palaeontological and map evidence, although again no incontrovertible proof, in the BP (and Anglo-Iranian Oil Company) archives to support a marine connection at least as far as the Mesopotamian Basin (South-Western Iran, North-Eastern Iraq and Eastern Syria) in the Middle Miocene, early–middle but not late Serravallian (see Figs. 14.2–14.3; see also Jones & Racey, 1994; GoV et al., 1995). A marine connection as far as Khuzestan and Lurestan Provinces in South-Western Iran in the Middle Miocene (Langhian to early–middle but not late Serravallian) is implied by the records in the upper part of the Asmari Formation not only of Orbulina universa (no older than Langhian, global standard planktonic foraminiferal zone N9) but also of Globorotalia (Fohsella) peripheroacuta, G. (F.) praefohsi and G. (F.) fohsi (the nominate taxa for the early–middle Serravallian zones N10, N11 and N12 respectively), but not G. (F.) fohsi lobata or G. fohsi robusta (late Serravallian zone N13) (T. D. Adams, 1969; see also James & Wynd, 1965, and Sampo, 1969). While it is again acknowledged that the identiWcations of the planktonic foraminifera forming the bases of the age assignments require veriWcation (to which end a search for the specimens or photographs is being undertaken), it should be stated that Globorotalia (Fohsella) spp. are readily identiWable in matrix-free

Marine invertebrate evidence

279

[Figure 14.2] Early Middle Miocene (Early–Middle Serravallian) palaeogeography of the Middle East. Based on an unpublished map of the so-called Lower Miocene (including Upper Asmari Formation) by H. V. Dunnington in the BP (Anglo-Iranian Oil Company) archives (see also Thomas, 1949, and T. D. Adams, 1969). Hachure indicates approximate position of margin of basin, brickwork shallow marine facies (represented chiefly by platform carbonates), ruling deep marine facies (represented chiefly by pelagic carbonates). Circles indicate positions of surface outcrop and subsurface well control points. F indicates area in South-Western Iran where deep marine facies yielded early Middle Miocene planktonic foraminifera (see text). Marine connection may have extended north-eastwards into Qom Basin in Central Iran at times, but been broken by land bridge between Iran and Anatolia at other times (duration of time-slice greater than frequency of measurable climatic change over same interval, such that climate and climatically induced sea-level could have changed within time-slice).

Palaeoenvironments: non-mammalian evidence

280

[Figure 14.3] Late Middle Miocene (Late Serravallian) palaeogeography of the Middle East. Based on an unpublished map of the so-called Middle Miocene (Lower and Middle Fars Formations and correlatives) by H. V. Dunnington in the BP (Anglo-Iranian Oil Company) archives. Hachure indicates approximate position of margin of basin, stipple non- to marginal-marine facies (represented by clastics and evaporites), brickwork marginal- to shallow-marine facies (chiefly platform carbonates and evaporites), and horizontal ruling deep-marine facies (chiefly pelagic carbonates). Circles indicate positions of surface outcrop and subsurface well control points.

Marine invertebrate evidence

material, and as such are unlikely to have been misidentiWed (again, in thin-section, there is more potential for misidentiWcation). A connection into North-Eastern Iraq and Eastern Syria is implied by the mapped equivalence of the Asmari Formation of South-Western Iran with the Lower–Middle Fars Formations of North-Eastern Iraq and Eastern Syria (H. V. Dunnington, unpubl.). However, to reiterate, there is no incontrovertible proof of a marine connection from the Indo-PaciWc into the Mediterranean during Middle Miocene time. Perhaps signiWcantly, incontrovertibly Middle Miocene, Serravallian representatives of the Globorotalia (Fohsella) lineage (speciWcally, G. (F.) peripheroacuta, G. (F.) praefohsi and G. (F.) fohsi and its subspecies) have never been recorded in the Mediterranean (Iaccarino, 1985) or in Paratethys (Ro¨gl, 1985).

Palaeogeography and faunal migration As noted above, Eurasian palaeogeography over the Oligocene–Holocene interval was modiWed not only by climatically induced cyclic Xuctuations in sea-level but also by mountain-building associated with the convergence between Arabia and Eurasia and the closure of Tethys. This section attempts to describe the palaeogeographic evolution of Western Eurasia (as far east as India and Pakistan) over the critical Oligocene–Miocene interval, in the form of a series of time-slice maps, which are constrained in part by data on the distribution and commonality of larger benthonic foraminifera from the Indo-PaciWc and Mediterranean Provinces (see below; see also Figs. 14.6–14.11). It also considers the consequences for terrestrial vertebrate migration. Complementary and supplementary information is given by Ro¨gl (1996a,b, 1998, in press, and this volume) (and additional references cited therein). The time-slices selected for palaeogeographic mapping are Early Oligocene (Rupelian), Late Oligocene–earliest Miocene (Chattian– Aquitanian), late Early–early Middle Miocene (Burdigalian–Langhian), Middle Miocene (Early–Middle Serravallian), late Middle–early Late Miocene (Late Serravallian–Tortonian) and latest Miocene (Messinian). The mean duration (frequency) of these selected time-slices (dictated by biostratigraphic resolution) is of the order of 4 Ma. In contrast, the frequency of measurable climatic change over the least part of the same interval is of the order of 0.2 Ma or less (see, for instance, Zubakov & Borzenkova, 1990). Thus, importantly, both climate and climatically induced sea-level could

281

Palaeoenvironments: non-mammalian evidence

282

have changed within each time-slice, and each time-slice is best regarded as a composite.

Notes on larger benthonic foraminifera Larger benthonic (bottom-dwelling) foraminifera constitute a disparate group of marine invertebrate microfossils ranging in age from Carboniferous to Holocene. They are grouped together not because they are taxonomically related (modern species belonging to two orders, ancient ones to four), but because they are of comparatively large size, being visible to the naked eye, and having complex internal structure, and are best studied, and in many cases can only be identiWed, in thin-section. The biogeographic and ecological distributions of both ancient and modern species are comparatively well documented (Adams, 1967, 1973; Murray, 1973, 1991; Reiss & Hottinger, 1984; Adams et al., 1990; Jones, 1996). Interestingly, the distributions are similar to those of zooxanthellate (z) corals (Adams et al., 1990; Rosen, this volume). In terms of biogeography, Cenozoic larger benthonic foraminiferal species characterise three provinces, namely the Indo-(West) PaciWc, the Mediterranean, and the Caribbean (Adams, 1967, 1973; Adams et al., 1990). As herein interpreted, the Indo-PaciWc Province includes the PaciWc, Australasia (Australia and New Zealand), South-East Asia, India and Pakistan, East Africa (including the Indian Ocean and Red Sea), and the Middle East, and the Mediterranean Province the Qom Basin of Central Iran, the Eastern Mediterranean and Levant, the Southern Mediterranean and North Africa, the Northern Mediterranean and Southern Europe, Central Europe (including Paratethys), and the Atlantic (including the Aquitaine Basin and West Africa). Also as herein interpreted, the commonality of larger benthonic foraminifera between the Indo-(West) PaciWc and Mediterranean Provinces is a key indicator of the palaeogeographic evolution of Eurasia (Figs. 14.6– 14.11). Palaeobiogeographic and stratigraphic distribution data for selected species are summarised in the Appendix. In terms of ecology, modern species characterise shallow, warm waters of near-normal salinities (some tolerating slightly elevated salinities) and, commonly, platformal and/or peri-reefal carbonate sediments. Ancient (Cenozoic) species and genera are interpreted as having had similar ecological tolerances. Most, if not all, modern larger benthonic foraminifera bear photosynthetic algal symbionts, such that they are restricted to depths within the photic zone (i.e. no deeper than 130 m in clear water, and shallower still in turbid water). Depth tolerances of individual species are dictated by the

Marine invertebrate evidence

precise light requirements of their symbionts. Those with green or red algal or dinoXagellate symbionts, such as most of the porcelaneous-walled Miliolida or ‘milioliforms’ are restricted to shallower (‘back-reef’ (including seagrass)) environments, those with diatom symbionts, such as most of the hyaline-walled Rotaliida or ‘rotaliiforms’ to deeper (‘reef’ and ‘fore-reef’) environments. Interestingly, the tests of modern Rotaliida are more enriched in photosynthetic d12C (or impoverished in d13C) than those of modern Miliolida, arguably indicating a greater photosynthetic capacity. Another possibility is that the Rotaliida use d12C-enriched photosynthetic carbon from their symbionts in test construction, whereas the Miliolida use carbon from sea-water. This is consistent with the observation that the Rotaliida are typical of oligotrophic and the Miliolida of mesotrophic to eutrophic environments, such that the Rotaliida are more and the Miliolida less reliant on their symbionts as a food source. In other words, the symbiosis may be obligate in the Rotaliida and facultative in the Miliolida. Ancient larger benthonic foraminiferal species probably harboured photosynthetic algal symbionts like their modern counterparts. In the case of the Rotaliida, the tests of certain ancient (Eocene–Early Miocene) species are more enriched in d12C (or impoverished in d13C) than others, again arguably indicating a greater photosynthetic capacity. Coincidentally or otherwise, these species evolved and became extinct later than the others, arguably indicating that symbiosis plays an important role in the evolution and extinction of larger benthonic foraminifera. Temperature tolerances of modern larger benthonic foraminifera fall within the range 14–40 °C (generally between 0–35° N/S, but in lower latitudes where cold water currents Xow northward along the western margins of the continental shelves in the southern hemisphere or southward along the eastern margins of the continental shelves in the northern hemisphere). A minimum summer temperature of 18 °C (or possibly slightly less), is apparently necessary for reproduction. Incidentally, reproduction takes place in alternate sexual and asexual stages, with sexually produced individuals generally rare. Whether or not a planktonic habit is developed at any stage of the life cycle remains unclear. Larger benthonic foraminiferal diversity appears to be at least in part a function of sea-surface temperature, like coral diversity (Rosen, 1984, and this volume). The relationship, as interpreted from data presented by Adams et al. (1990) and Murray (1991), is indicated in Fig. 14.4. Data presented by Adams et al. (1990) reveals that 29 out of 31 (94%) high diversity sites, 15 out of 23 (65%) moderate diversity sites (3–5 genera) and 7 out of 15 (47%) low diversity sites (1–2 genera) are located in tropical waters ( 25 °C), and the remainder in extratropical waters ( 25 °C).

283

Palaeoenvironments: non-mammalian evidence

284

[Figure 14.4] Relationship between larger benthonic foraminiferal diversity and temperature at the present time. Based on data in Adams et al. (1990) and Murray (1991).

Larger benthonic foraminiferal diversity can thus be used as a measure of temperature and, assuming uniformitarianism, sea-surface palaeotemperature or palaeoclimate (Fig. 14.5). However, it appears also to be function of several other variables, namely: (i) trophic regime – being highest in oligotrophic environments; (ii) evolution – being highest at times of intense evolutionary activity or in areas of intense evolutionary activity (e.g. South-East Asia); (iii) dispersal/migration – being highest in areas along migration routes from loci of evolution (the main migration route probably being westward from the main locus of evolution in the east, which, coincidentally or otherwise follows the route of the main surface currents); (iv) duration of studied interval and/or size of studied area – being highest at times or in areas of greatest niche availability (a proportional relationship between diversity and extent of available niche having been observed among several groups of terrestrial organisms by Wilson, 1992, and Fortelius et al., 1996); (v) sampling artefact; (vi) taxonomic artefact;

Marine invertebrate evidence

and indeed may be more a function of any one or more of these variables than of palaeotemperature or palaeoclimate. Thus, the comparatively low diversity of larger benthonic foraminifera in the Early Oligocene may be at least in part a function of the early evolutionary stage of the group at this time (although this may itself be related to the lack of available niche prior to the prolonged sea-level high-stand associated with the later climatic optimum, and hence at least indirectly to climate), and the comparatively low diversity in the Mediterranean as opposed to the Indo-PaciWc Province in the Late Oligocene–early Middle Miocene may be at least in part a function of migration from east to west. Moreover, the comparatively low diversity in the late Middle–Late Miocene, especially in areas such as the Middle East, may be at least in part a function of the lack of available niche (although this may itself be related to enhanced continentality and/or evaporite deposition and hence at least indirectly to climate). Furthermore, the comparatively low diversity in general in such areas as Qom Basin and East Africa may be at least in part a function of the comparatively limited extent of these areas or the extent to which they have been studied. Ideally, in order to take into account these eVects, raw diversity data require normalisation.

Early Oligocene (Rupelian) Figure 14.5 shows the inferred palaeogeography of the Rupelian. At this time, the Tethyan Seaway extended from the Indo-PaciWc in the east, through the Mesopotamian Basin (Proto-Mid East Gulf) and the Mediterranean Basin into the Aquitaine Basin and Atlantic in the west, with the Paratethyan Basin partly isolated to the north (and characterised by restricted marine sedimentation). The Gulf of Aden was probably open, the Red Sea unopened, Africa and Arabia contiguous (Hughes et al., 1991; Hughes & Beydoun, 1992). Based on the author’s compilation of data, commonality between IndoPaciWc and Mediterranean Province larger benthonic foraminiferal assemblages, expressed as the percentage of species in common (species identiWcations taken on trust), is 52% (14 species) (see the Appendix; see also Fig. 14.5). The larger benthonic foraminiferal diversity in Eurasia as a whole is 27 species (Fig. 14.5). However, there is a range within the region, from a low of 4 in the Eastern Mediterranean (0 in Australasia (at the time thought to have been located entirely in extratropical latitudes)) to a high of 18 in the Northern Mediterranean and Southern Europe (Fig. 14.5). Interpreted (seasurface) palaeotemperature (see Note on larger benthonic foraminifera

285

Palaeoenvironments: non-mammalian evidence

286

Marine invertebrate evidence

287

[Figure 14.5] Relationship between larger benthonic foraminiferal diversity and interpreted temperature through time. Rupel. = Rupelian; Chatt. = Chattian; Aquit. = Aquitanian; Bur-Lan. = Burdigalian = Langhian; Serr-Tort. = Serravallian-Tortonian. (a) Mediterranean Province (Atlantic, Central Europe, Southern Europe/Northern Mediterranean, North Africa/Southern Mediterranean, Levant/Eastern Mediterranean, Qom Basin); (b) Indo-Pacific Province (Middle East, East Africa, India & Pakistan, South-East Asia, Australasia, Pacific); (c) commonality; (d) summary. Distribution and diversity data compiled from various sources (see Appendix); interpreted temperature from Fig. 14.4 (see also discussion in text).

above) ranges from a low of c. 15 °C in East Africa to a high of c. 25 °C in the Northern Mediterranean and Southern Europe (Fig. 14.5). The migrations of various mammals from the Americas into Eurasia (by way of the land bridges across the Bering Sea and Turgai Strait) appear to have taken place at this time (see, for instance, Ro¨gl, this volume). This event is sometimes referred to as the Grande Coupure or Great Cut.

Late Oligocene–earliest Miocene (Chattian to Aquitanian) Figure 14.7 shows the inferred palaeogeography of the Chattian to Aquitanian. At this time, the Tethyan Seaway was still probably more often open (at times of relative high-stand of sea-level) than closed (at times of relative low-stand), with the Paratethyan Basin still partly isolated to the north. By the Aquitanian, the Red Sea was probably open to the north, and,

Palaeoenvironments: non-mammalian evidence

[Figure 14.6] Rupelian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996, and additional references cited therein; see also Ro¨gl, 1996a,b, 1998, and in press, and additional references cited therein; and Ro¨gl, 1996a,b, 1998, and in press, and additional references cited therein; and Ro¨gl, this volume). Stipple indicates land; otherwise, sea. Arrows indicate marine migration routes (as evidenced by larger benthonic foraminifera). Larger benthonic foraminiferal commonality data extracted from the Appendix.

at least intermittently (there may have been an at least intermittent land bridge between Africa and Arabia in the Afar area), to the south (Hughes et al., 1991; Crossley et al., 1992; Hughes & Beydoun, 1992), enabling the incursion of the Indo-PaciWc and Mediterranean ‘rotaliiform’ larger benthonic foraminifer Miogypsina tani (Hughes & FilatoV, 1995). Commonality between Indo-PaciWc and Mediterranean Province larger benthonic foraminiferal assemblages is 42% (15 species) for the Chattian and 38% (14 species) for the Aquitanian (see the Appendix; see also Fig. 14.5). The larger benthonic foraminiferal diversity in Eurasia as a whole is 36 species for both the Chattian and Aquitanian (Fig. 14.5). However, there is again a range within the region, from a low of 4 in East Africa (and Australasia) to a high of 17 in the Northern Mediterranean and Southern Europe (24 in South-East Asia) for the Chattian, and from a low of 2 in Central Europe to a high of 15 in the Northern Mediterranean and Southern Europe

Marine invertebrate evidence

[Figure 14.7] Chattian–Aquitanian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996; and additional references cited therein; see also Ro¨gl, 1996a,b, 1998, and in press, and additional references cited therein, and Ro¨gl, this volume). Stipple indicates land; otherwise, sea. Circle indicates positions of at least intermittent land bridges. Arrows on land indicate terrestrial migration routes (as evidenced by mammals); arrows at sea marine migration routes (as evidenced by larger benthonic foraminifera). Larger benthonic foraminiferal commonality data extracted from the Appendix.

(27 in South-East Asia) for the Aquitanian (Fig. 14.5). Interpreted (sea-surface) palaeotemperature (see discussion above) ranges from a low of c. 15 °C in East Africa in the Chattian and Central Europe in the Aquitanian to a high of c. 25 °C in the Northern Mediterranean and Southern Europe in the Chattian and Aquitanian, i.e. similar to the preceding interval (Fig. 14.5). The palaeolatitude of Central Europe at this time was approximately 40° N (Dercourt et al., 1985; Rangin et al., 1990; Smith et al., 1994), i.e. close to the limit of tolerance of modern larger benthonic foraminifera (see above). Palynological evidence points towards a subtropical climate and also a mosaic vegetation structure in the circum-Mediterranean area at this time (Suc et al., this volume). Palaeobotanical evidence points towards a generally warm climate in Central Europe (Kovar-Eder et al., 1996). Palaeobotanical and palaeopedological (palaeosol) evidence indicates a

Palaeoenvironments: non-mammalian evidence

[Figure 14.8] Burdigalian–Langhian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996; and additional references cited therein; see also Ro¨gl, 1996a,b, 1998, and in press, and additional references cited therein; and Ro¨gl, this volume). Stipple indicates land; otherwise, sea. Circles indicate positions of at least intermittent land bridges. Arrows on land indicate terrestrial migration routes (as evidenced by mammals); arrows at sea marine migration routes (as evidenced by larger benthonic foraminifera). Larger benthonic foraminiferal commonality data extracted from the Appendix. For maps outlining the vegetation of Africa and Southern Eurasia, see Retallack (1991).

mosaic vegetation structure in East Africa (Retallack, 1991 (see also Andrews, in Jones et al., 1992)). Remains of the primitive quadripedal hominoid Proconsul (probably an arboreal frugivore (Benton, 1990)) are found associated with the so-called Kiewo (the Dholuo word for striped) Series palaeosols formed from seasonally waterlogged wooded grasslands or dambo (Retallack, 1991). The migrations of various mammals from the Americas into Eurasia and Africa (by way of the Bering land bridge) appear to have taken place at this time (Agenian, MN 1–MN 2).

Marine invertebrate evidence

Late Early–early Middle Miocene (Burdigalian–Langhian) Figure 14.8 shows the palaeogeography of the Burdigalian to Langhian. By now, the Tethyan Seaway was probably as often closed (at times of relative low-stand of sea-level) as open (at times of relative high-stand), and the Paratethyan Basin at least intermittently completely isolated to the north. The Red Sea was probably open at its northern end (Adams et al., 1983; Ro¨gl & Steininger, 1984; Steininger & Ro¨gl, 1984; Steininger et al., 1985a), and, at least intermittently, at its southern end (Hughes et al., 1991; Crossley et al., 1992; Hughes & Beydoun, 1992), enabling the incursion of the Indo-PaciWc and Mediterranean ‘rotaliiform’ larger benthonic foraminifer Miogypsina intermedia (Dullo et al., 1983) and the interpreted essentially Indo-PaciWc (Jones, 1994) smaller benthonic foraminifera Ammonia and Pseudorotalia (Jones & Racey, 1994). However, at least intermittent land bridges (including the so-called ‘Gomphotherium land bridge’) probably connected Africa with Iran in the east and Anatolia in the west. Commonality between Indo-PaciWc and Mediterranean Province larger benthonic foraminiferal assemblages at this time is 25% (10 species) (see the Appendix; see also Fig. 14.5). That between the Middle East and Eastern Mediterranean/Levant was 31% (3 species), and that between the Middle East and Qom Basin was 40% (4 species). The larger benthonic foraminiferal diversity in Eurasia as a whole is 40 species (Fig. 14.5). However, there is again a range within the region, from a low of 5 in North Africa and the Southern Mediterranean to a high of 10 in the Northern Mediterranean and Southern Europe (27 in South-East Asia) (Fig. 14.5). Interpreted (sea-surface) palaeotemperature (see discussion above) ranges from a low of c. 15 °C in North Africa and the Southern Mediterranean to a high of c. 20 °C in the Northern Mediterranean and Southern Europe, i.e. similar to the preceding interval (Fig. 14.5). Independent isotopic evidence indicates that there was a global climatic optimum at this time (see palaeotemperature curves on Fig. 14.1). The evolution of various larger benthonic foraminiferal genera, including the ‘milioliform’ Alveolinella, at this time is regarded by Seiglie (1987) as related to niche diversiWcation during the sea-level high-stand associated with this climatic optimum. As well as moderately diverse larger benthonic foraminifera, other tropical to subtropical climatic indicators such as coral reefs (indicating average annual temperatures of  16 °C (Wells, 1975; Adams et al., 1990)) and mangroves (indicating average annual temperature of  20 °C (Plaziat, 1995)) were present as far north as Central Europe, situated in a palaeolatitude of 45° N (see above), at this time (Early Badenian) (Nagy & Kokay, 1991).

291

Palaeoenvironments: non-mammalian evidence

[Figure 14.9] Early–Middle Serravallian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996, and additional references cited therein; see also Ro¨gl, 1996a,b, 1998, and in press, and additional references cited therein; and Ro¨gl, this volume). Stipple indicates land; otherwise, sea. Hachure indicates hypersaline waters (as evidenced by evaporites). Circles indicate positions of at least intermittent land bridges. Arrows on land indicate terrestrial migration routes (as evidenced by mammals); arrows at sea marine migration routes (as evidenced by larger benthonic foraminifera). Larger benthonic foraminiferal commonality data extracted from the Appendix. For maps outlining the vegetation of Africa and Southern Eurasia, see Retallack (1991).

Palynological evidence again points towards a subtropical to tropical climate in the circum-Mediterranean area at this time (Suc et al., this volume). Palaeobotanical evidence again points towards a generally warm climate in Central Europe (Kovar-Eder et al., 1996). Palaeobotanical and palaeopedological (palaeosol) evidence again indicates a mosaic vegetation structure in East Africa (Andrews, in Jones et al., 1992). The migrations of various mammals, including the hominoid Pliopithecus, between Africa and Eurasia, and of numerous mammals within Eurasia, appear to have taken place at this time (Orleanian, MN 3–MN 5). In Eurasia, pliopithecids are recorded in China to the east and throughout Central and Western Europe (e.g. in Slovakia, Austria, Germany, Switzerland and France, and slightly later in Hungary and Poland) to the west (Andrews

Marine invertebrate evidence

& Bernor, this volume). Further migrations from the Americas into Eurasia and Africa (by way of the Bering land bridge) also appear to have taken place at this time.

Middle Miocene (Early–Middle Serravallian) Figure 14.9 shows the palaeogeography of the Early- Middle Serravallian. By now, the Tethyan Seaway was probably more often closed (at times of relative low-stand of sea-level) than open (at times of relative high-stand). The Paratethyan Basin was probably at least intermittently completely isolated to the north, resulting in increased salinity (a salinity crisis, as evidenced by the precipitation of evaporites) in Central Paratethys and decreased salinity in Eastern Paratethys. Even at times of marine connection, at least intermittently hypersaline waters in the Mid-East Gulf and Red Sea (again as evidenced by the precipitation of evaporites) could have formed a barrier to eastward or westward migration of larger benthonic and planktonic foraminifera. Nonetheless, there is some unpublished evidence for a marine connection from the Indo-PaciWc at least as far west as Syria at this time (see note on Timing of closure of Tethys, above). Over the Serravallian–Tortonian interval as a whole (which cannot be readily resolved by means of larger benthonic foraminiferal biostratigraphy), commonality between Indo-PaciWc and Mediterranean Province larger benthonic foraminiferal assemblages is ?11% (?1 species) (see the Appendix; see also Fig. 14.5). The larger benthonic foraminiferal diversity in the region as a whole is ?9 species (Fig. 14.5). However, there is again a range within the region, from a low of 0 in the Northern and Southern Mediterranean and Qom Basin to a high of 3 in East Africa (8 in South-East Asia) (Fig. 14.5). Interpreted (sea-surface) palaeotemperature (see discussion above) is c. 15 °C, i.e. considerably cooler than the preceding interval (Fig. 14.5). Palynological evidence points towards a cooling or drying in the circumMediterranean area at this time (Suc et al., this volume). Palaeobotanical evidence points towards climatic cooling in Central Europe (Kovar-Eder et al., 1996). Palaeobotanical and palynological and palaeopedological (palaeosol) evidence indicates a mosaic vegetation structure in East Africa, characterised not only by the local development of seasonally waterlogged wooded grassland or dambo but also, for the Wrst time, by Afromontane elements (Retallack, 1991). The evolution of the hominoid Ramapithecus, once interpreted as bipedal, at this time is regarded by some authors as related to selection pressure associated with a change to a cooler and/or more arid climate, which caused its arboreal ancestor to move out of its contracting forest

293

Palaeoenvironments: non-mammalian evidence

294

[Figure 14.10] Late Serravallian–Tortonian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996, and additional references cited therein). Stipple indicates land; otherwise, sea. Hachure indicates hypersaline waters (as evidenced by evaporites). Circles indicate positions of at least intermittent land bridges. Arrows indicate terrestrial migration routes (as evidenced by mammals). For maps outlining the vegetation of Africa and Southern Eurasia, see Retallack (1991).

habitat onto an expanding savannah. The migrations of numerous mammals, including not only Ramapithecus but also the (arguably synonymous) pongine hominid (ancestral orangutan) Sivapithecus and the dryopithecine hominid Dryopithecus from Africa into Eurasia appear to have taken place at this time (Astaracian, MN 6–MN 8). In Eurasia, all three groups are recorded in the Chinji, Nagri and Dhok Pathan Stages of the Lower and Middle Siwalik Series in India and Pakistan (Morley-Davies, 1971; Retallack, 1991; Barry, 1995). Ramapithecines are also recorded in China to the east, and in Turkey and Greece and along the banks of the Danube to the west; dryopithecines in China, in Turkey, in the Caucasus, in Slovakia and Poland, along the banks of the Danube, Elbe and Rho ˆ ne Rivers in Hungary, Austria and Germany, and in France and Spain (Andrews & Bernor, this volume). The large-bodied kenyapithecine hominid Griphopithecus is only known from Eurasia (Andrews & Bernor, this volume).

Marine invertebrate evidence

Late Middle-early Late Miocene (Late Serravallian–Tortonian) Figure 15.10 shows the palaeogeography for the Late Serravallian-Tortonian. By now, the Tethyan Seaway was probably completely and permanently closed. There is certainly no positive planktonic or larger benthonic foraminiferal evidence for a marine connection from the Indo-PaciWc and Mediterranean at this time, commonality between larger benthonic foraminiferal assemblages from the two provinces (notwithstanding the proviso that it is practically impossible by means of larger benthonic foraminiferal biostratigraphy alone to distinguish the Late Serravallian– Tortonian from the Early–Middle Serravallian (see above)), probably being 0%. Probably permanent land bridges connected Africa, Arabia, Anatolia and Iran. Palynological evidence again points towards a generally cool, dry climate in the circum-Mediterranean area at this time (Suc et al., this volume). Palaeobotanical evidence points toward a range of warm to cool climatic regimes in Central Europe (Kovar-Eder et al., 1996). Palaeobotanical and palaeopedological (palaeosol) evidence indicates a range of habitats, including seasonally waterlogged wooded grassland or dambo, in Pakistan (Retallack, 1991, 1995). Palaeobotanical and palaeopedological (palaeosol and soil carbonate d13C) evidence also indicates a signiWcant vegetational change in Pakistan between approximately 10–7 Ma, from a dominantly C3 (woodland) to a dominantly C4 (grassland) system (Barry, 1995; Quade et al., 1989; Crowley & North, 1991; Retallack, 1991, 1995; Barry, 1995; Quade & Cerling, 1995). However, the Xoras of the Late Miocene of the so-called Pikermian Province (France, Greece (e..g. Pikermi and Samos), Turkey, Bulgaria, Romania, Moldavia, Ukraine, Russia, Iran (e.g. Maragheh) and China) have recently been reinterpreted on palaeobotanical, palynological and soil carbonate d13C data as having been of sclerophyllous evergreen woodland (C3) aspect (similar to that of the modern Mediterranean) rather than savannah (C4) aspect, the associated faunas that eventually migrated into Africa as pre-adapted rather than adapted to the savannah (see, for instance, Solounias et al., this volume). This reinterpretation is further supported by the masticatory morphology of the mammals, patterns of tooth microwear, and tooth carbonate d13C data (a function of dietary d13C intake), all of which indicate not only grazing but also mixed feeding and browsing (Solounias et al., this volume). The crouzeline pliopithecid Anapithecus made its Wrst appearance in Hungary and Austria during the early Vallesian (MN 9) (as did Ankarapithecus also in Turkey), and the hominine hominid Graecopithecus or Oraunopithecus made its Wrst appearance in Greece during the late Val-

295

Palaeoenvironments: non-mammalian evidence

296

lesian (MN 10) (Andrews & Bernor, this volume). The migrations of numerous mammals between Africa and Eurasia, and within Eurasia appear to have taken place during the Vallesian (MN 9– MN 10). Most notably, the eVectively instantaneous migration of the threetoed horse Hipparion from the Americas into Eurasia (by way of the Bering land bridge) and Africa appears to have taken place at this time (see, for instance, Ro¨gl, in press). In Eurasia, Hipparion is recorded in China, Mongolia, and Kazakhstan, in the Nagri and Dhok Pathan Stages of the Middle Siwalik Series in India and Pakistan, in the United Arab Emirates, Iran, Crete, Greece, Turkey and the Caucasus, and in Ukraine, Rumania, Bulgaria, Hungary, Austria, Germany, France and Spain (see, for instance, Woodburne et al., 1996). In Africa, it is recorded in Tunisia, Algeria, Morocco and Kenya (Morley-Davies, 1971; Retallack, 1991). The Hipparion event has been independently linked to the coincident global glacio-eustatic sea-level lowstand (and deep-sea dissolution hiatus) at 10.5 Ma on the Haq et al. timescale or approximately 11.1 Ma on the Cande & Kent (1995) and Berggren et al. (1995) time-scales (Woodburne & Swisher, 1995; Garces et al., 1997). SigniWcantly (if sampling and preservational eVects can be excluded), most hominoids disappeared from Western Eurasia toward the end of the Vallesian (MN 10) (Fortelius et al., 1996), although some persisted until the Turolian in southerly areas such as the Tusco-Sardinian and Pikermian Provinces. This is interpreted as a direct or indirect response to a change to a cooler, more seasonal (monsoonal) continental climate caused by the closure of Tethys and uplift in the Himalayas, Karakoram and Hindu Kush (see, for instance, O’Brien & Peters, this volume), perhaps ‘. . . variously delayed and moderated by geographic and biotic Wlters’ (Fortelius et al., 1996). As noted above, palaeobotanical and palaeopedological evidence indicates that this climatic change was accompanied by a signiWcant vegetational change. Further migrations of arid steppe mammals (certain rodents, hyenas, and antelopes) within Eurasia appear to have taken place during the Early– Middle Turolian (MN 11–?MN 12). This has been interpreted as evidence of (further) cooling/drying. Warmer, more humid conditions, evidenced by the presence of woodland-dwelling murids, were only locally developed in parts of Western Europe (but see also discussion of Pikermian Province above). The cercopithecid (Old World Monkey) Mesopithecus made its Wrst appearance in Turkey, Greece, and Hungary (Pikermian Province) at this time (Kordos, this volume).

Marine invertebrate evidence

297

[Figure 14.11] Messinian palaeogeography of western Eurasia. Compiled from various sources (see, for instance, Jones & Racey, 1994; Goff et al., 1995; and Jones & Simmons, 1996, and additional references cited therein). Stipple indicates land (triangles high land); ruling, sea. Arrows indicate terrestrial migration routes (as evidenced by mammals). For maps outlining the vegetation of Africa and southern Eurasia, see Retallack (1991).

Latest Miocene (Messinian) Figure 14.11 shows the palaeogeography of the Messinian. At this time, due to climatic change (cooling/drying), base-levels in the Mediterranean and Paratethyan Basins fell to such an extent that they became isolated not only from one another (possibly for the Wrst time (but see above)) but also from the rest of the world’s oceans, resulting in a second salinity crisis. Ultimately, both basins desiccated, as evidenced by the precipitation of evaporites (chlorides in the Mediterranean, carbonates in the Black Sea, whose salinity was lower (Jones & Simmons, 1997)). Palynological evidence again points towards a mostly cool, dry climate in the circum-Mediterranean area at this time, with subtropical elements disappearing, and arid-adapted and semi-desert elements coming to dominate (Van Zinderen Bakker & Mercer, 1986; Suc et al., this volume). Palaeobotanical evidence again points toward a range of warm to cool climatic regimes in Central Europe (Kovar-Eder et al., 1996). The migrations of numerous mammals between Africa and Eurasia (by

Palaeoenvironments: non-mammalian evidence

298

way of a land bridge at the foot of the Iberian Peninsula, an island chain across the Mediterranean, or, ultimately, the dried-up sea-Xoor of the Mediterranean) appear to have taken place at this time (Late Turolian, MN 13). The cercopithecids (Old World Monkeys) Dolichopithecus and Macaca made their Wrst appearances in Europe at this time (Andrews et al., 1996). The migrations of numerous mammals between the Americas and Eurasia (by way of the Bering land bridge) and Africa also appear to have taken place at this time. The continued migration of numerous arid steppe mammals within Eurasia has been taken to imply further climatic deterioration.

Conclusions Global palaeoclimate Xuctuated cyclically during the Oligocene–Holocene, generally deteriorating until the Pleistocene, which was characterised by widespread glaciation. Eurasian palaeogeography was modiWed not only by associated cyclic Xuctuations in sea-level (glacio-eustasy) but also by mountain-building (tectonism) associated with the convergence between Arabia and Eurasia and the closure of the Tethyan Seaway. The palaeogeographic evolution of Western Eurasia over the critical Oligocene–Miocene time interval, constrained in part by data on the distribution and commonality of larger benthonic foraminifera from the IndoPaciWc and Mediterranean Provinces, can be discussed in the form of a series of composite time-slice maps. The mean duration (frequency) of the selected time-slices (dictated by biostratigraphic resolution) is greater than the mean frequency of measurable climatic change over the same time interval. Thus, importantly, both climate and climatically induced sea-level could have changed within each time-slice. Marine (invertebrate) migration routes existed between the Indo-PaciWc and Mediterranean when the Tethyan Seaway was open (permanently throughout the Oligocene and intermittently throughout the Early–Middle Miocene), and terrestrial (vertebrate) migration routes between Africa and Eurasia when it was closed (intermittently through the Early–Middle Miocene and permanently from the Late Miocene–Holocene). Hominoids Wrst migrated out of Africa into Eurasia in the Miocene.

Marine invertebrate evidence

Acknowledgements BP are thanked for permission to publish and for providing word-processing and drafting facilities. Fred Ro¨gl of the Naturhistorisches Museum, Vienna, and John Whittaker of the Natural History Museum, London, are thanked for their constructive reviews, and also for many hours of stimulating discussion on a wide range of topics either directly or indirectly related to the subject matter. Brian Rosen and Andrew Smith, also of the NHM, London, are thanked for helpful discussions concerning the stratigraphic and palaeobiogeographic signiWcance of corals and echinoids respectively.

References Adams, C. G. 1967. Tertiary Foraminifera in the Tethyan, American and Indo-PaciWc Provinces. Systematics Association Publications, 7, 195–217. Adams, C. G. 1970. A Reconsideration of the East Indian Letter ClassiWcation of the Tertiary. Bulletins of the British Museum (Natural History) (Geology), 19, 87–137. Adams, C. G. 1973. Some Tertiary Foraminifera. In Atlas of Palaeobiogeography, Hallam, A. (ed.), pp. 453–68. Elsevier. Adams, C. G. 1984. Neogene Larger Foraminifera, Evolutionary and Geological Events in the Context of Datum Planes. In Neogene PaciWc Datum Planes: Contributions to Biostratigraphy and Chronology, Ikebel, N. & Tsuchi, R. (eds.), pp. 47–67. University of Tokyo Press. Adams, C. G., Bayliss, D. D. & Whittaker, J. E. P. (in press). The Terminal Tethyan Event: a Critical Review of the ConXicting Age Determinations for the Disconnection of the Mediterranean from the Indian Ocean. In Vertebrate Faunas of Arabia, Whybrow, P. J. & Hill, A. (eds.), ch. 33. Yale University Press. Adams, C. G., Gentry, A. W. & Whybrow, P. J. 1983. Dating the Terminal Tethyan Event. Utrecht Micropaleontological Bulletins, 30, 273–98. Adams, C. G., Lee, D. E. & Rosen, B. R. 1990. ConXicting Isotopic and Biotic Evidence for Tropical Sea-Surface Temperatures during the Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology, 77, 289–313. Adams, T. D. 1969. The Asmari Formation of Lurestan and Khuzestan Provinces. Iranian Oil Operating Companies Geological and Exploration Division (Unpublished Report, No. 1154). Ali, M. S. M. 1983. Tertiary Echinoids and the Time of Collision between Africa and Eurasia. Neues Jahrbuch fu¨ r Geologie und Palaontologie, Monatshefte, 1983(4), 213–27. Andrews, P., Harrison, T., Delson, E., Bernor, R. L. & Martin, L. 1996. Distribution and Biochronology of European and Southwest Asian Miocene Catarrhines. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 168–207. New York, Columbia University Press.

299

Palaeoenvironments: non-mammalian evidence

300

Barry, J. C. 1995. Faunal Turnover and Diversity in the Terrestrial Neogene of Pakistan. In Paleoclimate and Evolution (with Emphasis on Human Origins), Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, L. H. (eds.), pp. 115–34. Yale University Press. Barry, J. C., Johnson, N. M., Raza, S. M. & Jacobs, L. L. 1985. Neogene Mammalian Faunal Change in Southern Asia: Correlations with Climatic, Tectonic and Eustatic Events. Geology, 13, 637–40. Bender, F. 1974. Geology of Jordan. Berlin, Stuttgart, Borntrager. Benton, M. J. 1990. Vertebrate Palaeontology. London, Chapman & Hall. Berggren, W. A., Kent, D. V., Swisher, C. C., III & Aubry, M.-P. 1995. A Revised Cenozoic Geochronology and Chronostratigraphy. In Geochronology, Time Scales and Global Stratigraphic Correlation, ed. W. A. Berggren et al., pp. 129–212. Society of Economic Paleontologists and Mineralogists (Special Publication, No. 54). Bernor, R. L., Brunet, M., Ginsburg, L., Mein, P., Pickford, M., Ro¨gl, F., Sen, S., Steininger, F. & Thomas, H. 1987. A Consideration of some major Topics concerning Old World Miocene Mammalian Chronology, Migrations and Paleogeography. Geobios, 20(4), 431–9. Bossart, P. & Ottiger, R. 1989. Rocks of the Murree Formation in Northern Pakistan: Indicators of a Descending Foreland Basin of Late Paleocene to Middle Eocene Age. Eclogae Geologicae Helvetiae, 82(1). Buchbinder, B. 1996. Miocene Carbonates of the Eastern Mediterranean, the Red Sea and the Mesopotamian Basin: Geodynamic and Eustatic Controls. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.), pp. 89–96. Tulsa, Oklahoma, Society of Economic Paleontologists and Mineralogists (Concepts in Sedimentology and Paleontology, Volume 5). Cahuzac, B. & Poignant, A. 1997. Essai de Biozonation de l’Oligo-Miocene dans les Bassins Europeens a l’Aide des Grands Foraminiferes Neritiques. Bulletin de la Societe Geologique de France, 168(2), 155–69. Cande, S. C. & Kent, D. V. 1995. Revised Calibration of the Geomagnetic Polarity Time Scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 100, 6093–5. Chepalyga, A. L. 1985. Climatic and Eustatic Fluctuations in the Paratethys Basin’s History. Abstracts, VII Congress, Regional Commission on Mediterranean Neogene Stratigraphy, Budapest, 134–6. Crossley, R., Watkins, C., Raven, M., Cripps, D., Carnell, A. & Williams, A. 1992. The Sedimentary Evolution of the Red Sea and Gulf of Aden. Journal of Petroleum Geology, 15(2), 157–72. Crowley, T. J. & North, G. R. 1991. Paleoclimatology. Oxford University Press. Ctyroky, P., Karin, S. A. & van Vessem, E. J. 1975. Miogypsina and Borelis in the Euphrates Limestone Formation of the Western Desert of Iraq. Neues Jahrbuch fu ¨ r Geologie und Palaontologie, Abhandlungen, 148(1), 33–49. Demarcq, J. 1985. Paleothermic Evolution during the Neogene in Mediterranean through the Marine Megafauna. Abstracts, VII Congress, Regional Commission on Mediterranean Neogene Stratigraphy (RCMNS), Budapest, 1985, 176–8. Dercourt, J. et al. 1985. Presentation de 9 Cartes paleogeographiques au 1/20.000.000e s’Etendant de l’Atlantique au Pamir pour la Periode du Lias a l’Actuel. Bulletin de la Societe Geologique de France, Ser. 8, 1, 637–52.

Marine invertebrate evidence

Drooger, C. W. 1979. Marine Connections of the Neogene Mediterranean deduced from the Evolution and Distribution of Larger Foraminifera. Annales Geologiques des Pays Helleniques, 1979(1), 361–9. Dubertret, L., Daniel, E. I. & Bender, F. 1963. Liban, Syrie, Jordanie. Lexique Stratigraphique International, III(10c1). Dullo, W.-C., Hotzl, H. & Jado, R. A. 1983. New Stratigraphical Results from the Tertiary Sequence of the Midyan Area, NW Saudi Arabia. Newsletters on Stratigraphy, 12(2), 75–83. Flynn, L. J., Pilbeam, D., Jacobs, L. L., Barry, J. C., Behrensmeyer, A. K. & Kappelman, J. W. 1990. The Siwaliks of Pakistan: Time and Faunas in a Miocene Terrestrial Setting. Journal of Geology, 98, 589–604. Fortelius, M., Werdelin, L., Andrews, P., Bernor, R. L., Gentry, A., Humphrey, L., Mittmann, H.-W. & Viratana, S. 1996. Provinciality, Diversity, Turnover and Palaeoecology in Land Mammal Faunas of the Later Miocene of Western Eurasia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 414–48. New York, Columbia University Press. Garces, M., Cabrera, L., Agustı´, J. & Pares, J. M. 1997. Old World First Appearance Datum of Hipparion Horses: Late Miocene Large-Mammal Disperal and Global Events. Geology, 25(1), 19–22. GoV, J. C., Jones, R. W. & Horbury, A. D. 1995. Cenozoic Basin Evolution of the Northern Part of the Arabian Plate and its Control on Hydrocarbon Habitat. In GEO ’94: The Middle East Petroleum Geosciences (Selected Papers from the Middle East Geosciences Conference, 1994), Al-Husseini, M. I. (ed.), pp. 402–12. Manama, Bahrain, Gulf PetroLink. Gokcen, N., Gokcen, S. L. & Kelling, G. 1991. Microfaunal Biostratigraphy of Cenozoic Sequences from the Misis-Andirin Area, Southern Turkey. Newsletters on Stratigraphy, 24(3), 111–35. Haq, B. U. 1991. Sequence Stratigraphy, Sea-Level Change, and SigniWcance for the Deep Sea. In Sedimentation, Tectonics and Eustasy: Sea-Level Changes at Active Margins, MacDonald, D. I. M. (ed.), pp. 3–39. Oxford, Blackwell ScientiWc Publications (International Association of Sedimentologists Special Publication, No. 12). Haq, B. U., Hardenbol, J. & Vail, P. R. 1988. Mesozoic and Cenozoic Chronostratigraphy and Eustatic Cycvles. In Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. C., Posamentier, H., Ross, C. A. & Van Wagoner, J. C. (eds.), pp. 71–108. Sea-Level Changes: an Integrated Approach, Society of Economic Paleontologists and Mineralogists (Special Publication, No. 42). Hughes, G. W. & Beydoun, Z. R. 1992. The Red Sea – Gulf of Aden: Biostratigraphy, Lithostratigraphy and Palaeoenvironments. Journal of Petroleum Geology, 15(2), 135–56. Hughes, G. W. & FilatoV, J. 1995. New Biostratigraphic Constraints on Saudi Arabian Red Sea Pre- and Syn- Rift Sequences. In GEO ’94: The Middle East Petroleum Geosciences (Selected Papers from the Middle East Geosciences Conference, 1994), Al-Husseini, M. I. (ed.), pp. 517–28. Manama, Bahrain, Gulf PetroLink. Hughes, G. W., Varol, O. & Beydoun, Z. R. 1991. Evidence for Middle Oligocene Rifting of the Gulf of Aden and Late Oligocene Rifting of the Southern Red Sea. Marine & Petroleum Geology, 8, 354–8.

301

Palaeoenvironments: non-mammalian evidence

302

Iaccarino, S. 1985. Mediterranean Miocene and Pliocene Planktic Foraminifera. In Plankton Stratigraphy, Bolli, H. M., Saunders, J. B. & Perch-Nielsen, K. (eds.), pp. 283–314. Cambridge University Press. James, G. A. & Wynd, J. G. 1965. Stratigraphic Nomenclature of Iranian Oil Consortium Agreement Area. American Association of Petroleum Geologists Bulletin, 49, 2182–245. Jones, R. W. 1994. The Challenger Foraminifera. Oxford University Press. Jones, R. W. 1996. Micropalaeontology in Petroleum Exploration. Oxford University Press. Jones, R. W. 1997. Aspects of the Cenozoic Stratigraphy of the Northern Sulaiman Ranges, Pakistan. Journal of Micropalaeontology, 16, 51–8. Jones, R. W. & Racey, A. 1994. Cenozoic Stratigraphy of the Arabian Peninsula and Gulf. In Micropalaeontology and Hydrocarbon Exploration in the Middle East, Simmons, M. D. (ed.), pp. 273–308. Chapman & Hall. Jones, R. W. & Simmons, M. D. 1996. A Review of the Stratigraphy of Eastern Paratethys. Bulletins of the Natural History Museum, London (Geology), 52(1), 25–49. Jones, R. W. & Simmons, M. D. 1997. A Review of the Stratigraphy of Eastern Paratethys (Oligocene–Holocene), with particular emphasis on the Black Sea. In Regional and Petroleum Geology of the Black Sea and surrounding areas, Robinson, A. G. (ed.), pp. 39–52. Tulsa, Oklahoma, American Association of Petroleum Geologists (Memoir, No. 68). Jones, S., Martin, R. & Pilbeam, D. 1992. The Cambridge Encyclopaedia of Human Evolution. Cambridge University Press. Karim, S. A. 1978. The Genus Borelis de Montfort from the Oligocene–Miocene Sediments of Iraq. Journal of the Geological Society of Iraq, 11, 106–18. Kovar-Eder, J., Kvacek, Z., Zastawniak, E., Givulescu, R., Hably, L., Mihajlovic, D., Teslenko, J. & Walther, H. 1996. Florisitic Trends in the Vegetation of the Paratethys Surrounding Areas during Neogene Times. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 395–413. New York, Columbia University Press. McCall, G. J. H., Rosen, B. R. & Darrell, J. G. 1994. Carbonate Deposition in Accretionary Prism Settings: Early Miocene Coral Limestones and Corals of the Makran Mountain Range in Southern Iran. Facies, 31, 141–78. Morley-Davies, A. 1971. Tertiary Faunas: Second Edition (revised by F. E. Eames with a contribution by R. J. G. Savage). George Allen & Unwin. Murrary, J. W. 1973. Distribution and Ecology of Living Benthic Foraminiferids. Heinemann Educational Books. Murray, J. W. 1991. Ecology and Palaeoecology of Benthic Foraminifera. Longman ScientiWc & Technical. Nagy, E. & Koday, J. 1991. Middle Miocene Mangrove Vegetation in Hungary. Acta Geologica Hungarica, 34(1/2), 45–52. Pilbeam, D., Morgan, M., Barry, J. C. & Flynn, L. 1996. European MN Units and the Siwalik Faunal Sequence of Pakistan. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 96–105. New York, Columbia University Press. Plaziat, J.-C. 1995. Modern and Fossil Mangroves and Mangals: their Climatic and Biogeographic Variability. In Marine Palaeoenvironmental Analysis from

Marine invertebrate evidence

Fossils, Bosence, D. W. J. & Allison, P. A. (eds.), pp. 73–96. London, Geological Society (Special Publication, No. 83). Ponikarov, V. P. 1967. The Geology of Syria. Syrian Arab Republic; Ministry of Industry. Prazak, J. 1978. The Development of the Mesopotamian Basin during the Miocene. Journal of the Geological Society of Iraq, 9, 170–89. Quade, J. & Cerling, T. E. 1995. Expansion of C4 Grasses in the Late Miocene of Northern Pakistan: Evidence from Stable Isotopes in Paleosols. Palaeogeography, Paleoclimatology, Palaeoecology, 114, 91–116. Quade, J. E., Cerling, T. E. & Bowman, J. R. 1989. Development of Asian Monsoon revealed by Marked Ecological Shift during the Latest Miocene in Northern Pakistan. Nature, 342, 163–6. Rangin, C. et al. 1990. Paleogeodynamic Maps of the West PaciWc. Bulletin de la Societe Geologique de France, 6. Reiss, Z. & Hottinger, L. 1984. The Gulf of Aqaba: Ecological Micropalaeontology. Springer-Verlag. Retallack, G. J. 1991. Miocene Paleosols and Ape Habitats of Pakistan and Kenya. Oxford University Press. Retallack, G. J. 1995. Palaeosols of the Siwalik Group as a 15Myr Record of South Asian Paleoclimate. Memoirs, Geological Survey of India, 32, 36–51. Ro¨gl, F. 1985. Late Oligocene and Miocene Planktic Foraminifera of the Central Paratethys. In Plankton Stratigraphy, Bolli, H. M., Saunders, J. B. & Perch-Nielsen, K. (eds.), pp. 315–28. Cambridge University Press. Ro¨gl, F. 1996a. Migration Pathways between Africa and Eurasia – Oligocene–Miocene Palaeogeography. Europal, 10, 23–6. Ro¨gl, F. 1996b. Stratigraphic Correlation of the Paratethys Oligocene and Miocene. Mitteilungen der Gesellschaft der Geologie und Bergbaustudenten in Wien, 41, 65–73. Ro¨gl, F. 1998. Palaeogeographic Considerations for Mediterranean and Paratethys Seaways (Oligocene to Miocene). Annalen der Naturhistorischen Museum, Wien, 99A, 279–310. Ro¨gl, F. (in press). Oligocene and Miocene Palaeogeography and Stratigraphy of the Circum-Mediterranean Region. In Vertebrate Faunas of Arabia, Whybrow, P. J. & Hill, A. (eds.), ch. 34. Yale University Press. Ro¨gl, F. & Steininger, F. F. 1983. Vom Zerfall der Tethys zu Mediterran und Paratethys: die Neogene Palaogeographie und Palinspastik des Zirkum-Mediterranen Raumes. Annalen des Naturhistorischen Museums, Wien, 85A, 135–63. Ro¨gl, F. & Steininger, F. F. 1984. Neogene Paratethys, Mediterranean and Indo-PaciWc Seaways. In Fossils and Climate, Brenchley, P. (ed.), pp. 171–200. John Wiley. Rosen, B. R. 1984. Reef Coral Biogeography and Climate through the Late Cainozoic. In Fossils and Climate, Brenchley, P. (ed.), pp. 201–62. John Wiley. Sampo, M. 1969. Microfacies and Microfossils of the Zagros Area, Southwest Iran (from Pre-Premian to Miocene). Leiden, E. J. Brill. Seiglie, G. A. 1987. Sea-Level Changes and Cenozoic Large Foraminifera with Inner Structures. Timing and Depositional History of Eustatic Sequences: Constraints on Sequence Stratigraphy, Ross, C. A. & Haman, D. (eds.), pp. 169–86. Cushman Foundation for Foraminiferal Research (Special Publication, No. 24). Smith, A. G., Smith, D. G. & Funnell, B. M. 1994. Atlas of Mesozoic and Cenozoic Shorelines. Cambridge University Press.

303

Palaeoenvironments: non-mammalian evidence

304

Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Miocene and Pliocene) Marine-Continental Correlations of Europe Mammal Units and Zones. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 7–46. New York, Columbia University Press. Steininger, F. F., Rabeder, G. & Ro¨gl, F. 1985a. Land Mammal Distribution in the Mediterranean Neogene: a Consequence of Geokinematic and Climatic Events. In Geological Evolution of the Mediterranean Basin, Stanley, D. J. & Wezel, F.-C. (eds.), pp. 559–71. Springer-Verlag. Steininger, F. F. & Ro¨gl, F. 1979. The Paratethys History – a Contribution towards the Neogene Dynamics of the Alpine Orogene. Annales Geologiques des Pays Helleniques, 1979(3), 1153–75. Steininger, F. F. & Ro¨gl, F. 1984. Paleogeography and Palinspastic Reconstruction of the Neogene of the Mediterranean and Paratethys. In The Geological Evolution of the Eastern Mediterranean, Dixon, J. E. & Robertson, A. H. F. (eds.), pp. 659–68. Blackwell. Steininger, F. F., Senes, J., Kleeman, K. & Ro¨gl, F. 1985b. Neogene of the Mediterraneal Tethys and Paratethys: Stratigraphic Correlation Tables and Sediment Distribution Maps. University of Vienna Press. Stocklin, J. & Setudehnia, A. O. 1972. Iran. Lexique Stratigraphique International, III(9b). Suess, E. 1893. Are Great Ocean Depths Permanent? Natural Science, London, 2, 180–7. Thomas, A. N. 1949. Tentative Isopach Maps of the Upper Asmari Limestone and the Oligocene–Lower Miocene of South-West Iran. Anglo-Iranian Oil Company (Unpublished Report, No. ANT/11). Van Zinderen Bakker, E. M. & Mercer, J. H. 1986. Major Late Cainozoic Climatic Events and Palaeoenvironmental Changes in Africa viewed in a Worldwide Context. Palaeogeography, Palaeoclimatology, Palaeoecology, 56, 217–35. Wells, J. W. 1957. Coral Reefs. Memoirs of the Geological Society of America, 67(1), 609–31. Wilson, E. O. 1992. The Diversity of Life. Penguin. Wolfart, R. 1967. Geologie von Syrien und dem Libanon. Berlin, Gebruder Borntraeger. Wonders, A. A. H. & Adams, C. G. 1991. The Biostratigraphical and Evolutionary SigniWcance of Alveolinella praequoyi sp. nov. from Papua New Guinea. Bulletins of the British Museum (Natural History) (Geology), 47(2), 169–75. Woodburne, M. O., Bernor, R. L. & Swisher, C. C. 1996. An Appraisal of the Stratigraphic and Phylogenetic Bases for the Hipparion Datum in the Old World. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 124–36. New York, Columbia University Press. Woodburne, M. O. & Swisher, C. C. III 1995. Land Mammal High-Resolution Geochronology, Intercontinental Overland Dispersals, Sea Level, Climate and Vicariance. In Geochronology, Time Scales and Global Stratigraphic Correlation, Berggren, W. A. et al. (eds.), pp. 335–64. Society of Economic Paleontologists and Mineralogists (Special Publication, No. 54). Zubakov, V. A. & Borzenkova, I. I. 1990. Global Palaeoclimates of the Late Genozoic. Elsevier.

Tf1-Tf2 Tf3-Th Td-Te1-4 Te Td-Te Te5 Te5-Tf1 Td Td-Te Te1-4 Tf1 Tf-Th Tb-Te5 Tc-Td Tf3-Th Td Te-Tf Tf Td-Te1-4 Td-Te1-4 Te5-Tf

Alveolinella praequoyi Alveolinella quoyi Archaias operculiniformis Archaias vandervlerki Austrotrillina asmariensis Austrotrillina brunni Austrotrillina howchini Austrotrillina paucialveolat??? Austrotrillina striata Borelis inflata Borelis melo curdica Borelis melo melo Borelis pygmaeus Bullaveolina bulloides Cycloclypeus carpenteri Cycloclypeus droogeri Cycloclypeus eidae Cycloclyppeus indopacificus Cycloclypeus mediterrane??? Cycloclypeus oppenoorthi Cycloclypeus posteidae Eulepidina andrewsiana Eulepidina badjirraensis Eulepidina dilatata Eulepidina elephantina Eulepidina ephippioides Eulepidina formosa Eulepidina formosinoides

Te1-4 Te Td

Td-Te1-4

Strat. dist.

Species

;

;

;

;

;

; ;

; ;

; ;

? ; ; ; ? ; ;

3

; ;

;

2

; ; ;

;

1

;

;

;

4

;

; ;

;

;

;

;

; ;

;

;

;

;

; ?

; ;

?

;

8

;

;

7

;

;

6

?

5

; ;

;

;

;

;

9

; ; ; ;

; ;

;

;

;

;

; ; ; ;

; ;

10

;

;

;

;

11

;

; ;

;

;

;

;

;

;

12

Appendix. Stratigraphic and palaeobiogeographic distribution of selected Oligocene–Miocene larger benthonic foraminifera

Eulepidina manduensis Eulepidina papuaensis Flosculinella bontangensis Flosculinella borneensis Flosculinella globulosa Flosculinella reichli Halkyardia minima Katacycloclypeus annulat??? Katacycloclypeus martini Marginopora vertebralis Miogypsina antillea Miogypsina basraensis Miogypsina bifida Miogypsina borneensis Miogypsina burdigalensis Miogypsina cushmani Miogypsina droogeri Miogypsina excentrica Miogypsina globulina Miogypsina gunteri Miogypsina intermedia Miogypsina kotoi Miogypsina mediterranea Miogypsina nardaghiensis Miogypsina polymorpha Miogypsina septentrionalis Miogypsina siahkuhensis Miogypsina tani

Species

Appendix (cont.)

;

N6-N8 Te1-4

N4

;

; ;

; ;

;

1

N7-N8

Tf3-Th N8-?N14 N4B-N5 N7-?N12 N4B-N6 N5-N6 N7-N8 N5 N6 N5-N6 N4B-N5 N7

Te5-Tf1 Te5 Ta3-Te1-4 Te5-Tf

Te Tf1

Strat. dist.

;

;

; ;

2

;

;

; ; ;

; ;

;

;

3

;

;

4

;

; ; ;

; ;

5

;

;

; ; ;

6

;

;

;

7

;

;

;

;

8

;

; ; ; ; ;

;

;

;

9

;

;

; ; ; ; ; ; ;

; ;

; ; ; ; ; ; ; ; ; ;

10

;

;

;

;

;

;

11

;

;

12

Miogypsina thecidaeformis Miogypsinoides bantamens??? Miogypsinoides bermudezi Miogypsinoides complanat??? Miogypsinoides cupulaefor??? Miogypsinoides dehaarti Miogypsinoides formosensis Miogypsinoides indica Miogypsinoides mauretani??? Miogypsinoides saipanensis Nephrolepidina angulosa Nephrolepidina borneensis Nephrolepidina brouweri Nephrolepidina ferreroi Nephrolepidina howchini Nephrolepidina inflata N. isolepidinoides Nephrolepidina japonica N. marginata-tournoueri Nephrolepidina martini Nephrolepidina morgani Nephrolepidina orakeiensis Nephrolepidina orientalis Nephrolepidina parva N. praemarginata Nephrolepidina radiata Nephrolepidina rutteni Nephrolepidina sumatrensis Nephrolepidina talahabensis Nephrolepidina transiens Nephrolepidina verbeeki N. fichteli-intermedius

Te5-Tf Tc-Td

Tf Td-Te1-4 Td TF Tf Te1-4-Te5

Te-Tf N3-N4 Te5-Tf Te5-Tf1 Te5-Tf

Tf1

Te5-Tf1 P21B-N4A Te1-4 Te1-4 Te5-Tf P22-N5 Te1-4 Tf1 P22-N4 P21B-N4A N7-N10 ;

; ;

;

;

;

;

;

;

;

;

;

;

;

;

;

; ;

; ;

;

;

;

;

;

;

;

;

;

;

;

;

;

;

;

; ;

;

;

;

; ; ;

; ; ; ;

; ; ; ; ;

;

; ; ; ; ; ; ;

; ; ;

; ; ;

; ; ; ;

;

;

;

;

;

;

;

;

;

; ; ;

;

Tb-Tc Tc-Td Ta3-Tc Tf-Tf3 Tc Te-Tf Tf Tb-Td Td-Te Tb-Te Tc-Te1-4 Tc-Te5 Tf1-?Tf2 Tf1 Tf1 Tf1-?Tf2 Tc-Te1-4 Te5-Tf1 Tb-Te5 Te5 Te1-4-Te5 Te1-4 Te1-4

Nummulites germanicus Nummulites pengaronensis Nummulites vascus Operculina bikiensis Peneroplis armorica Peneroplis evolutus Peneroplis farsensis Peneroplis glynjohnsi Peneroplis thomasi Palaeonummulites bouillei P. incrassatus Planoperculina complanat??? Planostegina costata Planostegina giganteoform??? Planostegina granulatatest??? Planostegina politatesta Praerhapydionina delicata Pseudotaberina malabaric??? Spiroclypeus spp. Tansinhokella yabei Vlerkina assilinoides Vlerkina borneensis Vlerkina pleurocentralis ;

;

; ;

;

; ; ;

3

;

; ; ; ;

;

; ;

2

; ; ;

;

;

;

;

;

1

;

; ;

; ;

4

;

;

;

; ;

5

; ;

;

6

;

;

; ;

;

; ; ; ; ;

; ;

7

; ;

;

; ;

8

;

; ;

9

;

; ; ; ;

;

; ; ;

10

;

11

;

; ;

;

;

12

Compiled from various sources. Stratigraphic distribution: Ta-Th = East Indian Letter Stages; P and N zones = global standard planktonic foraminiferal zones. Palaeobiogeographic distribution: 1 = Atlantic; 2 = Central Europe; 3 = Northern Mediterranean & Southern Europe; 4 = Southern Mediterranean & North Africa; 5 = Eastern Mediterranean & Levant; 6 = Qom Basin; 7 = Middle East; 8 = East Africa; 9 = India & Pakistan; 10 = = South-East Asia; 11 = Australasia; 12 = PaciWc. IdentiWcations unveriWed.

Strat. dist.

Species

Appendix (cont.).

15 Palaeoclimatic implications of the energy hypothesis from Neogene corals of the Mediterranean region Brian R. Rosen

Introduction Fossil corals have long been used for palaeoenvironmental interpretation, largely because many extant taxa are restricted to warm shallow tropical marine waters, and this is commonly extrapolated to the fossil coral record on simple uniformitarian grounds. For the most part, this approach has been largely qualitative, the presence of corals, or more particularly, the presence of ‘reef corals’, at a particular place being used to infer a warm climate, and their absence to infer a cool climate. By extension from this, the northernmost and southernmost limits of reef coral distributions in the past have also been used to infer times of global warming and cooling (e.g. Adams et al., 1990), since the global reef coral belt has varied in width through time. The present paper is a preliminary attempt to make more use of the potential of corals, by invoking the ‘energy hypothesis’, which provides a quantitative relationship between taxonomic richness and prevailing temperatures. This is used here to derive actual palaeotemperatures from the Miocene coral record of the Mediterranean region.

Palaeontological and taxonomic background The corals which are the subject of this contribution belong to the order Scleractinia which Wrst appeared in the Middle Triassic (Roniewicz & Morycowa, 1993). Scleractinia are also common in modern seas, where they are a major biotic element of coral reef habitats like the Great Barrier Reef, as well as forming deep and cool water coral banks (Stanley & Cairns, 1988). There is a substantial Neogene fossil coral record for the combined European and Mediterranean region (referred to for convenience here simply as the ‘Mediterranean’ region), though many of the outcrops are small and scattered, and some important localities are no longer collectable. In general, Miocene corals of the western Mediterranean region, together with the immediately adjacent areas of France, are much better studied than Neogene faunas elsewhere in Europe and the Mediterranean. Better knowledge of the western Mediterranean is largely due to the landmark revision and synthesis by Chevalier (1962). Stratigraphically however, information about many Mediterranean coral formations is not always up to date, and a good revision and synthesis is desirable. Nevertheless, much relevant information can be

Palaeoenvironments: non-mammalian evidence

310

extracted from the recent review of Mediterranean reef geology by Franseen et al. (1996). Much of the present faunal data has been extracted from the papers within this work.

Ecological background So far, corals have been mentioned without any further ecological details. For a palaeoclimatic study however, it is essential to distinguish corals which are restricted to warm shallow tropical marine waters, from those which are not. Richness of tropical corals dwindled almost to zero by the end of the Miocene in the Mediterranean region. The most proximal explanation for this is progressive cooling to the point at which the prevailing climate was too temperate to support tropical marine faunas. This can be attributed either to global cooling, or to northward tectonic shift of the Mediterranean borderlands from more tropical latitudes to more temperate ones, or to a combination of both. This is further discussed in the light of the results of the present study. Corals found only in warm, shallow waters are those which are symbiotic with dinoXagellate algae (zooxanthellae). These are zooxanthellate or ‘zcorals’, whereas non-symbiotic corals (i.e. azooxanthellate or ‘az-corals’) are not restricted in this way. Z-corals today are distributed within a low latitude circumglobal band between 35–40° N and about 35° S, and the lowest mean annual surface temperatures for z-corals are about 16 °C (Rosen, 1984; Fraser & Currie, 1996). For species tolerances to particular seasonal minima, Veron (1995) concluded that about half the corals in Japanese waters could tolerate seasonal minima down to 14 °C. Provided the corals concerned are known to be z-corals, the existence of fossil scleractinians at a given location can therefore be used, on uniformitarian grounds, to infer a minimum sea water temperature of around 14–16 °C. However, although z-corals are conWned to these limits, az-corals, though often inconspicuous in shallow tropical waters, can also occur here too (Stanley & Cairns, 1988). Hence a fossil coral fauna cannot be assumed a priori to consist entirely of z-corals or entirely of az-corals, and the task of using corals for palaeoclimatic purposes presupposes that one can distinguish the two kinds of coral (see below). Further palaeoclimatic detail can be derived by applying the energy hypothesis. This is an empirical model which relates Wrst order global patterns of taxonomic richness of scleractinians to incident solar energy (Rosen, 1984; Veron, 1995; Fraser & Currie, 1996). Broadly, the higher the incident solar energy at a given place, the higher the number of taxa present,

Palaeoclimatic implications of the energy hypothesis

311

[Figure 15.1] The variation in the square root of modern scleractinian coral richness with mean annual sea surface temperatures, based on 130 sites, worldwide (redrawn from Fraser & Currie, 1996, fig. 4). The authors’ original conversion of generic richness to square roots is retained here, but the unconverted values are also shown in parentheses. The envelope curve of maximum values of richness has been subjectively fitted here for the purposes of the present paper. Although the authors do not explicitly distinguish z-coral and az-corals (see text), their data are based on ‘reef corals’ which are taken here to correspond to z-corals.

though for practical reasons in marine environments, prevailing sea water temperatures are customarily used as a proxy for incident energy (Fig. 15.1). For a fossil coral fauna, this model can therefore be used, as here, to obtain palaeotemperatures from richness values of a given fauna at a given place. Rosen (1977) Wrst attempted to do this (in combination with other approaches) for the Lower Cretaceous corals of southern England, but Fraser & Currie’s (1996) model (Fig. 15.1) is preferable because it is based on more data from modern corals than were available to Rosen (1977). While use of the energy hypothesis lacks the detailed precision of other palaeotemperature methods, notably stable isotope analysis, it does provide independent evidence which can be useful for comparative purposes, and can also reveal important anomalies (Adams et al., 1990). The methodology of using the energy hypothesis is given later. Further ecological details of z-corals relevant to palaeoclimatic interpretation are as follows. The presence of photosynthetic symbionts in the cells of z-corals restricts them to the euphotic zone, a working generalization for which is about 100 m depth of water. In practice, this means that most z-corals are restricted to clear waters away from areas of heavy clastic sedimentation. Taxonomic richness, however, also decreases with increas-

Palaeoenvironments: non-mammalian evidence

312

ing depth, as conWrmed by numerous compilations and ecological transect studies throughout the reef coral realm (Rosen, 1977; Perrin et al., 1995), though the decrease is often irregular. Ultimately, this decrease is controlled by diminishing light penetration, hence by implication, richness decreases with increasing turbidity too. Overall, this means that richness patterns can be potentially explained by these (and other) factors, and not solely by temperature. This must clearly be borne in mind when applying the energy hypothesis to palaeoclimatic interpretation, though Fraser & Currie (1996) concluded that temperature was the dominant Wrst order factor. Corals are benthic organisms which are often preserved as autochthonous or parautochthonous assemblages. This means that any palaeoenvironmental inference made from a coral fauna can usually be assumed to apply to their immediate sample location, unless there is evidence for large scale transportation either of the corals themselves, or of their enclosing sediments. Often the site of z-coral occurrence is reefal and the temperature minimum for reefs in modern waters is about 18 °C (i.e. somewhat higher than that of z-corals in general), hence the occurrence of a fossil reef can also be used to infer a minimum sea water temperature. ‘Reef corals’ are a reasonable approximation for z-corals, and much of the existing palaeoecological and palaeobiogeographical literature on corals is based on this approximation, and this also applies to many distributional studies of modern corals too, including Fraser & Currie’s (1996) plot shown in Fig. 15.1. Nevertheless more precision is desirable because z-corals are frequently found in bedded deposits too, and the word ‘reef’ has also been used loosely to include a whole range of features, some of which are not inhabited or constructed by z-corals at all. There is therefore no watertight correlation between the occurrence of z-corals and that of reefs. This is especially true of the Mediterranean Neogene, as can be seen from the variety of reefal formations and structures treated by Franseen et al. (1996). Not all of these are restricted to warmer (tropical and sub-tropical) waters. Hence, if reef evidence is to be used at all for palaeoclimatic purposes, it is safest to concentrate on evidence from reefs which are known to bear z-coral faunas, and otherwise to make a careful distinction between reef evidence and coral evidence.

Biogeographical context It has long been accepted by coral workers that the Mediterranean region has had its own strong biogeographical identity from at least the early Neogene onwards, based mainly on endemism at speciWc and sometimes

Palaeoclimatic implications of the energy hypothesis

generic level, especially of the reef corals (Chevalier, 1977; Rosen, 1988; Rosen & Smith, 1988). On its western side, the Mediterranean coral fauna was apparently distinct from that of the Caribbean and western Atlantic due at least in part to progressive widening of the Atlantic. As a result, a marine barrier progressively developed between coastal regions on either side of the Atlantic. To the east, Indo-PaciWc reef corals had also become distinct from those of the Mediterranean during the Miocene, when the Tethyan seaway connections between the two were severed by emergent land, following closure of the Arabian peninsula against Eurasia. For the reef corals in particular, this history has been treated in further detail by Rosen & Smith (1988); Rosen (1988); and McCall et al. (1994). As a result of these regional events, the Mediterranean region apparently became inaccessible to potential colonisation by reef coral taxa from elsewhere, and likewise, its own taxa could not emigrate, or extend their ranges beyond the region. In other words, for present purposes, the Mediterranean region can be regarded broadly as a biogeographically closed system for much of the Neogene until the present day. Nevertheless, this is only a working model which still requires more detailed investigation. For instance, a thorough taxonomic investigation of coral endemicity is needed since there are no recent studies based on modern taxonomic methods applied to well-dated sets of sample taxa from all three major biogeographical regions. The results should then be compared with more recent models of the history of seaway closures between the Mediterranean and the Indo-PaciWc. Ro¨gl (1998; and in this volume) has indicated a more complex history than the simple model above, with reopening of seaways in the Langhian–Serravallian. According to the particular time within this interval, a number of diVerent seaways provided Indo-PaciWc connections to some parts of Europe but not to others. On this basis, we would expect more complex endemicity patterns in the Mediterranean and regions to the east, than are currently known for corals of this age. Of more direct importance for climatic considerations, is the fact that the Mediterranean reef- and z-coral fauna diminishes markedly in richness from the late Oligocene onwards (Chevalier, 1962; 1977; and see Fig. 15.2). This strongly distinguishes it from the other major biogeographical regions, where, for instance, the richness of z-corals in the Indo-West PaciWc focal region increased at least fourfold at around the Palaeogene–Neogene boundary and continued to increase thereafter (Wilson & Rosen, 1998). Even though much taxonomic revision is necessary, the progressive impoverishment of the Mediterranean is such a strong and widely accepted pattern, that it is unlikely to be qualitatively contradicted by further faunistic studies.

313

Palaeoenvironments: non-mammalian evidence

314

[Figure 15.2] Generic richness of Neogene–Recent zooxanthellate scleractinian corals of the Mediterranean region, and inferred minimum marine palaeotemperatures. Ages for the bases of stratigraphical divisions are given in Ma. Geological classification, and plotted data are scaled to time except for ‘modern fauna’. Plot 1 shows the diversity trend for the whole region for comparison with local area data in Table 15.1. See Wilson & Rosen (1998, fig. 9) for further details of data and methods. Plot 2 shows ranges of minimum inferred palaeotemperatures for various local areas, denoted by numbers corresponding to those in Table 15.1. Inferred temperatures are plotted as rectangles or vertical bars, whose dimensions represent ranges of uncertainty (see text). The z-coral richness value used for the modern ‘inferred temperature’ (A of Plot 2) is 4 genera (as in Plot 1). The modern temperature value (B of Plot 2) is a mid-point based on the coldest month range across the Mediterranean (12.8–15.6 °C), and the warmest month range (23.9–28.1 °C) respectively. Data for this are from Hydrographic Office United States Navy (1944).

This coral pattern has therefore been taken here as a reasonable basis for a broad palaeoclimatic analysis.

Methods

Taxa It is Wrst necessary to derive richness values of z-corals, in this case, for a range of Mediterranean localities through the Miocene. For the present

Palaeoclimatic implications of the energy hypothesis

study, the raw data have been derived from lists of local fossil coral faunas as shown in Table 15.1, and these are assumed, neutrally, to consist of a combination of both z- and az-corals. Some of the source works consist of descriptive taxonomy, others are simply check lists included within other kinds of study, though in most of the latter cases, the coral identiWcations have been made by coral taxonomists and are therefore taken to be reliable within the usual range of taxonomic subjectivity. Although descriptions of the corals are lacking in these latter works, this is compensated by their provision of recently revised ages of the coral-bearing formations treated therein. Ideally, the present study would be based on species. Fraser & Currie’s (1996) plot, however, is based on genera (Fig. 15.1), and this follows widespread practice by coral specialists. The prime reason why genera are preferred to species appears to be the relative stability of scleractinian coral identiWcations at generic level, in contrast to the considerable instability and disagreement between authors for species level treatments. Coral species are notoriously variable at the intraspeciWc level due to ecophenotypic factors and frequently show morphological overlap between species.

Extraction of z-coral data Since a fossil coral fauna might consist of a combination of both z- and az-corals (see above), each element of a given fauna must be assigned to its appropriate ecological category (z or az). Richness values can then be extracted for the z-coral component, since it is only these corals that are relevant to the method used here (below). The algal symbionts of corals, however, are not preserved in fossil samples, so direct evidence of symbiosis is lost and must be inferred indirectly. The protocol used for this is summarised by Wilson & Rosen (1998) whose Neogene Mediterranean data (in their Fig. 9) have also been used here (Plot 1, Fig. 15.2). Further information, including details of the criteria themselves, together with examples of their application to various parts of the fossil scleractinian record, are given by Rosen (in press).

The richness–temperature curve The energy hypothesis provides a richness–temperature relationship and is the present basis for making quantitative inferences of palaeotemperature from taxonomic richness patterns of fossil z-coral faunas. This follows an earlier study based on fewer modern data (Rosen, 1977, and see above). Palaeoclimatic information can also be derived from corals in other ways

315

Messinian

Messinian

early Messinian

Tortonian–Messinian

Tortonian

Late Badenian Serravallian

Langhian–Serravallian

Langhian (‘early Badenian’) Langhian (‘early Badenian’) Langhian

-15-

-14-

-13-

-12-

-11-

-10-9-

-8-

-7-6-5-

Age Miocene stages

16.4–14.8 16.4–14.8 16.4–14.8

16.4–11.2

14.0–13.0 14.8–11.2

11.2–7.1

11.2–5.3

7.1–6.2

7.1–5.3

6.9–5.8*

Absolute age (Ma) Time scale from Berggren et al., 1995; Ro¨gl, 1998

NW Bulgaria N–N. Central Hungary SW France

SW France

N–N. Central Hungary SW France

Cyprus

NE Rif of N. Morocco (Melilla, on Med. coast ca 250 Km ESE of Tangier) Western Orania, NW Algeria: Med. coast: Sebaa Chioukh Hills and Traras margin Livornesi Mts, Toscana, central Italy Baixo, Porto Santo, Madeira

Locality/area

Wijsman Best & Boekschoten, 1982 Follows, 1992; Follows et al., 1996 Oosterbaan, 1990 Cahuzac & Chaix, 1996, table 6 Cahuzac & Chaix, 1996, table 7 Pisera, 1996 Oosterbaan, 1990 Cahuzac & Chaix, 1996, table 6

9–9 (9) 12–13 (13) 10–17 (24)

9–14 (22)

4–5 (5) 3–8 (13)

3–3 (3)

6–10 (11)

4–5 (5)

5–5 (5)

Saint Martin & Corne´e, 1996, table 1 Bossio et al., 1996

2–2 (2)

Saint Martin & Corne´e, 1996, table 1*

References

Generic richness of z-corals Min. and max. (in relation to total)

18.6 °C 18.9–19.0 °C 18.8–19.4 °C

18.6–19.2 °C

17.4–17.7 °C 16.9–18.5 °C

16.9 °C

18.0–18.8 °C

17.4–17.7 °C

17.7 °C

16.4 °C

Prevailing surface temperatures of sea water Min. temperature ranges inferred from corals

Table 15.1. Selected Miocene scleractinian coral faunas of the Mediterranean region with inferred palaeotemperatures

23.8–20.5

-1SW France

Cyprus

SW France

NW Red Sea, Egypt

Locality/area Purser et al., 1996 (for localities); Rosen unpublished for coral records Cahuzac & Chaix, 1996, table 5 Follows, 1992; Follows et al., 1996 Cahuzac & Chaix, 1996, table 4

References

24–30 (36)

14–15 (15)

28–33 (41)

8–8 (8)

Generic richness of z-corals Min. and max. (in relation to total)

19.9–20.4 °C

19.2–19.3 °C

20.1–20.5 °C

18.5 °C

Prevailing surface temperatures of sea water Min. temperature ranges inferred from corals

Localities have been chosen to give a reasonable stratigraphical and geographical representation through the Miocene based on recent coral literature. Ranges in values for numbers of z-corals for a given area arise from the approach used here to infer symbiosis in fossil corals (see text). Temperatures have been inferred from richness of z-coral faunas by taking the square roots of z-coral richness ranges, and reading back the mininum possible temperature from Fraser & Currie’s (1996) plot (see Fig.1). See Fig.2 for plot of inferred temperatures against absolute age. * Author’s own radiometric dating.

Aquitanian

Aquitanian–early Burdigalian 23.8–17.3

-2-

20.5–16.4

Burdigalian

-3-

20.5–14.8

Burdigalian–Langhian

Absolute age (Ma) Time scale from Berggren et al., 1995; Ro¨gl, 1998

-4-

Age Miocene stages

Table 15.1 (cont.).

Palaeoenvironments: non-mammalian evidence

318

(e.g. Wells, 1967; Barta-Calmus, 1977; Rosen, 1977) but these are not attempted here. The modern data used for inferring palaeotemperatures from taxonomic richness are shown in Fig. 15.1, after Fraser & Currie (1996). These authors state that they obtained a signiWcant correlation between richness and temperature, but their plot consists only of the scattered points and does not include the regression line derived from them, to which they refer in their text. Such a regression line might be used to ‘read back’ temperatures from known richness values, but since sampling is rarely complete or uniform in this kind of data, the merits of using a regression line in this way are questionable. Instead, a subjective curve of maximum diversity has been drawn here (Fig. 15.1) on their plot, as an envelope joining maximum realised richness values for each temperature value (cf. Rosen, 1971; Wg. 5). Note that for any given prevailing temperature, there are a set of locations whose observed richnesses range downwards from a maximum realised value, eVectively as anomalies or shortfalls. Such shortfall localities might be explained by numerous ecological factors other than temperature, including depth and turbidity mentioned above, as well as other local conditions. Non-ecological factors might also be important, including incomplete preservation and incomplete sampling, either of the bulk fauna as a whole, or with respect to particular habitats within a sample area. The present approach is also non-rigorous in the sense that no attempt has been made to standardise for sample area, which would in any case be diYcult – but see Fraser & Currie’s (1996) discussion of their parameter, ‘length of coastline containing reefs‘. The fractal nature of many coral habitats also makes it very diYcult to standardise for habitat heterogeneity. For palaeoclimatic inference from corals, the richness–temperature curve is used here ‘in reverse’ to ‘read back’ prevailing sea water palaeotemperatures for a particular locality or area from the richness of its fossil z-coral fauna. Since Fraser & Currie converted richness values to their square roots (Fig. 15.1), the same has to be done for the present fossil data, simply to facilitate linear readings from their plot. (Had their raw data been available the better alternative would have been to replot them on nonexponential axes.) The problem of richness shortfall at all fossil localities has to be borne in mind, since there is no simple direct way of knowing whether or not the richness of a recorded fauna is close to its potential empirical maximum. True richness might have been higher than that of the recorded fauna for any number of the reasons mentioned already, hence the inferred temperature must be a minimum value. In fact, since the fossil record has to be assumed to be incomplete, as a generality, all richness values, and hence

Palaeoclimatic implications of the energy hypothesis

also the palaeotemperatures inferred from them, are likely to be too low. Note that the theoretical maximum possible palaeotemperature that can be inferred from a given fossil fauna is the same for any sample fauna, regardless of its actual recorded richness value, and this corresponds to the maximum temperature (c. 29 °C) of the whole curve (Fig. 15.1). It also follows from all of the above discussion that absence of z-corals does not in itself indicate palaeotemperatures below the minimum on the curve (14 °C) because absence may be explained by a whole range of factors other than temperature. In using the richness–temperature curve, there is also an implicit assumption that the absolute values shown in Fig. 15.1 can also be applied to the past. However, since we know that global richness of corals, as with all other taxa, has varied through geological time (e.g. Rosen, in press), this assumption is equivalent to assuming that global diversity through time has also been controlled largely by prevailing temperatures, in this case presumably, global rather than local temperatures. Further discussion of this problem is beyond scope here but should be borne in mind when judging present results. An attempt to make a fuller and more rigorous application of the energy hypothesis to palaeoclimatology must be left to future work. It is possible, for instance, that statistical techniques might be used to alleviate some of the above problems.

Results Results are shown in Table 15.1 and Fig. 15.2. Apart from the modern data, the inferred palaeotemperature values on Table 15.1 are plotted in Fig. 15.2 (Plot 2), with their respective estimated ranges of uncertainties shown as rectangles which represent combined ‘error bars’ for both axes (time and temperature). Age variability for the fauna at a given locality represents either the degree of resolution of the known age, or the actual known age range of the coral horizons concerned. Inferred temperature variability for each fauna reXects the maximum and minimum estimated richness of z-corals occurring in that particular fauna (see above). In some cases however, the maximum and minimum are the same (see Table 15.1), in which case the plotted point is shown only as a line, rather than a rectangle, denoting age uncertainty only. Although the estimated number of z-corals at a locality ranges from a minimum to maximum, and inferred palaeotemperatures reXect this, the plotted temperatures for both these values are minima in the more global context of the temperature–richness curve of Fig. 15.1 (see Methods, above).

319

Palaeoenvironments: non-mammalian evidence

320

Fig. 15.2 shows a clear trend of accelerating temperature decrease after the Burdigalian, broadly suggesting that the whole region was subjected to climatic cooling from at least that time onwards to the Quaternary. The net temperature drop is 4.5 °C. However, the modern ‘inferred temperature’ (A of Plot 2) shows a marked shortfall against the known mid-point temperature for the region, suggesting either that factors other than climate might be important in that particular case, or that the temperature–richness curve is useful only for reconstructing relative palaeotemperature patterns, rather than absolute ones. Both factors might also apply. This is discussed further below.

Discussion The following discussion concentrates on the temperature patterns inferred from the numerical data of coral richness but also considers whether some of the results might be explained by biogeographical factors, especially increasing isolation of the Mediterranean region, rather than, or additionally to, climatic trends. However, a fuller exploration of this requires an analysis of the faunal compositional changes through time as well as quantitative changes in richness. Decline in richness might be due to increased extinction rates, or decreased origination rates, or both, but such an analysis is outside the present scope. The overall descending temperature trend of Plot 2 in Fig. 15.2 broadly matches global palaeotemperature curves derived from oxygen isotopes (e.g. Savin & Douglas in Dodd & Stanton, 1990, Wg. 3.26). The absolute temperatures are diVerent, however, in that the coral-derived values are higher by about 13 °C, even though they represent minima. This diVerence may be due to a much higher latitude (and/or deeper water) data source for the isotope curves compared with the Mediterranean. In any case, whatever the reason, the isotope-based temperatures lie entirely below the minimum known temperature limit for modern z-corals. Further possible reasons connected with the nature of isotopic estimates of temperature in comparison with those suggested by palaeoecological analysis of foraminifera, corals and mangroves have been discussed by Adams et al. (1990). Further discussion (below) concentrates on comparing the relative temperature patterns rather than absolute temperatures. In closer detail, the global isotopic temperature pattern shows a warming trend of about 1.5 °C in the early Miocene that reaches a peak around the Burdigalian–Langhian boundary. Although the present coral-derived palaeotemperatures do not have this quality of resolution, Plot 2 of Fig. 15.2

Palaeoclimatic implications of the energy hypothesis

does hint at a slight warming in the early Miocene (Burdigalian) of about 0.5–1.0 °C, which visually matches the isotope curve over this time interval surprisingly well. Immediately after the Burdigalian, the isotope curve shows a warm peak at the Burdigalian–Langhian boundary, whereas the coral palaeotemperature peak is in the Burdigalian. This slight diVerence might only reXect insuYcient coral data, dating or methodological problems. After peaking, both palaeotemperature patterns show sharp mid- to late Miocene declines over the same temperature range (4.5 °C), but whereas the isotope curve fall lies almost entirely within the earlier part of the mid-Miocene, the coral pattern shows a relatively slow start of declining temperature from the start of the mid-Miocene and the main fall within the late Miocene. Thus the main fall has a time lag of about 5 Ma in relation to the isotope curve. This may reXect regional diVerences in data sources, methodological problems or perhaps the Mediterranean region remained regionally warmer than global conditions overall. Perhaps too, there was a time-lag in the inXuence of climatic cooling on z-coral richness, such that faunas, though beginning to decline in richness at this time, remained ‘over-rich’ for a while in relation to prevailing temperatures, so giving rise to higher inferred palaeotemperatures from the corals than would otherwise be expected. After the steep temperature decline in the mid-Miocene, the isotopic curve shows a slight warming through the late Miocene until the midPliocene. This pattern is not seen at all in the coral evidence, as already discussed for the diVerences in the mid-Miocene patterns. The combined variability of palaeotemperatures shown in Plot 2 of Fig. 15.2, for any given time interval, probably reXects a combination of all the local ecological and sampling (etc.) factors already discussed. In addition, however, the region is a large one and some of the variability might also result from sampling from diVerent parts of it, perhaps reXecting climatic diVerences analogous to those seen today within the region. With suYcient data in the future, it should be possible to test for geographical controls on palaeotemperatures by considering sets of geographically widespread localities of the same age, within the Mediterranean region. The diVerence between the modern ‘inferred temperature’ and actual temperature (A and B respectively in Plot 2) has two possible implications. Firstly, it might be taken to suggest that the temperature–richness curve (Fig. 15.1) consistently leads to underestimates of palaeotemperatures (at least in the Miocene of the Mediterranean region) by about 2.5 °C. This should not be surprising as the method provides minimum estimates only, as already discussed above. An imaginary line drawn through all the minima

321

Palaeoenvironments: non-mammalian evidence

322

of Plot 2 can be regarded as a constraint of minimum possible realisable conditions through the Miocene, but there is currently no obvious way of knowing how far true conditions exceeded these temperatures. If a regression line can be obtained for the modern data in Fig. 15.1, this could be used to provide mid-point estimates of palaeotemperature, which could be used instead of (or in addition to) the minima currently yielded by the envelope curve of Fig. 15.1. Secondly, and perhaps oVsetting the Wrst implication, it is possible that the shortfall in the modern temperature estimate also reXects the relative isolation of the modern Mediterranean from z-coral regions elsewhere. In other words, while falling temperatures during the late Cenozoic (and presumably also the Messinian desiccation events) caused falling z-coral richness during this time, increasing isolation may have played a part in limiting recolonisation of the region by z-corals from elsewhere when temperatures rose again, e.g. in the Pliocene and during the Holocene glacial retreat. It is also useful to compare the present data with the reef patterns discussed by Esteban (1996). In the early Miocene (Aquitanian) of Europe, the northern limit of z-coral formations (including coral reefs) follows an approximate line from southwestern France to southern Turkey, and the regional richness for this interval is about 50 genera (Plot 1, Fig. 15.2). The limit shifted northwards in the mid-Miocene, during which regional diversity decreased to about 30. The northernmost margins consist of low diversity z-coral communities in the Vienna Basin and northwestern Bulgaria (Esteban, 1996; Pisera, 1996). The latter author’s data (No.7) are shown in Plot 2, Fig. 15.2. By the late Miocene (Tortonian and early Messinian), the northern limit had withdrawn southwards to a line which approximately follows the present northern coastline of the Mediterranean, and diversity had fallen to 10 genera or fewer according to horizon and location within that interval. The northernmost coral reefs at this time occur in Toscana between Livorno and Siena (Esteban, 1996; Bossio et al., 1996). Data from the latter authors are shown in Plot 2 of Fig. 15.2 (No.13). Four diVerent factors, in various combinations, have been invoked in the literature to explain these distribution patterns, especially the overall trend of decline in z-coral diversity: (1) sea level changes, (2) increasing physical and biogeographical isolation of the whole region from the main coral areas elsewhere in the world, (3) regional temperature changes, and (4) northward plate movement of the whole region over this time. Climate is interwoven with all of these, either as cause or eVect. Esteban (1996) notes that most reefs (and hence their associated z-corals) occur in three major high sea-level phases corresponding to the above intervals

Palaeoclimatic implications of the energy hypothesis

(Aquitanian, Langhian–Serravallian, and Tortonian–Messinian), reXecting transgressive phases of second order sea-level cycles. In addition Esteban has discussed the inXuence of more local factors such as evaporation and water circulation. When the isotopic temperature curve, the reef patterns and z-coral patterns are compared, the mid-Miocene is again problematic. At the start of the middle Miocene, the warm peak of the isotope curve is consistent with the northward shift of the reef belt, both suggesting warmer conditions than the coral richness evidence. Regional and local z-coral richness seems to have already begun to decline at this time (Fig. 15.2). This anomaly might be attributable solely to dating resolution of the coral and reef evidence, but might also be explained by the onset of biogeographical isolation of the Mediterranean. Seaway interconnections to the east (Indo-PaciWc) had closed at least once by the start of the mid-Miocene, and the extent to which postulated re-openings actually aVected corals are still unknown (see above). By the late middle Miocene a reverse anomaly emerges, with the sharp fall in the isotope curve implying much cooler conditions than either the coral or reef evidence suggest. Dating resolution may be important again here, and the idea of a time-lag in richness decline of the corals has also been mentioned above. Finally, with respect to climatic inXuence on organisms in general, both terrestrial and shallow aquatic, it is worth noting that at the present day, more temperate conditions are not simply cooler, but actually show much greater seasonality of temperature than tropical climates. While the coral data themselves give no direct indication of seasonality, it is likely that the Miocene climatic trend inferred from them was accompanied by the onset of increasingly seasonal climates in the Mediterranean region, perhaps not unlike those seen across the region today. In the context of this volume, one would expect the onset of strongly seasonal conditions by the late Miocene to have had important consequences for changes in both vegetation and mammalian faunas.

Summary The ‘energy hypothesis’ provides a working model of the global relationship between taxonomic richness (in this case of extant zooxanthellate coral genera) to prevailing temperatures, and has been used here in a preliminary study to infer minimum palaeotemperatures for Mediterranean waters during the Miocene. The z-coral faunas of Wfteen Miocene localities suggest that climatic conditions in the early Miocene were warmest, with a possible

323

Palaeoenvironments: non-mammalian evidence

324

slight increase during that time. From the middle Miocene, climate cooled at an accelerating rate through about 4.5 °C. As a general trend, this broadly matches the evidence of oxygen isotopes, but the coral-derived cooling lags behind the isotope curve by about 5 Ma, perhaps due to biological time-lag in response of coral richness to the eVects of global cooling. Evidence from coral reef distributions broadly matches that both from corals and isotopes in the early and late Miocene, but there are anomalies between all three during the middle Miocene. At least some of these may be due to dating problems, though increasing marine biogeographical isolation of the Mediterranean may have been important too. In general, the energy hypothesis yields a surprisingly good visual match to gross palaeoclimatic trends known from other independent evidence, and the method should be further developed with more rigour in future studies.

Acknowledgements I thank two informal referees for their helpful discussion, and for providing suggestions for the paper: Ken Johnson (Scripps Institution, California) and Bonnie O’Brien (University of Georgia at Athens).

References Adams, C. G., Lee, D. E. & Rosen, B. R. 1990. ConXicting isotopic and biotic evidence for tropical sea-surface temperatures during the Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology, 77, 289–313. Barta-Calmus, S. 1977. Aperc¸u de l’e´volution des Madre´poraires dans la province me´diterrane´enne occidentale au Nummulitique. In Second Symposium International sur Coraux et Re´cifs Coralliens Fossiles, Me´moires du Bureau de Recherches Ge´ologiques et Minie`res, 89, 507–17. Berggren, W. A., Kent, D. V., Swisher, C. C., III & Aubry, M.-C. 1995. A revised Cenozoic geochronology and chronostratigraphy. In Geochronology Time Scales and Global Stratigraphic Correlation, Berggren, W. A., Kent, D. V., Aubry, M.-C. & Hardenbol, J. (eds.). SEPM Special Publication 54, 129–212. Bossio, A., Esteban, M., Mazzanti, R., Mazzei, R. & Salvatorini, G. 1966. Rosignano reef complex (Messinian), Livornesi Mountains, Tuscany, Central Italy. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.), SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 277–94. Cahuzac, B. & Chaix, C. 1996. Structural and faunal evolution of Chattian-Miocene reefs and corals in western France and the northeastern Atlantic Ocean. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy,

Palaeoclimatic implications of the energy hypothesis

J.-M. (eds.). SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 105–27. Chevalier, J.-P. 1962. Recherches sur les Madre´poraires et les formations re´cifales mioce`nes de la Me´diterrane´e occidentale. Me´moires de la Socie´te´ Ge´ologique de France (Nouvelle Se´rie), 40 (93), 1–562. Chevalier, J.-P. 1977. Aperc¸ u sur la faune corallienne re´cifale du Ne´oge`ne. In Second Symposium International sur Coraux et Re´cifs Coralliens Fossiles, Me´moires du Bureau de Recherches Ge´ologiques et Minie`res, 89, 359–66. Dodd, J. R. & Stanton, R. J., Jr. 1990. Paleoecology; Concepts and Applications. 2nd Edition. John Wiley & Sons, New York. Esteban, M. 1996. An overview of Miocene reefs from Mediterranean areas: general trends and facies models. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.). SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 3–53. Follows, E. J. 1992. Patterns of reef sedimentation and diagenesis in the Miocene of Cyprus. Sedimentary Geology, 79, 225–53. Follows, E. J., Robertson, A. H. F. & ScoYn, T. P. 1996. Tectonic controls on Miocene reefs and related carbonate facies in Cyprus. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.). SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 295–315. Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.) 1996. Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions. SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5. Fraser, R. H. & Currie, D. J. 1996. The species richness-energy hypothesis in a system where historical factors are thought to prevail: coral reefs. The American Naturalist, 148, 138–59. Hydrographic OYce of the United States Navy, 1944. World Atlas of Sea Surface Temperatures. Second Edition. Hydrographic Publication No. 225. The Hydrographic OYce of the United States Navy, Washington, D.C. McCall, G. J. H., Rosen, B. R. & Darrell, J. G., 1994. Carbonate deposition in accretionary prism settings: Early Miocene coral limestones and corals of the Makran mountain range in southern Iran. Facies, 31, 141–78. Oosterbaan, A. F. F. 1990. Notes on a collection of Badenian (Middle Miocene) corals from Hungary in the National Museum of Natural History at Leiden (The Netherlands). Contributions to Tertiary and Quaternary Geology, 27, 3–15. Perrin, C., Bosence, D. W. J. & Rosen, B. R. 1995. Quantitative approaches to palaeozonation and palaeobathymetry of corals and coralline algae in Cenozoic reefs. In Marine Palaeoenvironmental Analysis from Fossils, Bosence, D. W. J. & Allison, P. A. (eds.). Geological Society of London Special Publications, 83, 181–229. Pisera, A. 1996. Miocene reefs of the Paratethys: a review. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.). SEPM [Society

325

Palaeoenvironments: non-mammalian evidence

326

for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 97–104. Purser, B. H., Plaziat, J.-C. & Rosen, B. R., 1996. Miocene reefs of the northwest Red Sea. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.). SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 347–65. Ro¨gl, F. 1998. Palaeogeographic considerations for Mediterranean and Paratethys seaways (Oligocene to Miocene). Annalen des Naturhistorischen Museums in Wien, 99A, 279–310. Roniewicz, E. & Morycowa, E. 1993. Evolution of the Scleractinia in the light of microstructural data. In Proceedings of the VI International Symposium on Fossil Cnidaria and Porifera held in Mu ¨ nster, Germany, 9–14 September 1991, Oekentorp-Ku ¨ ster, P. (ed.), Courier Forschungsinstitut Senckenberg, 164, 233–40. Rosen, B. R. 1971. The distribution of reef coral genera in the Indian Ocean. In Regional Variation in Indian Ocean Coral Reefs, Stoddart, D. R. & Yonge, C. M. (eds.). Symposium of the Zoological Society of London, 28, 263–99. Rosen, B. R. 1977. The depth distribution of Recent hermatypic corals and its palaeontological signiWcance. In Second Symposium International sur Coraux et Re´cifs Coralliens Fossiles. Me´moires du Bureau de Recherches Ge´ologiques et Minie`res, 89, 507–17. Rosen, B. R. 1984. Reef coral biogeography and climate through the late Cainozoic: just islands in the sun or a critical pattern of islands? In Fossils and Climate, Brenchley, P. J. (ed.). John Wiley and Sons, Chichester, Geological Journal Special Issues, 11, 201–62. Rosen, B. R. 1988. Progress, problems and patterns in the biogeography of reef corals and other tropical marine organisms. Helgola¨nder Meeresuntersungen, 42, 269–301. Rosen, B. R. (in press) Algal symbiosis, and the collapse and recovery of reef communities: Lazarus corals across the K-T boundary. In Biotic Response to Global Change: the Last 145 Million Years, Culver, S. J. & Rawson, P. A. (eds.). Cambridge University Press, Cambridge. Rosen, B. R. & Smith, A. B. 1988. Tectonics from fossils? Analysis of reef-coral and sea-urchin distributions from late Cretaceous to Recent, using a new method. In Gondwana and Tethys, Audley-Charles, M. G. & Hallam, A. (eds.). Geological Society of London Special Publications, 37, 275–306. Saint Martin, J.-P. & Corne´e, J.-J. 1996. The Messinian reef complex of Melilla, northeastern Rif, Morocco. In Models for Carbonate Stratigraphy from Miocene Reef Complexes of Mediterranean Regions, Franseen, E. K., Esteban, M., Ward, W. C. & Rouchy, J.-M. (eds.). SEPM [Society for Sedimentary Geology], Tulsa, Oklahoma. Concepts in Sedimentology and Paleontology, 5, 227–37. Stanley, G. D., Jr. & Cairns, S. D. 1988. Constructional azooxanthellate coral communities: an overview with implications for the fossil record. Palaios, 3, 233–42. Veron, J. E. N. 1995. Corals in Space in Time. The Biogeography and Evolution of the Scleractinia, University of New South Wales Press, Sydney. Wells, J. W. 1967. Corals as bathometers. Marine Geology, 5, 349–65.

Palaeoclimatic implications of the energy hypothesis

Wijsman Best, M. & Boekschoten, G. J. 1982. On the coral fauna in the Miocene reef at Baixo, Porto Santo (eastern Atlantic). Netherlands Journal of Zoology, 32, 412–18. Wilson, M. E. J. & Rosen, B. R. 1998. Implications of paucity of corals in the Paleogene of SE Asia: plate tectonics or Centre of Origin? In Biogeography and Geological Evolution of SE Asia, Hall, R. & Holloway, J. D. (eds.), pp. 165–95. Backhuys Publishers, Leiden, The Netherlands.

327

16 Contribution to the knowledge of Neogene climatic changes in western and central Europe by means of non-marine molluscs Daniela Esu

Introduction During the nineteenth century and the Wrst half of the twentieth century, a great number of papers on Cenozoic fossil European non-marine molluscs were published. After the Second World War the number of publications dealing with fossils of this group decreased and among the relatively limited number of papers published since then the majority regards malacological assemblages of the Quaternary period. Many faunas must be revised and consequently their utility for paleoclimatical, paleoecological or paleogeographical purposes is very limited. The potentiality of non-marine mollusc fossil assemblages to the reconstruction of paleoclimates has been clearly evidenced by many works of many malacologists working on the Quaternary (Lozek, 1964; Puisse´gur, 1976). The majority of the species recorded in sediments of this period (especially of middle and late Pleistocene and Holocene) are still living and a quantitative analysis of the composition of the fossil assemblages gives very accurate information about the local climatic conditions. This fact is strictly linked to the onset of the main glaciations just at the beginning of the middle Pleistocene. Unfortunately this situation is not the same for older periods in which the climatic changes were not so sharp and the majority of the fossils belong to extinct species. Since the shell morphology of land species generally is not aVected by climatic variations, the information about environmental changes is inferred indirectly considering other factors: the occurrence or disappearance of the taxa, their geographic distribution (also in connection with the latitude and elevation, for example, of living species of the same genus), the degree of the species diversity of the assemblages. The oligotypic character of the non-marine assemblages is typical of cold climatic phases; on the contrary the richness in genera and species is typical of warm climate (Esu et al., 1989). Among the non-marine molluscs two ecologically very diVerent groups must be distinguished: the aquatic and the land forms. The former give little information about the paleoclimate but more about the paleoenvironment (type of habitat, water salinity, nature of substratum). Anyway during the whole Miocene signiWcant paleoclimatic elements like Melanopsis, a warm climate aquatic prosobranch living today in Europe only in the southern

Contribution of non-marine molluscs

Iberian and southern Balkan peninsulas, were widespread in west and central Europe and also in the Pannonian and Dacian basins of the Paratethys. The land species give more information about paleoclimate and vegetation cover. For the reasons stated above we must use principally taxonomical groups of land molluscs at least partly revised like Helicoidea, Clausilioidea and some other taxa among Vertiginidae, Chondrinidae (Gastrocoptinae), Pupillidae, Strobilopsidae, Zonitidae, Oleacinidae to detect climatic variations through the time. As for the geographic distribution of the land molluscs many diVerent factors are responsible: climate is one of the most important, but also moisture level, nature of the substratum and physiography are important. Neogene non-marine mollusc assemblages are widespread all over Europe, but often their correct chronostratigraphic position is not well deWned. In several basins the continental molluscs are found associated to vertebrates. This opportunity is of great importance for a correct biochronology of the molluscan faunas. The biochronological scale adopted in this work is based on the European Mammal Neogene (MN) zones given by Mein (1975) according to mammal ages of Fahlbusch (1976) as stated by Mein (1990) and Bruijn et al. (1992). The chronological boundaries of the MN 1, MN 2 and MN 3 zones are taken from Schlunegger et al. (1997). A new detailed chronology for the middle to late Miocene MN zones, based on high resolution magnetostratigraphic data, is given by Krijgsman et al. (1996) who have dated seven successive MN zone boundaries (from MN 4/5 to MN 11/12 boundary) in continental sections of Spain bearing mammal faunas of two or more successive zones in superposition. These data and the MN 14/15 boundary as proposed by Aguirre et al. (1995) are adopted here. As for the boundary MN 4/5 diVerent proposals are made: Schlunegger et al. (1996) state the lower boundary of the MN 5 zone at the base of Langhian (at about 16.4 Ma); Alvarez Sierra et al. (1997) give for the MN 4/5 boundary an age of 15.975 Ma. The chronological framework used here is related to the up-to-date chronostratigraphical time scale proposed by Gradstein & Ogg (1996) (see Fig. 16.1). In this paper only the mollusc faunas having a chronostratigraphic signiWcance will be taken into account. Many European fossiliferous sites considered here, bearing signiWcant mollusc assemblages with Clausilioidea and Helicoidea, are referred to the MN zones by Nordsieck (1982). Particular consideration is made of the deposits of western and central Europe (Spain, France and Germany) where an almost complete Neogene continental record is available.

329

Palaeoenvironments: non-mammalian evidence

330

[Figure 16.1] Schematic chronostratigraphic relationships among European selected localities with non-marine molluscs, European Mammal Neogene (MN) zones (Mein, 1990) and mammal ages (Fahlbusch, 1976). Shaded intervals indicate the most significant Neogene malacofaunistic changes.

Contribution of non-marine molluscs

Data analysis of Miocene faunas

Early Miocene In western Europe from Agenian sediments outcropping at Paulhiac (Lot-etGaronne) (MN 1) (Wenz, 1936; Rey, 1968; Truc, 1971a) a land mollusc fauna is recorded in which Gastrocoptinae, Strobilopsidae, Zonitidae and Helicoidea are well represented. The most signiWcant taxa are Gastrocopta (Albinula) turgida (Reuss), Discostrobilops uniplicatus (Braun), Leucochroopsis leptoloma leptoloma (Braun), Wenzia ramondi (Brongniart) and ‘Cepaea’ subsulcosa (Thomae) (= Helicinae group, sensu Nordsieck, 1986). Some mollusc assemblages with vertebrates referred to MN 2a and MN 2b zones are recorded in De´p. Allier (Montaigu, Laugnac), in which land species known also from other coeval sites of central and southern France such as Palaeoglandina gracilis (Zieten), Discus (Discus) vireti Truc, ‘Cepaea’ moroguesi (Brongniart), Omphalosagda subrugulosa (Quenstedt), prevail (Truc, 1971c; Cavelier, 1972). At Verrie`res (Haute-Chaıˆne, middle Jura, France), continental sediments overlying the lower Burdigalian transgressive marine sediments contain some Helicidae faunas of late Burdigalian age (Rangheard et al., 1985) characterized by the occurrence of the land gastropods Megalotachea turonensis (Deshayes) and ‘Cepaea’ silvana silvana (Klein). The former is present in several coeval western European outcrops (Nordsieck, 1986) including Spain (Sant Pere de Ribes, Valle´s Penede´s, Barcelona) (Calzada, 1990), the latter is widespread in France and Germany from early Miocene (MN 3) to late Miocene (MN 9). Rich Orleanian continental mollusc assemblages are widespread in the Rhon ˆ e basin (France) where many outcrops yielded freshwater and terrestrial gastropods often associated with mammals. These assemblages are characterized mainly by land species belonging to the families Pomatiasidae, Ellobiidae, Oleacinidae and to the superfamily Helicoidea. In Provence the well known outcrops in the basin of Cucuron (Vaucluse), Mirabeau Les Pardigons, Champ de Mathieu, Sepulture Chre´tien, yielded rich assemblages of terrestrial species. The mammal fauna of these deposits belongs to the La Romieu mammal zone (MN 4) (Mein et al., 1971; Truc, 1971a,b). Many paleoclimatic signiWcant taxa are spread in this area: Tudorella draparnaudi (Matheron), Palaeoglandina aquensis (Matheron), Pseudoleacina (Paraglandina) christoliana (Matheron), Megalotachea turonensis, Schlickumia aquensis (De Serres), Leucochroopsis pisum (Matheron). Some of these species are also recorded in the Orleanian deposits of the basin of Digne (Pont d’Aiguines, Alpes de Haute Provence, Var)

331

Palaeoenvironments: non-mammalian evidence

332

with vertebrate remains referred to the MN 4 (upper part) and MN 5 (basal part) zones (Gigot et al., 1976). In Germany non-marine molluscs of early Miocene age have been discovered associated with mammals in several sites. Among them the assemblages collected at Rottenbuch (Bayern) and OVenbach (Tempelsee, Hessen) are assigned to MN 1, and those collected at Budenheim (RheinlandPfalz), Wiesbaden (Hessen) and Ulm (Baden-Wu ¨ rttemberg) are assigned to MN 2 (Nordsieck, 1982). Other assemblages, recorded in strata interbedded or correlated with marine sequences, are indirectly assigned to the MN zones: Hauchenberg (Bayern) to MN 3 and Pfa¨nder (Wirtatobel, Vorarlberg, Austria) to the MN 3/4 boundary (Nordsieck, 1982). The molluscs recorded in these sites were studied many years ago (Wenz, 1923–30; PfeVer, 1929) and need systematic revision, but some important groups such as Clausilioidea and Helicoidea have been revised (Nordsieck, 1981a,b, 1986): the genera Constricta, Canalicia, Eualopia, Leucochroopsis, Protodrepanostoma, Galactochilus, Tropidomphalus, Klikia, Cyrtochilus, Pseudochloritis, Titthodomus are present. In particular a very rich assemblage is present at Budenheim (MN 2b) in which many thermophilous terrestrial elements occur, like Gastrocopta (Albinula), G. (Sinalbinula), Leiostyla (Leiostyla), Negulus, Strobilops (Strobilops), Palaeoglandina in addition to the thermophilous aquatic genus Melanopsis (Wenz, 1923–30). At Theobaldshof (Rho¨n, Hessen) referred to MN 2 zone by Nordsieck (1982), but which is perhaps a little younger (MN 3 or MN 4) on the basis of micrommammal remains, a rich assemblage of molluscs with archaic elements is present: Carychiopsis schwageri (Reuss), Eostrobilops Wscheri (Wenz), Discus (Discus) euglyphus (Reuss), Pseudoleacina (Paraglandina) oligostropha (Reuss), Eualopia bulimoides Thomae, Laminifera mira (Slavik), Leucochroopsis apicalis (Reuss), Titthodomus koeneni (Fischer & Wenz) (Moayedpour, 1977). In the site of Tuchorice (Bohemia, Czech Republic) the deposits referred to the MN 3 zone yielded a very rich mollusc fauna recorded by Reuss (1849) and Wenz (1923–30) and partially revised by Nordsieck (1981a,b) and Falkner (1986). Many signiWcant taxa are present: Negulus raricostatus (Slavik), Gastrocopta (Albinula) sp., Strobilops (Strobilops) sp., Pleurodiscus falciferus (Boettger), Janulus densestriatus (Klika), Serrulella schwageri (Boettger), Serrulastra amphiodon (Reuss), S. polyodon (Reuss), Constricta tenuisculpta (Reuss), Cochlodina perforata (Boettger), Laminifera mira, Canalicia attracta (Boettger), Protodrepanostoma nordsiecki Falkner, Helicodonta hecklei (Klika), Pseudomonacha zippei (Reuss).

Contribution of non-marine molluscs

Paleoclimatic remarks During the basal part of the early Miocene (MN 1, MN 2a, MN 2b zones) a certain number of late Oligocene warm climate genera among Clausilioidea and Helicoidea, such as Constricta, Canalicia, Eualopia, Wenzia, Leucochroopsis, Protodrepanostoma, Galactochilus, Tropidomphalus, Klikia, Cyrtochilus, Pseudochloritis, are still present in deposits of western and central Europe. At the beginning of the Burdigalian age the molluscan assemblages recorded in deposits with mammal faunas of the MN 3 zone present a higher diversity. In western and central Europe many new genera and species of Clausilioidea and Helicoidea appear in deposits bearing mammal fauna referred to the MN 3 and MN 4 zones: Serrulella (with several species), Serrulastra, Cochlodina, Megalotachea, Schlickumia and new species of Leucochroopsis, Protodrepanostoma and ‘Cepaea’ group (Truc, 1971b; Nordsieck, 1981a,b, 1986). This fact and the spreading in the assemblages of tropical, subtropical and temperate warm forestal elements, as Negulus (at present living in Ethiopia, eastern Africa, its presence in Malawi is not conWrmed by Bruggen, 1994), Leiostyla (Leiostyla) (Madeira, Canary Is., Azores, Algeria and sporadically in western Europe as relict), Strobilops (Strobilops) (eastern N America, mainly south of 52° N, central America, West Indies and northern S America), Discostrobilops (NE Mexico, central America, Jamaica, Cuba, Bahamas Is., Bermuda Is.), Eostrobilops (East Asia south of 40° N), Janulus (Canary Is. and Madeira), Pleurodiscus (circum-Mediterraneum) (Zilch, 1959–60; Manganelli et al., 1989, 1990), is indicative of a general warm climatic phase and of the presence of diVuse moist forestal biotopes during the early Miocene in western and central Europe. The general warm climatic phase during the early Miocene in Europe is also testiWed to by the presence in central southern Poland (Belchato´w), at about 51° N, between 18.1 ± 1.7 and 16.5 ± 1.3 Ma (Belchato ´ w-B) and below 18.1 Ma (Belchato ´ w-C), of a rich land snail fauna documented by Stworzewicz (1993, 1995), composed mainly of Cyclophoridae, Diplommatinidae, Vertiginidae, Chondrinidae (Gastrocoptinae), Subulinidae and Strobilopsidae, comprising species belonging to genera living in subtropical or tropical areas: Negulus, Gastrocopta (Sinalbinula) (Caucasus, Eastern Africa, Eastern China, Korea, Japan, Polynesia, Micronesia, Australia, Hawaii), Strobilops (Strobilops) and genera no longer present in Poland but living in the circum-Mediterranean regions or in the Caucasus as Cochlostoma, Caspicyclotus, Pomatias, Tudorella.

333

Palaeoenvironments: non-mammalian evidence

334

Middle Miocene In western Europe scanty data are available for the non-marine molluscs of deposits referred to the MN 5 zone. In central Europe at Undorf (Regensburg, Bayern) assigned to the lowermiddle part of MN 5, its deposits being younger than those of Langenmoosen, Bayern (cf. Fahlbusch, 1964, 1975; Nordsieck, 1982) a very rich mollusc assemblage is recorded in which many land thermophilous taxa are present (Table 16.1) (Wenz, 1923–30; Schlickum, 1976; Falkner, 1986; Nordsieck, 1981b). Similar assemblages are present in other localities of Austria and Germany (Rein in Steiermark, Mo¨rsingen in Baden-Wu ¨ rttemberg) (Wenz, 1923–30; Nordsieck, 1982). In France the classic site of Sansan (Gers, basin of Aix-en-Provence), well known for the presence of the early Astaracian vertebrate fauna (MN 6), yielded a rich freshwater and land mollusc fauna studied at the beginning of the twentieth century by Dollfus (1915) and partially revised in the past thirty years by Truc (1971a,b) and Fischer (in press). The assemblage is characterized by several palustrine species and by land snails; selected taxa are listed in Table 16.1. From outcrops of the same age in the surroundings of Sansan, aquatic prosobranchs of warm climate, such as Melanoides aquitanicus (Noulet) and Melanopsis kleini Kurr, were recorded (Dollfus, 1915). Few data are available for Iberian peninsula: a Portuguese fauna, referred to MN 6, which includes thermophilous taxa as Janulus and Megalotachea, is known at Pero Filho and Po´voa de Santare´m (basin of Tago) (Antunes & Mein, 1977; Truc, 1977). In central Europe at Sandelzhausen and Gu ¨ ndlkofen (Bayern, Germany) the deposits of the Bavarian ‘Obere Su¨sswassermolasse’ referred to MN 6 (Mein, 1990) yielded many land taxa among which the genera Cochlostoma, Pomatias, Gastrocopta (Sinalbinula), Parmacella, Palaeoglandina and some species of Clausilioidea and Helicoidea are signiWcant (Gall, 1972, 1980; Nordsieck, 1982). Parmacella (s.s.) is living in eastern N Africa (Cyrenaica, N Egypt), western N Africa, Iberian Peninsula (western regions), Canary Is., Caucasus, N Iran (Manganelli & Giusti, 1993). The stratigraphic position of Sandelzhausen has been recently revised by Heissig (1997) who refers the classic deposits of this site to the MN 5 zone (below the ‘MN 5 typique’); in this case the warm character of the mollusc fauna would be in accordance with the new stratigraphic position. At Hollabrunn (southern Austria) (MN 7/8) a rich mollusc assemblage comprises elements of tropical, subtropical and warm-temperate climate, as Palaina, Cochlostoma, Pomatias, Negulus, Leiostyla, Gastrocopta (Sinalbinula), Strobilops (Strobilops)

Contribution of non-marine molluscs

Table 16.1. Selected taxa of non-marine gastropods from Undorf (Bayern) (MN 5 lower part) and Sansan (France) (MN 6)

Selected taxa Carychium (Saraphia) nouleti Bourguignat Gastrocopta (Albinula) acuminata acuminata (Klein) Gastrocopta (Sinalbinula) larteti (Dupuy) Gastrocopta (Sinalbinula) nouletiana (Dupuy) Strobilops (Strobilops) costatus (Clessin) Vertigo (Vertigo) diversidens (Sandberger) Janulus supracostatus (Sandberger) Triptychia (Triptychia) bacillifera (Sandberger) Triptychia (Triptychia) grandis (Klein) Tryptychia (Milneedwardsia) lartei (Dupuy) Serrulella clessini (Boettger) Pseudidyla moersingensis undorfensis (Boettger) Palaeoglandina gracilis (Zieten) Opeas minutum (Klein) Leucochroopsis kleini (Klein) Leucochroopsis sp. Pseudochloritis incrassatus (Klein) Mesodontopsis ludovici (Noulet) Megalotachea turonensis larteti (Boissy) Protodrepanostoma sp.

Undorf

Sansan

MN 5

MN 6 ;

; ; ; ; ; ;

; ; ; ;

; ; ; ; ; ; ; ; ; ; ;

(Schu ¨ tt, 1967) and at Leobersdorf-B (zone B/C of Vienna basin) referable to MN 8 (cf. Ro¨gl et al., 1993) some thermophilous taxa like Papyrotheca (a Succineidae known also in the Pliocene of southeastern Europe), Negulus, Gastrocopta (Sinalbinula), Leiostyla, Strobilops and some species of Clausilioidea are recorded (Lueger, 1981). In Eastern Europe, at Opole (southwestern Poland) (MN 7) a very interesting similar fauna is recorded containing thermophilous taxa as Palaina, Craspedopoma, Pomatias, Negulus, Gastrocopta (Sinalbinula), Strobilops (Strobilops), Pleurodiscus, Pseudoleacina (Stworzewicz, 1989, 1993). Several species, mainly among Clausilioidea, occur for the Wrst time in deposits of central and eastern Europe referred to the MN 7 and MN 8 zones (e.g. Steinheim am Aalbuch, Germany; Opole, southwestern Poland; St.Veit a. d. Triesting and Hollabrunn, Austria; Anwil, N Switzerland). The new species of Clausilioidea belong to many genera, as Serrulella, Serrulastra, Pseudidyla, Regiclausilia, Cochlodina, Trolliella, Macrogastra, Clausilia among which Regiclausilia and Trolliella appear for the Wrst time (Nordsieck, 1981a,b, 1982).

335

Palaeoenvironments: non-mammalian evidence

336

Paleoclimatic remarks The warm character of the assemblages referred to the Wrst and middle part of the MN 5 zone is testiWed by the presence of moderately thermophilous and tropical–subtropical taxa such as Janulus, Gastrocopta (Albinula) (some species live in Mexico and central eastern United States), G. (Sinalbinula), Strobilops (Strobilops), Opeas (living in tropical and subtropical areas of new and old World). Many common early Miocene genera and species of Helicoidea of western and central Europe are no longer present (Schlickumia aquensis, Titthodomus, Pseudomonacha, Cyrtochilus, Creneatachea). Many systematics problems and the absence of a continuous record make it impossible to detect if the disappearance of these taxa is due to a unique extinction event or is gradual. In the upper part of the MN 5 zone Clausilioidea seem to be very scanty; other species, such as Leucochroopsis pisum and Megalotachea turonensis, which will be present again during the MN 6 zone, are lacking (Nordsieck, 1981b, 1986). A decrease in number of species in the upper part of the MN 5 zone, the lacking of some taxa and the deWnitive disappearance (probably gradual) of many ancient genera is most probably due to a climatic change towards a cooler phase. The re-establishment of warm conditions is testiWed by several events: the enrichment in species of the mollusc assemblages in deposits bearing mammal faunas referred to MN 6 zone; the presence of many warm taxa; the Wrst occurrence of several species, mainly among Clausilioidea, in deposits of central and eastern Europe referred to the MN 7 and MN 8 zones in concomitant with the appearance of tropical taxa such as Palaina, extinct in Europe but living in islands of the West PaciWc, Australia/Queensland, Amur basin, Japan, Korea.

Late Miocene Very scanty data are available for continental molluscs of the early Vallesian (MN 9) in western Europe. In central Europe some mollusc assemblages are recorded from the Vienna basin (Austria) in which deposits referred to MN 9, at Leobersdorf (‘Pannonian D’) and Vo¨sendorf (‘Pannonian D-E’) (Papp & Thenius, 1953; Lueger, 1981; Ro¨gl et al., 1993) yielded some thermophilous elements, Pomatias conicus (Klein), Negulus suturalis gracilis Gottschick & Wenz, Gastrocopta (Albinula) acuminata, G. (A.) edlaueri (Wenz), G. (Sinalbinula)

Contribution of non-marine molluscs

nouletiana (Dupuy), G. (S.) serotina Lozek, Leiostyla (L.) austriaca (Wenz), Strobilops (S.) tiarulus (Sandberger), Strobilops (S.) pappi (Schlickum), Triptychia leobersdorfensis (Troll), Leucochroopsis kleini, Galactochilus leobersdorfensis (Troll). At Go¨tzendorf (‘Pannonian F’), referred to the MN 9/10 boundary by Ro¨gl et al. (1993), an enrichment in number of species of the assemblages is evident. Numerous fossiliferous outcrops referred to late Vallesian and Turolian are present in France, mainly in the basin of the Rhoˆne. They are very rich in non-marine molluscs and mammals. The late Vallesian assemblages of Montredon, Soblay (De´p. Ain), Combesse, Tersanne (Bas Dauphine´) referred to MN 10 are characterized by several species of freshwater and land molluscs widespread in eastern France during the late Miocene (Truc, 1971a, 1972). The most signiWcant taxa are listed in Table 16.2. In Spain similar late Vallesian assemblages (MN 10) with Prososthenia, Melanopsis, Gastrocopta (A.), Strobilops and Helicoidea were found at Albacete (Calvo et al., 1978). During the early Turolian most Vallesian species are still present. In southeastern France the rich fossiliferous outcrops of Mollon-Ravin, Sermenaz-Bas-Neyron (Ain) and Lobrieu (Valre´as-Visan, Vaucluse) (Mein & Truc, 1966; Truc, 1971a, 1972) referred to MN 11 (early Turolian) (Mein, 1990) yielded very rich assemblages of aquatic and land species; the most signiWcant are listed in Table 16.2. The continental deposits of Ratavoux (Cucuron, Vaucluse) of middle Turolian age (MN 12) (Truc, 1971a,b; Ballesio et al., 1979) yielded several aquatic and land species widespread in the late Miocene among which Tudorella draparnaudi minor Depe´ret & Sayn, Melanopsis narzolina (Archiac), Gastrocopta sp., Palaeoglandina sp. A, Leucochroopsis dufrenoyi (Matheron), Megalotachea christoli (Matheron), Schlickumia cf. S. aquensis must be mentioned. M. christoli is a southern species variant of the northern M. delphinensis (Truc, 1971b). In Spain, the Neogene deposits of the Rio Cabriel Valley (Valencia and Albacete districts) are very rich in continental molluscs and vertebrates. They were known by the early authors, but their stratigraphic position has been revised in the past ten years taking into account the distribution of the micro- and macromammal faunas (Mein et al., 1978; Robles et al., 1991). The middle Turolian lacustrine and lignitiferous deposits of Fuente Podrida (Valencia), in addition to vertebrates referred to MN 12, yielded a rich mollusc fauna composed of mainly endemic prosobranchs and pulmonates among which the most paleoclimatically signiWcant species are: Melanopsis laevigata Lamarck, Planorbarius villatoyensis (Jodot), Iberus dupuydelomei (Revilla), Megalotachea sp. In the Turolian sediments of Teruel (MN 12) rich

337

Palaeoenvironments: non-mammalian evidence

338

Table 16.2. Selected taxa of non-marine gastropods from French deposits referred to MN 10 (De´p. Ain, Bas Dauphine´) and to MN 11 (De´p. Ain, Vaucluse) Selected taxa

MN 10

MN 11

Melanopsis kleini Kurr Gastrocopta (Albinula) sp. A Gastrocopta (Sinalbinula) sp. B Strobilops (Strobilops) sp. A Parmacella sayni (Fontannes) Triptychia (Milneedwardsia) lageti (Truc) Triptychia (Triptychia) bourguignati (Locard) Leucochroopsis valentinensis (Fontannes) Leucochroopsis v. sermenazensis (Locard) Apula escoffierae (Fontannes) Pseudochloritis mollonensis (Truc) Galactochilus locardi (Tournouer) Mesodontopsis heriacensis (Depe´ret) Megalotachea gualinoi (Michaud) Megalotachea delphinensis tersannensis (Locard)

; ; ; ; ; ;

; ; ; ; ; ;

; ; ;

; ;

; ; ; ; ; ; ;

assemblages are present (Royo Gomez, 1922; Sondaar, 1961; Robles, 1975). Many freshwater and land species characterize the numerous outcrops of this area (Los Mansuetos, Los Aljezares etc.): Tudorella draparnaudi, Gastrocopta (Albinula) sp. A, Strobilops (S.) sp. A, Zonitoides wenzi (Royo), Janulus olisipponensis (Roman), Palaeoglandina sp. A, ‘Helix’ bolivari (Royo), ‘Helix’ vilanovai (Royo), ‘Cepaea’ concudensis (Jodot). Only a few species are common to the coeval sediments of the Rhoˆne basin, the majority are in fact characteristic of Turolian of Spain. In France the continental uppermost Turolian sediments are not well represented owing to a big erosional phase whilst in Spain rich continental mollusc assemblages, referred to the late Turolian, are well outcropping. The rich vertebrate fauna of Venta del Moro (Valencia) is referred to the MN 13 zone. The abundant molluscs found in the same sediments bearing the vertebrates are represented mainly by freshwater prosobranchs and pulmonates; some land snails are also present. The most signiWcant taxa are: Melanopsis laevigata, M. narzolina gigantea (Robles), M. requenensis Royo, Planorbarius aV. P. villatoyensis and abundant specimens of the land gastropods Tudorella sp. and Megalotachea sp. (Aguirre et al., 1973; Mathisen & Morales, 1981; Robles et al., 1991). At La Portera, in the neighbourhood of Venta del Moro, a mollusc assemblage similar to that of Venta del Moro with Melanopsis narzolina gigantea, M. requenensis, Iberus dupuydelomei, Planorbarius sp., Megalotachea sp. and other

Contribution of non-marine molluscs

species, was found associated with many vertebrate remains dominated by bones of the hippopotamid Hexaprotodon crusafonti (Aguirre). The mammal fauna is referred to MN 13 (Lacomba et al., 1986; Robles et al., 1991). At Albacete vertebrate faunas referred to MN 13 and numerous species of aquatic and land molluscs were collected (Lopez-Sancho et al., 1984). The mollusc fauna of this locality is dominated by aquatic prosobranchs among which the warm climate genera Melanoides and Melanopsis are present. Turolian molluscan assemblages associated with vertebrates are poorly known in central Europe. In the Vienna basin at Eichkogel (‘Pannonian H’) near Mo¨dling, a rich fauna with Melanopsis, Negulus, Gastrocopta (Albinula), Gastrocopta (Sinalbinula), Strobilops (S.) and many species of Clausilioidea and Helicoidea, associated to vertebrates referred to MN11, are reported by Wenz & Edlauer (1942) and Lueger (1981). Many other sites of the same basin rich in similar mollusc faunas but without vertebrate remains are correlated to Eichkogel by Lueger (1981). Also in the Slovak part of the Pannonian basin, at East of Povazsky Inovec Mts., an abundant fauna of terrestrial and freshwater gastropods correlated to Eichkogel has been found (Fordina´l, 1996). The documentation about the land molluscs of late Messinian age is very scanty. In Messinian deposits of northeastern Italy (Montello, Cornuda in Veneto) a fauna with Helicoidea and Clausilioidea is recorded but only ‘Dinarica’ dalpiazi (Wenz) (endemic, probably belonging to the Galactochilus group, cf. Nordsieck, 1986), Mesodontopsis doderleini (Brusina) and Triptychia leobersdorfensis are recognized. The Galactochilus group ranges from the late Oligocene to middle Pliocene, M. doderleini is spread in Turolian sediments of Vienna basin (Lueger, 1981) and in ‘Pontian’ sediments of Croatia, Serbia, Hungary (Dal Piaz, 1942; Wenz, 1942; Schlickum & Strauch, 1973; Esu & Kotsakis, 1987). Some terrestrial species of warm and dry climate are recorded in the upper Messinian sediments of Brisighella (Romagna, central Apennine) (Parmacella, Pomatias sp., Rumina cf. R. decollata (Linnaeus)) (Esu & Taviani, 1989; Taviani, 1989; Manganelli & Giusti, 1993) and of Sicily (R. decollata, Helicella (Xerotricha) sp.) (Di Geronimo et al., 1991).

Paleoclimatic remarks Some species among Clausilioidea and other land pulmonates of tropical– subtropical climate became extinct in western and central Europe at the end of the Astaracian (MN 8). The most signiWcant taxa are Palaina, Pseudoleacina (P.) christoliana, Opeas, Serrulastra brandti (Schu¨tt), Coch-

339

Palaeoenvironments: non-mammalian evidence

340

lodina oppoliensis Nordsieck, Pseudidyla, Clausilia hollabrunnensis Schu ¨ tt. This fact and the scanty record of continental molluscs during the early Vallesian (MN 9) is probably due to a climatic change towards general cooling conditions at the beginning of the late Miocene. On the other hand, deposits with faunas referred to MN 10 and to MN 11 of central and western Europe are very rich in signiWcant molluscs. Many new species and genera of Clausilioidea and Helicoidea appear: Nordsieckia, T. (T.) bourguignati, T.(M.) lageti, L. valentinensis, P. mollonensis, A. escoYerae, M. heriacensis, M. delphinensis, M. gualinoi, G. locardi, Helicigona. Other taxa among aquatic prosobranchs and land snails of the families Melanopsidae, Vertiginidae, Strobilopsidae, Parmacellidae appear: M. kleini, Gastrocopta (A.) sp. A, Strobilops (S.) sp. A, P. sayni. Many typically thermophilous genera are widespread: Melanopsis, Gastrocopta (S.), Strobilops (S.), Parmacella, Janulus. The general high species diversity of the assemblages, the appearance of several taxa and the spreading of warm climate genera is indicative of a warming up during the late Vallesian and early Turolian. The non-marine assemblages of western Europe referred to the middle and late Turolian (MN 12 and MN 13) in addition to the spreading of aquatic prosobranchs, among which genera of warm climate as Melanopsis and Melanoides prevail, show a moderate expansion of land genera preferring a medium degree of aridity and warm conditions, like Tudorella, Parmacella, Janulus, Rumina, Iberus (all circum-Mediterranean genera except Janulus which is living in the Canary Is. and Madeira). The Mediterranean freshwater–brackish environments of Messinian age are characterized by the expansion of thermophilous oligo- and mesohaline elements typical of the ‘lago-mare’ biofacies, as gastropods of the families Neritidae, Melanopsidae, Thiaridae and bivalves of the subfamily Limnocardiinae of Paratethyan origin (Esu & Girotti, 1989).

Data analysis of Pliocene faunas

Early Pliocene Malacological assemblages of early Pliocene age are very well represented in France but almost unknown elsewhere in western and central Europe. A malacofaunistic renewal took place after the Messinian age in the Mediterranean area. The early Pliocene (Ruscinian) faunas of France are quite diVerent from the Miocene ones. Many new land snail genera and species occur at the

Contribution of non-marine molluscs

beginning of the Pliocene. Very rich assemblages of land and freshwater molluscs are recorded in the Rho ˆ ne basin. The main outcrops are located at Hauterives (Droˆme) in the northern part of the basin and at Celleneuve (He´rault) in the southern part (Truc, 1971a,b). In these two sites many remains of vertebrates were also found, both are referred to the lower part of the MN 14 zone (Mein, 1990; Lindsay et al., 1997, for Celleneuve). Many paleoclimatic and biostratigraphic signiWcant land genera and species occur in the two sites; at Celleneuve more southern genera and species are present (Table 16.3). Clausilioidea and Helicoidea are revised by Truc (1972) and by Nordsieck (1981a, 1986). These kind of assemblages are also present in other French outcrops referred to MN 14 (Perouges-Ferme Bardon, De´p. Ain; Vinsobres, De´p. Droˆme, Valre´as-Visan basin; Hautimagne, De´p. Vaucluse) (Truc, 1971a, 1972; Ballesio et al., 1979) which testiWes to a broad spreading of such species. In French deposits referred to MN 15 (Condal, Bourgogne; Neublans, Franche Comte´; Truc, 1971b; Nordsieck, 1986) many taxa of Helicoidea and Clausilioidea are recorded, some of which are new and some are present in deposits of the MN 14 zone: Soosia pseudoplanorbis Wenz, Apula amberti, Mesodontopsis chaixi, Puisseguria idanica (Locard), Helicigona chaignoni (Locard), Frechenia ducrosti (Locard), F. nayliesi. Most of them became extinct at the end of the early Pliocene. In Italy the Ruscinian deposits with continental molluscs are rather scanty and must be revised (Esu et al., 1986). Anyway land genera of paleoclimatical importance such as Janulus, Galactochilus, Palaeoglandina are recorded in Tuscany (De Stefani, 1876–80; Wenz, 1923–30). In Spain non-marine molluscs of supposed Ruscinian age were described by Royo Gomez (1922) but they need a systematic revision. Paleoclimatic remarks The richness of the Ruscinian assemblages, the occurrence of many new taxa of woody biotope, mainly among Helicoidea and Clausilioidea, and the presence of genera typical of a temperate–warm and subtropical climate, as Craspedopoma (living in the Canary Is., Madeira, Azores), Tudorella, Negulus, Gastrocopta (Albinula), Strobilops (Strobilops), Eostrobilops, Janulus and Schlickumia, which is extinct and was widespread in southwestern Europe during the Neogene (Truc, 1971b), point to an increase in humidity in respect to the drier previous climatic phase, suggesting subtropical conditions in the considered regions.

341

Palaeoenvironments: non-mammalian evidence

342

Table 16.3. Selected taxa of non-marine gastropods from the Rhoˆne basin deposits referred to MN 14. H = Hauterives (Bas Dauphine´), C = Celleneuve (Montpellier) Selected taxa

MN 14

Craspedopoma conoidale (Michaud) Tudorella baudoni (Michaud) Hydrocena dubrueilliana (Paladilhe) Carychium (Saraphia) pachychilum Sandberger Carychium (Carychiella) puisseguri Truc Negulus bleicheri (Paladilhe) Vertigo (Vertigo) nouleti Michaud Gastrocopta (Albinula) dupuy (Michaud) Gastrocopta (Sinalbinula) baudoni (Michaud) Leiostyla priscilla (Paladilhe) Eostrobilops duvali (Michaud) Strobilops (Strobilops) labyrinthiculus (Michaud) Strobilops (Strobilops) romani (Wenz) Macrozonites casteti (Viguier) Macrozonites collongeoni (Michaud) Janulus sp. A Fortuna seringi (Michaud) Serrulella michelottii (Michaud) Serrulastra michaudi Nordsieck Nordsieckia fischeri (Michaud) Tryptichia (Milneedwardsia) sinistrorsa (De Serres) Tryptichia (Milneedwardsia) terveri (Michaud) Cochlodina? berthaudi (Michaud) Laminifera (Laminiplica) meini Truc Truciella ballesioi (Truc) Macrogastra loryi (Michaud) Clausilia baudoni (Michaud) Palaeoglandina paladilhei (Michaud) Protodrepanostoma bernardii (Michaud) Soosia godarti (Michaud) Apula amberti (Michaud) Frechenia nayliesi (Michaud) Frechenia quadrifasciata (De Serres) Helicigona truci Schlickum Schlickumia gaspardiana (Paladilhe) Mesodontopsis chaixi (Michaud)

HC H C HC C HC HC HC H C HC HC H C H C H HC C HC C H H C C HC HC H H HC HC H C H C H

Middle Pliocene The early Villafranchian deposits of the Piedmont basin, Villafranca d’Asti, Fossano and Tassarolo, bearing mammal fauna of the Triversa F.U. (Azzaroli, 1977) (base of MN 16; Mein, 1990), yielded rich mollusc faunas studied by F. Sacco during the nineteenth century and revised by many

Contribution of non-marine molluscs

authors (see Esu et al., 1993; Ciangherotti et al., 1996). They are dominated by archaic thermophilous land pulmonates most of which were widespread in other basins; many taxa, also among the aquatic forms, indicate strong aYnities with early or middle Pliocene western European basins (Hauterives, Celleneuve, France; Frechen-Fortuna, West Germany). Many species of thermophilous character are common to other middle and late Pliocene Italian deposits (Sardinia and central Italy, Dunarobba; Ciangherotti et al., 1998) and other taxa are endemic (Table 16.4). The rich fossil-bearing German deposits of Frechen and Fortuna (Rheinische Braunkohle) (Schlickum & Strauch, 1979) are referred to the Wrst part of the MN 16 zone (Nordsieck, 1982) but they are slightly younger than the Italian localities of Villafranca d’Asti and Fossano (see discussion in Ciangherotti et al., 1996). The mollusc assemblages are characterized by the appearance of many new taxa among freshwater and land gastropods and by the presence of several thermophilous taxa. The assemblage of Cessey-sur-Tille (La Bresse basin, France) is assigned to the top of the MN 16 zone (cf. Chaline, 1984). The mollusc assemblage is characterized by archaic thermophilous elements as Cochlostoma, Hydrocena, Negulus, Fortuna, Frechenia, Puisseguria and a certain number of modern land species (Puissegur, 1984). Many new species of Clausilioidea appear in the middle Pliocene sites mentioned above, and also in Sessenheim (Elsass, France) (MN 16) (Nordsieck, 1972, 1974, 1976, 1978, 1982; Schlickum & Geissert, 1980). In the same localities many signiWcant species of Helicoidea are also recorded (Table 16.4).

Paleoclimatic remarks The mollusc assemblages of the middle Pliocene deposits of western Europe reveal warm climatic conditions not much diVerent from those of the early Pliocene. In fact thermophilous land and freshwater elements occurring in deposits referred to MN 14 and MN 15 are widespread also in those referred to MN 16; moreover the faunistic aYnities among French, German and Italian basins point to similar climatic conditions in these regions. On the whole, the middle Pliocene assemblages point to a warm and humid climatic phase since the thermophilous and forestal elements are dominant: Negulus, Leiostyla (Leiostyla), Gastrocopta (Albinula), Eostrobilops, Janulus, Triptychia, Serrulella, Laminifera (Laminiplica), Polloneria, Protodrepanostoma, Mesodontopsis, Schlickumia.

343

Hydrocena (Hydrocena) dubrueilliana (Paladilhe) Carychium (Saraphia) pachychilum Sandberger Carychium (Saraphia) pseudotetrodon Strauch Negulus truci Schlickum Negulus villafranchianus (Sacco) Vertigo (Vertigo) nouleti Michaud Gastrocopta (Albinula) acuminata fossanensis (Sacco) Gastrocopta (Vertigopsis) dehmi Schlickum & Strauch Leiostyla capellinii (Sacco) Leiostyla gottschicki (Wenz) Eostrobilops aloisii Manganelli, Delle Cave & Giusti Eostrobilops patuliformis (Sacco) Janulus angustiumbilicatus (Sacco) Discus lateumbilicatus (Sacco) Discus pantanellii (Sacco) Monoptychia monoptyx Nordsieck Triptychia geisserti Nordsieck Triptychia mastodontophila (Sismonda) Triptychia schlickumi Nordsieck Serrulella? decemplicata (Sacco) Serrulella truci (Nordsieck) Serruluna anodon Nordsieck Macrogastra densestriata (Rossma¨ssler) Macrogastra loryi (Michaud) Macrogastra multistriata Nordsieck Macrogastra schlickumi (Nordsieck)

Selected taxa

;

;

;

;

;

;

;

;

; ;

;

;

; ;

;

;

; ;

;

;

; ; ; ; ; ; ; ; ;

;

; ; ;

Dunarobba MN 17

;

; ;

Cessey-sur-Tilles MN 16

;

;

Sessenheim MN 16

;

Frechen/Fortuna MN 16

; ;

; cf. ;

Piedmont MN 16

Table 16.4. Selected taxa of non-marine gastropods from Italian (MN 16, MN 17), German (MN 16) and French deposits (MN 16)

Macrogastra sessenheimensis (Nordsieck) Laminifera (Laminiplica) cesseyensis Nordsieck Laminifera (Laminiplica) villafranchiana (Sacco) Polloneria pliocenica (Sacco) Cochlodina laminata (Montagu) Cochlodina? prolaminata (Sacco) Clausilia baudoni baudoni Michaud Clausilia baudoni tillensis Nordsieck Clausilia pliodiptyx Nordsieck Clausilia? portisi (Sacco) Clausilia produbia Nordsieck Clausilia rolfbrandti Schlickum Clausilia strauchiana Nordsieck Protodrepanostoma bernardii (Michaud) Protodrepanostoma plioauriculatum (Sacco) Soosia monikae Schlickum & Strauch Apula koehnei (Schlickum & Strauch) Mesodontopsis chaixi (Michaud) Mesodontopsis nehringi Schlickum & Strauch Puisseguria kowalczyki Schlickum & Strauch Puisseguria zilchi Schlickum Helicigona chaignoni (Locard) Helicigona schwarzbachi Schlickum & Strauch Frechenia ducrosti (Locard) Frechenia nayliesi (Michaud) Frechenia reichenbachi Schlickum & Strauch Schlickumia bottinii (Sacco)

Selected taxa

Table 16.4 (cont.).

;

;

;

;

; ;

Piedmont MN 16

; ; ;

;

; ;

; ;

;

Frechen/Fortuna MN 16

;

; ; ;

;

;

;

;

;

;

Sessenheim MN 16

;

; ;

;

;

;

; ;

;

Cessey-sur-Tilles MN 16

cf.

Dunarobba MN 17

Palaeoenvironments: non-mammalian evidence

346

Late Pliocene In southeastern Spain (Murcia and Alicante districts) some thermophilous land snails, Parmacella sp., Rumina sp., Palaeoglandina montenati (Truc), Schlickumia alicantensis (Truc), Tacheocampylaea sp. and the aquatic prosobranch Melanopsis sp., are recorded by Montenat & Truc (1971) in assemblages with vertebrate remains probably belonging to MN 17 zone. In the basin of Guadiz-Baza (Granada, Spain) Plio-Pleistocene sediments with freshwater and land molluscs occur (Alberdi et al., 1989; Robles, 1989). The assemblages referred to the latest Pliocene are characterized by a high number of modern European elements of temperate climate. Only one extinct archaic genus, the aquatic prosobranch Prososthenia, known from the late Miocene to early Pleistocene of Europe, is present. Two aquatic taxa typical of a warm climate, Melanopsis and Melanoides, also occur. In the Plio-Pleistocene deposits of the Bresse basin (Bourgogne, France) some mollusc assemblages recorded at Montagny-les-Beaune and Blignyles-Beaune are associated with micromammals referred to the Mimomys pliocaenicus zone correlated with the Tiglian by Chaline (1984). The fauna is composed of a few archaic freshwater and land genera, as Tournouerina, Neumayria, Gastrocopta (Sinalbinula) and by many modern genera and species (Schlickum & Puisse´gur, 1978; Puisse´gur, 1984). In Italy very few outcrops referred to this period are known. The most important for the richness of molluscs is the Dunarobba site (Terni, Umbria), in central Italy, where a late Pliocene fossil forest is very well preserved (Ambrosetti et al., 1995). The sediments bearing the fossil forest are referred to a coastal lacustrine wetland in which two main diVerent malacological assemblages are recorded: the forestal biotope is dominated by a rich land snail assemblage of a swampy environment characterized by paleoclimatic and biochronologic signiWcant Pliocene elements many of which are also present in middle Pliocene deposits of Villafranca d’Asti (Piedmont) and Frechen-Fortuna (Table 16.4); the lacustrine one is dominated by aquatic prosobranchs represented mainly by thermophilous taxa widespread in central Italy until early Pleistocene: Theodoxus (Neritaea), Prososthenia, Melanopsis. A certain number of modern taxa are also present.

Paleoclimatic remarks The mollusc fauna of the Dunarobba site (central Italy) shows clear warm climatic conditions. The high percentage of thermophilous elements present among both aquatic and land species like Hydrocena (Hydrocena) (living in Azores, S Africa, southeastern Asia, PaciWc Is. and in Dalmatia as relict),

Contribution of non-marine molluscs

Negulus, Gastrocopta (Albinula), Leiostyla, Eostrobilops, Theodoxus (Neritaea) (the subgenus Neritaea is living in N Africa), Melanopsis is signiWcant. The assemblages referred to the latest Pliocene of Spain and France show temperate–warm conditions being characterized by some archaic thermophilous genera, such as Parmacella, Palaeoglandina, Schlickumia, Prososthenia and by a high number of European elements of temperate climate still living. The climatic deterioration of the late Pliocene in respect to the middle Pliocene may be evidenced by some faunistic events: an impoverishment of archaic species in the late Pliocene molluscan assemblages (i.e. Dunarobba molluscs in respect to the Villafranca d’Asti and Frechen-Fortuna assemblages), the extinction at the end of the MN 16 zone of many species of Clausilioidea and Helicoidea, and a higher percentage of modern species in the upper Pliocene deposits.

Concluding remarks The almost complete Neogene record of mollusc associations in western and central Europe and the possibility to order them in a chronostratigraphic framework permits us to outline their variation through the time and to point out environmental and paleoclimatic changes. In a general way it is possible to make the following remarks. 1. The beginning of the Miocene is characterized by the persistence of some late Oligocene thermophilous land taxa and by the appearance of some new genera. The extinctions are limited mainly to species level. 2. From the early Burdigalian (MN 3) till about the upper Langhian (the middle part of the MN 5 zone), the mollusc assemblages are characterized by the dominance of a mixture of archaic genera that appeared during the Oligocene and of new Miocene taxa. Very rich assemblages are present during this time (MN 3, MN 4, MN 5 basal part) in which many genera of tropical, subtropical and warm–temperate climate are widespread. 3. A conspicuous number of these elements is absent from the assemblages referred to the upper part of MN 5 zone and the more recent ones. Many systematic problems and the absence of continuous record make it impossible to detect if the disappearance of these forms is due to a unique extinction event or is gradual. But in any case the faunistic change at generic level is the most important of the whole

347

Palaeoenvironments: non-mammalian evidence

348

4.

5.

6.

7.

8.

9.

Neogene. The absence of a great number of elements characterizing the early Miocene and the early middle Miocene in deposits assigned to the upper part of the MN 5 zone (and more recent ones) indicates an important change in the climatic conditions, very probably to correlate with the cooling of waters in the high latitudes at about 15–14 Ma (Kennett, 1995). The assemblages referred to the late middle Miocene (MN 6, MN 7/8) are aVected by a strong enrichment in number of species of warm climate and by the Wrst occurrence of several species, which testiWes to a change in the climate towards the re-establishment of warm conditions. The transition from the middle to the late Miocene, at the beginning of the MN 9 zone, around 11 Ma, is characterized by the extincion of a few genera and some species and by the appearance of a small number of new taxa. This fact, and the scanty record of continental molluscs during the early Vallesian, are probably due to a slight climatic change towards cooling conditions, but the changes in the composition of the molluscan assemblages are far less important than the previous one. During the late Vallesian and early Turolian (MN 10, MN 11 zones) the mollusc assemblages are characterized by a general degree of high diversity in species. The appearance of several taxa and the spreading of warm climate genera is indicative of a signiWcant warming up during this time span. The assemblages referred to middle and late Turolian (MN 12 and MN 13 zones) show a moderate expansion of land genera preferring a medium degree of aridity, pointing to an expansion of semi-arid areas around the Mediterranean basin during the latest Miocene. The extinction of a good number of species and genera at the end of the Miocene can be correlated to the cooling at about 6.5–5 Ma (Kennett, 1995). A malacofaunistic renewal took place after the Messinian age. A good number of new species and genera occur at the beginning of Pliocene (MN 14). The presence of taxa typical of temperate–warm and subtropical climate point to an increase in temperature and humidity with respect to the previous climatic phase. The molluscan assemblages of the middle Pliocene (MN 16) reveal similar climatic conditions. In the late Pliocene (MN 17) some events may be evidenced in the molluscan assemblages: a general impoverishment of species with a decline of the thermophilous taxa, the extinction at the end of the MN 16 zone of many species and a major percentage of modern species and genera. These events reXect the climatic deterioration related to the

Contribution of non-marine molluscs

onset of the Wrst large ice sheet on the Northern Hemisphere, at the Gauss/Matuyama boundary (at about 2.6 Ma) (Shackleton et al., 1990).

References Aguirre, E., Robles, F., Thaler, L., Lopez, N., Alberdi, M. T. & Fuentes, C. 1973. Venta del Moro, nueva fauna Wnimiocenica de moluscos y vertebrados. Estudios Geologicos, 29, 569–78. Aguirre, J., Castillo, C., Ferriz, F. J., Agustı´, J. & Oms, O. 1995. Marine-continental magnetobiostratigraphic correlation of the Dolomys subzone (middle of Late Ruscinian): implications for the Late Ruscinian age. Palaeogeography, Palaeoclimatology, Palaeoecology, 117, 139–52. Alberdi, M. T., Bonadonna, F. P. & Leone, G. 1989. Esquema de la evolucio´n geodina´mica de la regio´n de Guadix-Baza (Granada, Espan ˜ a). Trabaios sobre el Neogeno-Cuaternario, 11, 113–8. Alvarez Sierra, M. A., Daams, R., Meulen Van Der, A. & Pela´ez-Campomanes, P. 1997. A new time scale for the Aragonian. Abstracts BiochroM’97, 18. Montpellier. Ambrosetti, P., Basilici, G., Ciangherotti, A. D., Codipietro, G., Corona, E., Esu, D., Girotti, O., Lo Monaco, A., Meneghini, M., Paganelli, A. & Romagnoli, M. 1995. La Foresta Fossile di Dunarobba (Terni, Umbria, Italia centrale): contesto litostratigraWco, sedimentologico, palinologico, dendrocronologico e paleomalacologico. Il Quaternario, 8, 465–508. Antunes, M. T. & Mein, P. 1977. Contributions a` la pale´ontologie du Mioce`ne moyen continental du bassin du Tage. III. Mammife`res. Ciencias de Terra (UNL), 3, 143–65. Azzaroli, A. 1977. The Villafranchian stage in Italy and the Plio-Pleistocene boundary. Giornale di Geologia, s.2, 41, 61–79. Ballesio, R., Carbonnel, G., Mein, P. & Truc, G. 1979. Sur un nouveau gisement fossilife`re du Mioce`ne supe´rieur (Tortonien-Turolien moyen) de Cucuron (Vaucluse). Geobios, 12, 467–71. Bruggen, A. C. van 1994. The genus Truncatellina Lowe, 1952 (Gastropoda Pulmonata: Pupilloidea) in Malawi, Zambia, Angola and Fernando Poo. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, 97, 1–25. Bruijn, H. De, Daams, R., Daxner-Ho¨ck, G., Fahlbusch, V., Ginsburg, L., Mein, P. & Morales, J. 1992. Report of the RCMNS working group on fossil mammals, Reisensburg 1990. Newsletters on Stratigraphy, 26, 65–118. Calvo J. P., Elizaga E., Lopez Martinez N., Robles F. & Usera J. 1978. El Mioceno superior continental del Prebe´tico Externo: Evolucio´n del Estrecho Nordbe´tico. Boletin Geologico y Mineiro, 89, 407–26. Calzada, S. 1990. Gastero´podos no marinos burdigalienses de Sant Pere de Ribes (Barcelona). Sucesio´n faunı´stica. Paleontologia i Evolucio`, 23, 181–6. Cavelier, C. 1972. Les e´quivalents continentaux de l’Aquitanien marin provenc¸al et le classement de la zone de Paulhiac a` la base du Mioce`ne. Bulletin du Bureau des Recherches Ge´ologiques et Minie`res, s.2, sec.I, 4, 115–23. Chaline, J. 1984. La se´quence des rongeurs de Bresse, en tant que re´fe´rence biostratigraphique et pale´oclimatique. Ge´ologie de la France, 3, 251–68.

349

Palaeoenvironments: non-mammalian evidence

350

Ciangherotti, A. D., Esu, D. & Girotti, O. 1996. La malacofauna continentale. In: Revisione del Villafranchiano nell’area-tipo di Villafranca d’Asti. Il Quaternario, 9, 93–4. Ciangherotti, A. D., Esu, D. & Girotti, O. (1998). Review of the history of the Late Neogene and Early Quaternary non-marine molluscs of Italy. Mededelingen Nederlands Instituut voor Toegepaste Geowetenschappen, 60. Dal Piaz, G. 1942. L’eta` del Montello. Commentarii PontiWcia Academia Scientiarum, 6, 475–94. De Stefani, C. 1876–80. Molluschi continentali, Wno ad ora notati in Italia nei terreni pliocenici, ed ordinamento di questi ultimi. Atti della Societa` Toscana di Scienze Naturali, Memorie, 2, 130–74; 3, 274–325; 5, 9–108. Di Geronimo, I., Esu, D. & Grasso, M. 1991. Gli strati a ‘Congerie’ del Messiniano superiore del margine nord-occidentale ibleo. Caratteristiche faunistiche e possibili implicazioni paleogeograWche e paleoclimatiche. Atti dell’ Accademia Peloritana dei Pericolanti, Classe I di Scienze Fisiche Matematiche e Naturali, 67 (1989), 129–50. Dollfus, G. F. 1915. Etude sur la molasse de l’Armagnac. Bulletin de la Socie´te´ Ge´ologique de France, s.4, 15, 335–402. Esu, D. & Girotti, O. 1989. Late Miocene and Early Pliocene continental and oligohaline molluscan faunas of Italy. Bollettino della Societa` Paleontologica Italiana, 28, 253–63. Esu, D., Girotti, O. & Kotsakis, T. 1986. Lineamenti di paleobiogeograWa dei vertebrati e dei molluschi continentali dell’Italia centrale. II. Cenozoico. Memorie della Societa` Geologica Italiana, 35, 245–55. Esu, D., Girotti, O. & Kotsakis, T. 1989. Oligotipia nei vertebrati e nei molluschi continentali fossili. Atti 3° Simposio di Ecologia e Paleoecologia delle Comunita` Bentoniche, 285–98. Esu, D., Girotti, O. & Kotsakis, T. 1993. Palaeobiogeographical observations on Villafranchian continental molluscs of Italy. Scripta Geologica, Special Issue, 2, 101–19. Esu, D. & Kotsakis, T. 1987. PaleobiogeograWa dei vertebrati e dei molluschi continentali dell’area alpina sud-orientale. Biogeographia, 13, 57–81. Esu, D. & Taviani, M. 1989. Oligohaline mollusc faunas of the Colombacci Formation (upper Messinian) from an exceptional fossil vertebrate site in the Romagna Apennines: Monticino Quarry (Brisighella, N Italy). Bollettino della Societa` Paleontologica Italiana, 28, 265–70. Fahlbusch, V. 1964. Die Cricetiden (Mamm.) der Oberen Su ¨ sswasser-Molasse Bayerns. Abhandlungen der Bayerischen Akademie der Wissenschaften, N.F., 118, 1–136. Fahlbusch, V. 1975. Die Eomyiden (Rodentia, Mammalia) der Oberen Su¨sswasser-Molasse Bayerns. Mitteilungen der Bayerischen Staatssammlung fu ¨ r Pala¨ontologie und Historische Geologie, 15, 63–90. Fahlbusch, V. 1976. Report on the International Symposium on Mammalian Stratigraphy of the European Tertiary. Newsletters on Stratigraphy, 5, 160–7. Falkner, G. 1986. Protodrepanostoma nordsiecki n.sp. aus dem untermioza¨nen Landschneckenkalk von Tuchorice/Bo¨hmen (Gastropoda: Hygromiidae). Heldia, 1, 121–3. Fischer, J. C. (in press). La faune mioce` ne de Sansan (Gers) et son environnement. 8. Malacofaune. Me´moires du Museum d’ Histoire Naturelle de Paris.

Contribution of non-marine molluscs

Fordina´l, K. 1996. Terrestrial gastropods of the Upper Pannonian in the northern part of the Danube basin. Slovak Geological Magazine, 1, 5–16. Gall, H. 1972. Die obermioza¨ne Fossil-Lagersta¨tte Sandelzhausen. 4. Die Molluskenfauna (Lamellibranchiata, Gastropoda) und ihre stratigraphische und o¨kologische Bedeutung. Mitteilungen der Bayerischen Staatssammlung fu ¨r Pala¨ontologie und Historische Geologie, 12, 3–32. Gall, H. 1980. Eine Gastropodenfauna aus dem Landshuter Schotter der Oberen Su¨sswassermolasse (Westliche Paratethys, Badenien) von Gu ¨ ndlkofen/Niederbayern. Mitteilungen der Bayerischen Staatssammlung fu ¨r Pala¨ontologie und Historische Geologie, 20, 51–77. Gigot, P., Mein, P. & Truc, G. 1976. La se´rie continentale de Pont d’Aiguines pre`s de Moustiers-Ste-Marie: un equivalent lateral du Mioce`ne marin du Bassin de Digne (Alpes de Haute Provence). Geobios, 9, 795–9. Gradstein, F. M. & Ogg, J. 1996. A Phanerozoic time scale. Episodes, 19, 3–5. Heissig, K. 1997. Mammal faunas intermediate between the reference faunas of MN 4 and MN 6 from the Upper Freshwater Molasse of Bavaria. Me´moires et Travaux de l’Institut de Montpellier, 21, 537–46. Kennet, J. P. 1995. A review of Polar climatic evolution during the Neogene, based on the marine sediment record. In Paleoclimate and evolution, with emphasis on Human origins, Vrba, E. S. et al. (eds.), pp. 49–64. New Haven, Yale University Press. Krijgsman, W., Garces, M., Langereis, C. G., Daams, R., Dam, J. Van, Meulen, A. J. Van Der, Agustı´, J. & Cabrera, L. 1996. A new chronology for the middle to late Miocene continental record in Spain. Earth and Planetary Sciences Letters, 142, 367–80. Lacomba, J. I., Morales, J., Robles, F., Santisteban, C. & Alberdi, M. T. 1986. Sedimentologia y paleontologia del yacimiento Wnimiocenico de La Portera (Valencia). Estudios Geologicos, 42, 167–80. Lindsay, E. H., Michaux, J., Suc, J. P., Aguilar, J. P., Clauzon, G. & Opdyke, N. D. 1997. Small mammals, magnetic polarity sequences and pollen Xoras in the Languedoc-Roussillon areas. Abstracts BiochroM’97, 79. Montpellier. Lopez-Sancho, J. L., Acuna, J. D. & Robles, F. 1984. Tafonomia y paleoecologia de una secuencia turoliense de Tolosa (Albacete). Estudios Geologicos, 40, 201–13. Lozek, V. 1964. Quarta¨rmollusken der Tschechoslowakei. Rozpravy U ´ strednı´ho U ´ stavu Geologicke´ho, 31, 1–368. Lueger, J. P. 1981. Die Landschnecken im Pannon und Pont des Wiener Beckens. Denkschriften der O ¨ sterreichische Akademie der Wissenschaften, Mathematisch-Naturwissenschaftlichen Klasse, 120, 1–124. Manganelli, G., Delle Cave, L. & Giusti, F. 1989. Notulae malacologicae, XLII. Strobilopsidae (Gastropoda, Pulmonata), a family new to the Villafranchian land snail fauna of Apenninic Italy. Basteria, 53, 3–13. Manganelli, G. & Giusti, F. 1993. Notulae malacologicae, XLV. Fossil Parmacellidae from Italy. Archiv fu ¨ r Molluskenkunde, 121, 143–56. Manganelli, G., Giusti, F. & Delle Cave, L. 1990. Notulae malacologicae, XLVII. Lauriinae (Gastropoda, Pulmonata, Orculidae/Pupillidae) from the Villafranchian of peninsular Italy. Basteria, 54, 87–103. Mathisen, M. & Morales, J. 1981. Stratigraphy, facies and depositional environments of the Venta del Moro vertebrate locality, Valencia, Spain. Estudios Geologicos, 37, 199–207.

351

Palaeoenvironments: non-mammalian evidence

352

Mein, P. 1975. Re´sultats du Groupe de Travail des Verte´bre´s. In Report on Activity of the Regional Committee on Mediterranean Neogene Stratigraphy Working Groups (1971–1975), Senes, J. (ed.), pp. 78–81. Bratislava. Mein, P. 1990. Updating of MN zones. In European Neogene Mammal Chronology, Lindsay, E. et al. (eds.), pp. 73–90. New York, Plenum Press. Mein, P., Moissenet, E. & Truc, G. 1978. Les formations continentales du Ne´oge`ne supe´rieur des Valle´es du Ju´car et du Cabriel au NE d’Albacete (Espagne). Biostratigraphie et environnement. Documents des Laboratoires de Ge´ologie de Lyon, 72, 99–148. Mein, P. & Truc, G. 1966. Facie`s et association faunique dans le Mioce`ne supe´rieur continental du Haut-Comtat Venaissin. Travaux du Laboratoire de Ge´ologie de Lyon, n.s., 13, 273–6. Mein, P., Truc, G. & Demarcq, G. 1971. Micromammife`res et gaste´ropodes continentaux des biozones de Paulhiac et de la Romieu dans le Mioce`ne de la Bastidonne et de Mirabeau (Vaucluse, Sud Est de la France). Comptes Rendus de l’ Academie des Sciences de Paris, s.D, 273, 566–8. Moayedpour, E. 1977. Geologie und Pala¨ ontologie des tertia¨ren ‘Braunkohlenlagers’ von Theobaldshof/Rho¨n (Mioza¨n, Hessen). Geologische Abhandlungen Hessen, 76, 1–135. Montenat, C. & Truc, G. 1971. Plioce`ne supe´rieur et Villafranchien dans le Levant espagnol (Provinces de Murcia et d’Alicante). Boletin Geologico y Mineiro, 82, 52–8. Nordsieck, H. 1972. Fossilen Clausilien, I. Clausilien aus dem Plioza¨n W-Europas. Archiv fu ¨ r Molluskenkunde, 102, 165–88. Nordsieck, H. 1974. Fossile Clausilien, II. Clausilien aus dem O-Plioza¨n des Elsaß. Archiv fu ¨ r Molluskenkunde, 104, 29–39. Nordsieck, H. 1976. Fossile Clausilien, III. Clausilien aus dem O-Plioza¨n des Elsaß, II. Archiv fu ¨ r Molluskenkunde, 107, 73–82. Nordsieck, H. 1978. Fossile Clausilien, IV. Neue Taxa neogener europa¨ischer Clausilien, I. Archiv fu ¨ r Molluskenkunde, 109, 103–8. Nordsieck, H. 1981a. Fossile Clausilien, V. Neue Taxa neogener europa¨ischer Clausilien, II. Archiv fu ¨ r Molluskenkunde, 111, 63–95. Nordsieck, H. 1981b. Fossile Clausilien, VI. Die posteoza¨nen tertia¨ren Clausilien Mittelund West-Europas. Archiv fu ¨ r Molluskenkunde, 111, 97–114. Nordsieck, H. 1982. Zur Stratigraphie der neogenen Fundstellen der Clausiliidae und Triptychiidae Mittel- und Westeuropas (Stylommatophora, Gastropoda). Mitteilungen der Bayerische Staatssammlung fu¨r Pala¨ontologie und Historische Geologie, 22, 137–55. Nordsieck, H. 1986. Das System der tertia¨ren Helicoidea Mittel- und Westeuropas (Gastropoda: Stylommatophora). Heldia, 1, 109–20. Papp, A. & Thenius, E. 1953. Vo¨sendorf – ein Lebensbild aus dem Pannon des Wiener Beckens. Mitteilungen der Geologischen Gesellschaft in Wien, 46 (Sonderband), 1–109. PfeVer, G. 1929. Zur Kenntnis tertia¨rer Landschnecken. Geologische und Pala¨ontologische Abhandlungen, N.F., 17, 153–380. Puisse´gur, J. J. 1976. Mollusques continentaux quaternaires de Bourgogne. Me´moires Ge´ologiques de l’ Universite´ de Dijon, 3, 1–241. Puisse´gur, J. J. 1984. Les faunes malacologiques plio-ple´istoce`nes de la Bresse.

Contribution of non-marine molluscs

SigniWcations e´cologique, climatique et chronologique. Ge´ologie de la France, 3, 281–302. Rangheard, Y., Demarcq, G., Mongereau, N., Pharisat, A., Pouyet, S. & Truc, G. 1985. Le Mioce`ne du Val des Verrie`res et du Bief des Lavaux (Jura Central, Haute-Chaıˆne): e´ve´nements pale´obioge´ographiques et ge´odynamiques. Geobios, 18, 769–85. Reuss, A. 1849. Beschreibung der fossilen Ostracoden und Mollusken der tertia¨ren Su¨sswasserschichten des no¨rdlichen Bo¨hmens. Palaeontographica, 2, 16–42. Rey, R. 1968. L’Oligoce`ne a` partir des mollusques continentaux. Annales Gue´bhard, 44, 162–96. Robles, F. 1975. Contribucion al conocimiento paleomalacologico del Turoliense de Teruel. Trabajos sobre el Neogeno-Cuaternario, 4, 113–15. Robles, F. 1989. Moluscos continentales del Plio-Pleistoceno de la cuenca de Guadix-Baza. Trabajos sobre el Neogeno-Cuaternario, 11, 127–38. Robles, F., Belincho´n, M., Garcia-Flor, J. & Morales, J. 1991. El Neo´geno continental de Bun ˜ ol y del Valle del Rio Cabriel. Revista Espan ˜ ola de Paleontologı´a, n. extr., 205–15. Ro¨gl, F., Zapfe, H., Bernor, R. L., Brzobohaty, L., Daxner-Ho¨ck, G., Draxler, I., Fejfar, O., Gaudant, J., Herrmann, P., Rabeder, G., Schultz, O. & Zetter, R. 1993. Die Primatenfundstelle Go¨tzendorf an der Leitha (Obermioza¨n des Wiener Beckens, Niedero¨sterreich). Jahrbuch der Geologischen Bundesanstalt -A., 136, 503–26. Royo Gomez, J. 1922. El Mioceno continental ibe´rico y su fauna malacolo´gica. Memorias Comisio´n de Investigaciones Paleontologicas y Prehisto´ricas, 30, 1–230. Schlickum, W. R. 1976. Die in der pleistoza¨nen Gemeindekiesgrube von Zwiefaltendorf a.d. Donau abgelagerte Molluskenfauna der Silvanaschichten. Archiv fu ¨r Molluskenkunde, 107, 1–31. Schlickum, W. R. & Geissert, F. 1980. Die plioza¨ne Land- und Su¨sswassermolluskenfauna von Sessenheim/Krs. Hagenau (Unterelsass). Archiv fu ¨ r Molluskenkunde, 110 (1979), 225–59. Schlickum, W. R. & Puisse´gur, J. J. 1978. Die Molluskenfauna der Schichten mit Viviparus burgundinus und Pyrgula nodotiana von Montagny-les-Beaune (De´p. Coˆte-d’Or). Archiv fu¨r Molluskenkunde, 109, 1–26. Schlickum, W. R. & Strauch, F. 1973. Die neogene Gastropoden Gattung Mesodontopsis Pilsbry 1895. Archiv fu¨r Molluskenkunde, 103, 153–64. Schlickum, W. R. & Strauch, F. 1979. Die Land- und Su ¨ sswassermollusken der plioza¨nen Deckschichten der rheinischen Braunkohle. Abhandlungen Senckenbergischen Naturforschenden Gesellschaft, 536, 1–144. Schlunegger, F., Burbank, D. W., Matter, A., Engesser, B. & Mo¨dden, C. 1996. Magnetostratigraphic calibration of the Oligocene to Middle Miocene (30–15 Ma) mammal biozones and depositional sequences of the Swiss Molasse Basin. Eclogae Geologicae Helvetiae, 89, 753–88. Schlunegger, F., Matter, A., Burbank, D. W. & Klaper, E. M. 1997. Magnetostratigraphic constraints on relationships between evolution of the central Swiss Molasse basin and Alpine orogenic events. Geological Society of America Bulletin, 109, 225–41. Schu ¨ tt, H. 1967. Die Landschnecken der untersarmatischen Rissoenschichten von Hollabrunn, N.-O ¨ . Archiv fu¨r Molluskenkunde, 96, 199–222.

353

Palaeoenvironments: non-mammalian evidence

354

Shackleton, N. J., Berger, A. & Peltier, W. R. 1990. An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Transactions of the Royal Society of Edinburgh, 81, 251–61. Sondaar, P. 1961. Les Hipparion de l’Arago´n meridional. Estudios Geologicos, 17, 209–305. Stworzewicz, E. 1989. Slimaki – Gastropoda. In: History and evolution of the terrestrial fauna of Poland. Folia Quaternaria, 59–60, 43–67. Stworzewicz, E. 1993. Palaeobiogeographical characteristics of the Miocene land snail fauna of Poland. Scripta Geologica, Special Issue, 2, 397–406. Stworzewicz, E. 1995. Miocene land snails from Belchato´w (Central Poland), I. Cyclophoridae, Pomatiasidae (Gastropoda, Prosobranchia). Pala¨ontologische Zeitschrift, 69, 19–30. Taviani, M. 1989. Late Turolian slugs (Gastropoda: Pulmonata) from vertebrate-rich sedimentary dikes in the Monticino Quarry (Brisighella, Northern Italy). Bollettino della Societa` Paleontologica Italiana, 28, 321–2. Truc, G. 1971a. Gaste´ropodes continentaux ne´ogene`s du bassin Rhodanien. Documents des Laboratoires de Ge´ologie de Lyon, H.S., 79–129. Truc, G. 1971b. Heliceae (Gastropoda) du Ne´oge`ne du bassin Rhodanien (France). Geobios, 4, 273–327. Truc, G. 1971c. Discus (Discus) vireti, nouvelle espe`ce du Mioce`ne infe´rieur de l’Allier (France) (Gastropoda, Euthyneura). Archiv fu ¨ r Molluskenkunde, 101, 141–4. Truc, G. 1972. Clausiliidae (Gastropoda, Euthyneura) du Ne´oge`ne du bassin Rhodanien (France). Geobios, 5, 247–75. Truc, G. 1977. Contributions a` la pale´ontologie du Mioce`ne moyen continental du bassin du Tage. I. Quelques mollusques. Pero Filho, Po´voa de Santare´m, Sı´tio do Mirante. Ciencias de Terra (UNL), 3, 121–7. Wenz, W. 1923–30. Gastropoda extramarina tertiaria. Fossilium Catalogus, 1. Animalia, 17, 18, 20–23, 32, 38, 40, 43, 46, 1–3387. Berlin, W. Junk. Wenz, W. 1936. Die Molluskenfauna der Mergel von Paulhiac (Lot-et-Garonne). Archiv fu ¨ r Molluskenkunde, 68, 228–38. Wenz, W. 1942. Zur Kenntnis der fossilen Land- und Su¨ sswassermollusken Venetiens. Memorie dell’ Istituto di Geologia della Regia Universita` di Padova, 14, 1–51. Wenz, W. & Edlauer, A. 1942. Die Molluskenfauna der oberpontischen Su¨sswassermergel von Eichkogel bei Mo¨dling, Wien. Archiv fu ¨r Molluskenkunde, 74, 82–98. Zilch, A. 1959–60. Gastropoda, 2. Euthyneura. Handbuch der Pala¨ozoologie, 6, 1–834.

17 Sedimentary facies analysis in palaeoclimatic reconstructions. Examples from the Upper Miocene–Pliocene successions of south-central Tuscany (Italy) Marco Benvenuti, Mauro Papini and Giovanni Testa

Introduction One of the purposes of the workshop on ‘Climatic and environmental changes in the Neogene of Europe’, was to assemble diVerent points of view on problems which by their nature need a multidisciplinary approach. Climatic and environmental changes can be detected in the geologic record by palaeontologists, sedimentologists and geochemists, through integration of their eVorts. The goal of this paper is to show how sedimentology can contribute to such multidisciplinary studies. Facies analysis has acquired a great importance in the detailed study of sedimentary deposits. A facies is a rock or sediment unit distinguishable from others on the basis of its speciWc aspect (from Latin facies) including lithologic, sedimentologic, and biologic features. A sedimentary facies records the physical, chemical or biological processes operating during sedimentation. Facies associations and sequences (the spatial arrangement of facies) provide information on the depositional environments and on their dynamics. Selected examples from Upper Miocene–Pliocene continental and shallow marine deposits Wlling some post-collisional basins in southern and central Tuscany (central Italy, Fig. 17.1a), will show in the following paragraphs, the potentialities and limits of facies analysis as an integrative tool for palaeoclimatic reconstructions.

General setting The sedimentary basins we deal with in this paper are located in central and southern Tuscany (central Italy), on the western margin of the Northern Apennines, a thrust and fold chain whose formation started in the Late Cretaceous. In the Middle (Carmignani et al., 1994) or Late Miocene the western side of the chain was aVected by crustal extension, possibly alternated with (Bernini et al. (1990) or subordined to (Boccaletti et al., 1991, 1995) compression. This complex and still debated Neogene structural evolution led to the opening and the development of the Tyrrhenian Sea and the formation of some basins in southern and central Tuscany (Late

Palaeoenvironments: non-mammalian evidence

Sedimentary facies analysis

Miocene–Pliocene ‘central’ basins, sensu Martini & Sagri, 1993) and then in the inner part of the chain (Plio-Pleistocene ‘peripheral’ basins, sensu Martini & Sagri, 1993). The central and peripheral basins (Fig. 17.1a) diVer signiWcantly in their sedimentary Wlls, the former being Wlled with Xuviolacustrine and shallow marine deposits up to 2000 m thick, the latter by Xuvio-lacustrine and alluvial deposits less than 1000 m thick. The examples discussed in the following sections are mainly from the central basins (Fig. 17.1a,b) and from the Upper Valdarno peripheral.

Records of Neogene aridity

The Messinian evaporites of Tuscany The most striking palaeoclimatic event aVecting the whole Mediterranean area in the Late Miocene is without doubt the ‘Messinian Salinity Crisis’ (Ruggieri, 1967), that is the precipitation of a thick succession of evaporitic salts (mainly gypsum and halite) due to the severing of the connections between the oceans and the Mediterranean Sea, that was characterised at that time as well as at present by a negative hydrologic budget. The sedimentary Wll of the extensional basins on the Tyrrhenian margin of the Apennine bears the record of this event. Messinian evaporites are most extensively exposed in two roughly north–south elongated rift basins: the Fine Basin, located near the present coastline, and the Volterra Basin in central Tuscany (Fig. 17.1a). Two distinct gypsum units outcrop in the Fine Basin (Fig. 17.1b). The lower unit (Acquaviva Gypsum) is made up of swallow-tail crystals in growth position (Fig. 17.2), it overlies laminitic diatomaceous marls and diatomites referred by Bossio et al. (1985) to the lowermost Non Distinctive Zone – hence equivalent of the upper part of the Tripoli Fm. of Sicily – and is overlain by sand and marl (Fig. 17.1b) containing the so-called Lago Mare biofacies (Bossio et al., 1985).

[Figure 17.1 (opposite)] (a) Structural sketch of central Italy, showing the distribution of the Neogene basins; FB = Fine Basin; VB = Volterra Basin; VE = Val d’Elsa Basin; BC = Baccinello Basin; VA = Upper Valdarno Basin. (b) Synthetic columns showing successions of depositional environments in time for the various basins: 1) alluvial fan, 2) floodplain, 3) fluvio-lacustrine, 4) brackish (Lago Mare), 5) reef, 6) shallow marine, 7) open marine (Tripoli), 8) shelf. L: lignite, Fp: floodplain, A: aeolian, AG: Acquaviva Gypsum, MG: Marmolaio Gypsum, RaG: Radicondoli Gypsum, RG: Ripaiola Gypsum, SdVG: Saline di Volterra Gypsum.

357

Palaeoenvironments: non-mammalian evidence

358

[Figure 17.2] Swallow-tail twin gypsum crystals in growth position (pins pointing down) in the Radicondoli Formation (Radicondoli-SI). In the upper part of the picture an irregular contact between two beds is visible. At the top of the lower bed the space between swallow-tail crystals is filled with gypsarenite. This facies is known to develop in subaqueous settings of modern salinas and hypersaline lagoons. This facies occurs in the Radicondoli and Acquaviva Fms, and, at lesser extent in the Ripaiola and Marmolaio Fms.

The upper unit (Marmolaio Gypsum – MG) is made up of swallow-tail crystals (Fig. 17.2) in growth position, and gypsarenites diVusely aVected by nodularisation (Lugli & Testa, 1993), it overlies the Lago Mare sand and marl, and is overlain by conglomerate, sand, marl and clastic gypsum (Fig. 17.1b), containing the Lago Mare biofacies (Bossio et al., 1985; Sarti, 1995). The two gypsum units can also be diVerentiated based on their Sr content, the lower containing about 1200 p.p.m. of Sr, whereas the unaltered facies of the upper contain about 500 p.p.m. (Dinelli et al., 1997). Three gypsum units have been recognised by Testa (1996a) in the Late Miocene sedimentary Wll of the Volterra Basin (Fig. 17.1). 1. Radicondoli Gypsum: lowermost Messinian gypsarenite and swallow-tail gypsum in growth position (Fig. 17.2), with intercalations of marl bearing a brackish-water fossil assemblage (Bossio et al., 1978, 1981, 1993) and laterally passing into conglomerate; it unconformably overlies Upper Tortonian Xuvio-lacustrine deposits and is unconformably (Testa, 1995, 1996a) overlain by marine deposits, referred by Bossio et al. (1994, 1996), Sarti & Testa (1994) and Testa (1996a) to the upper part of the Globorotalia conomiozea Zone and the lowermost Non Distinctive Zone (Fig. 17.1b).

Sedimentary facies analysis

359

[Figure 17.3] Nodular microcrystalline gypsum in the Ripaiola Fm. (Faltona-LI). The white nodules are vertically elongated, closely resembling the shape of swallow-tail gypsum crystals (Fig. 17.2), thus being interpreted as pseudomorphs after them. Original swallow-tail gypsum was transformed into anhydrite that consequently turned again to gypsum (Lugli & Testa, 1993); the outer shape was roughly preserved, but the microscopic structure was obliterated. This facies occurs mainly in the Ripaiola and Marmolaio Fms.

2. Ripaiola Gypsum: upper Messinian swallow-tail and nodular microcrystalline gypsum (Figs. 17.2, 17.3); it conformably overlies the lower Messinian marine units, and is unconformably overlain (Fig. 17.1b) by the third evaporitic unit (Testa, 1996a). 3. Saline di Volterra Formation: nodular gypsum (Fig. 17.4) and halite (Fig. 17.5), with intercalations of sand and laminitic clay containing the Lago Mare biofacies; it is unconformably overlain by conglomerate, sand, gypsarenite and clay (Mazzanti, 1966), containing the Lago Mare biofacies (Bossio et al. 1978; Testa, 1996a). The Radicondoli Gypsum lies below Messinian marine deposits correlated to the Globorotalia conomiozea Zone (Fig. 17.1b), that predates the precipitation of evaporites in the rest of the Mediterranean (D’Onofrio et al., 1975; Colalongo et al., 1979; Iaccarino & Salvatorini, 1981; Iaccarino, 1985). The Acquaviva and the Ripaiola formations, which overlie those marine deposits (Fig. 17.1b), can be correlated between them and to the lower evaporites of the Messinian Event (Hsu ¨ et al., 1978). Hence the Saline di Volterra and Marmolaio formations can be correlated to the upper evaporites (Hsu ¨ et al., 1978). The Radicondoli Formation is made up of precipitated gypsum in situ –

Palaeoenvironments: non-mammalian evidence

360

[Figure 17.4] Nodular microcrystalline gypsum from the Saline di Volterra Fm (Faltona-LI). From the bottom observe: (a) elongated nodules that could be interpreted as pseudomorphs after selenite gypsum (Fig. 17.3); (b) tightly and irregularly folded microcrystalline enterolithic gypsum; (c) silts and gypsarenites with a big spheroidal microcrystalline gypsum nodule (d). This succession is typical of a depositional sequence developing from subaqueous to subaerial conditions, with early post-depositional gypsum-anhydrite reactions due to circulation of highly saline ground-water, in a sabkha (flat area close to the sea or a saline lake in arid environment). The sabkha sequence in the Messinian of Tuscany is typically developed in the Saline di Volterra and Marmolaio Fms.

swallow tail crystals in growth position (Hardie & Eugster, 1971; Schreiber et al., 1976; Schreiber, 1982) – and of the product of its intraformational reworking: gypsarenites. Evaporitic condition, thus, alternated with periods of sedimentation in brackish-water, represented by the marly interlayers.

Sedimentary facies analysis

[Figure 17.5] Halite bore-hole core from the Saline di Volterra Fm.The lower part of the core is composed of laminitic-anhydrite with cubic crystals of halite grown within, probably from hypersaline ground-water during subaerial exposure. After an abrupt contact, halite with chevron structures, is interpreted as grown subaqueously at the bottom of the basin. Rapid water-level oscillations are thus recorded also by halite, beside gypsum in the Saline di Volterra Fm. (see Fig. 17.4). Halite occurs only in the Saline di Volterra Fm.

Bossio et al. (1993) interpret this alternation as due to rapid oscillations of the salinity in a lagoon. The presence of clasts of ‘Calcare cavernoso’, a collapse breccia associated with the Triassic Burano Anhydrite Formation (Lugli, 1993), in a conglomerate facies of the Radicondoli Gypsum, and the 34S of the gypsum (17‰, Testa, 1996b), which is closer to the Triassic value (15‰), than to the Messinian normal marine ( 34S = + 22‰; e.g. Longinelli, 1979; Claypool et al., 1980), indicates that the Radicondoli Gypsum precipitated from solutes recycled from Triassic sulphates. A similar origin is reported for the Pleistocene Montallegro Gypsum (Sicily), which is interpreted by Schreiber & Kinsman (1975) as the product of dissolution and re-precipitation of Messinian gypsum. Dinelli et al. (in press), assuming the composition of the modern Elsa river (Fig. 17.1a) as representative of the continental water feeding the Volterra Basin during the Early Messinian ([SO42−] = 400 p.p.m.),

361

Palaeoenvironments: non-mammalian evidence

362

estimated the salinity of the lagoon water, where the Radicondoli Gypsum formed, by mass balance calculations, as ranging around 4‰. This value is compatible with the palaeosalinity inferred by Bossio et al. (1978, 1981) for the marly interlayers on the basis of microfossil ecology. Therefore no dramatic salinity change occurred between the deposition of marls and gypsum. The gypsum of the Radicondoli Formation can be considered as precipitated from mostly continental solution already close to saturation in calcium sulphate – hence requiring only a modest evaporation rate – possibly in a warm–temperate climate as suggested by Bertini (1994) on the basis of palinological analyses. Several lines of evidence allow us to consider the Ripaiola Gypsum and the Acquaviva Gypsum as the product of evaporation of marine waters: their 34S indicate precipitation from marine sulphates (Testa, 1996b; Dinelli et al., 1997); they overlie marine deposits; the foraminiferal microfauna of the underlying marine sediments displays increasing restrictive conditions going upsection; the layers immediately underlying the Wrst gypsum bed are barren (Bossio et al., 1978, 1981, 1985; Sarti & Testa, 1994). Most Messinian successions display the same characteristics in correspondence of the transition from marine to the Wrst evaporitic sediments (Cita et al., 1978). The present day Mediterranean has a negative hydrologic budget, so that damming the Gibraltar Strait would cause an evaporative crisis (Ryan et al., 1973; Hsu¨ et al., 1978); therefore, a climate similar to the present can be invoked to account for the deposition of the Messinian lower evaporite. The situation of the Tuscan basins Wts well in this regional framework. Both the Acquaviva Gypsum and the Ripaiola Gypsum display swallow-tail gypsum crystals in growth position (Fig. 17.2), indicating precipitation in a subaqueous environment. Nodular microcrystalline facies of the Ripaiola Gypsum has been interpreted (Testa, 1996a) as the product of diagenesis of previous swallow-tail gypsum (Fig. 17.3). A signiWcant environmental change occurred concurrently with the sedimentation of the upper evaporite. The Marmolaio Gypsum displays features of subaqueous gypsum precipitation, cyclically interrupted by subaerial exposures and early diagenetic dehydration of gypsum due to circulation of highly saline ground-water (‘sabkhisation’; Lugli & Testa, 1993). Diagenetic structures, typical of the sabkha environment (Kinsman, 1966; Shearman, 1982) are evident in the Saline di Volterra Formation as well (Fig. 17.4; Testa, 1996a). A general continentalisation of the depositional environment is witnessed by the occurrence of abundant plant remains and fresh

Sedimentary facies analysis

water ostracods in the shaly interlayers of both the Marmolaio (Lugli & Testa, 1993), and Saline di Volterra formations (Sarti & Testa, 1994; Testa, 1996a). The halite layers (Fig. 17.5) interbedded in the Saline di Volterra Formation, provide petrographic and geochemical evidence of primary precipitation from non marine solutions (Lugli & Testa, 1996; Testa, 1996b). Therefore, the onset of sabkha conditions and halite precipitation in the Fine and Volterra basins indicates that aridity had increased with respect to the lower evaporite depositional interval. Nonetheless, the cyclic alternation of evaporites and terrigenous sediments deposited in waters of changing inXux (Dinelli et al., in preparation), suggests a cyclic alternation of drier and more humid climates. A clear cyclic pattern is displayed by the Marmolaio Gypsum in the Fine Basin (Bossio et al., 1978; Lugli & Testa, 1993), made up of eight shale–gypsum couplets (Bossio et al., 1978) similar to other upper evaporite sections – e.g. the Colombacci Formation in Romagna (Vai, 1989) and the Pasquasia Gypsum in the Caltanisetta Basin, Sicily (Decima & Wezel, 1973). Krijgsman (1996) recognised 67 ± 2 cycles in the sedimentary record of the Mediterranean evaporitic event (from the base of Tripoli to the base of Pliocene), and interpreted them as originated by forcing of precession cyclicity (average duration of 20 kyr). We consider this the most reasonable hypothesis about the genesis of cyclicity in the Marmolaio Gypsum although further testing is required.

Aeolian deposits in the Upper Valdarno Basin The Upper Valdarno Basin (Fig. 17.1a,b) is known as one of the most famous sites bearing Villafranchian vertebrate faunas. It is a half-graben Wlled with continental deposits up to 600 m thick spanning from the Middle Pliocene to the Present. Magi & Sagri (1994a) distinguished three successions which at the margins of the basin are bounded by erosive and angular unconformities representing therefore three allostratigraphic units composed of Xuviolacustrine and alluvial deposits (Fig. 17.1b). The oldest allostratigraphic unit is made of basal Xuvio-deltaic gravel and sand overlain by lignitiferous palustrine and lacustrine clay topped by Xuvio-deltaic sand. Mammal, pollen, and magnetostratigraphic data (Albianelli et al., 1995, 1997; Torre et al., 1996) allow us to date this succession to the Middle Pliocene. The second allostratigraphic unit, Upper Pliocene– Lower Pleistocene in age, is made of Xuvial and locally aeolian whitish and quartz-rich sand passing upward to shallow lacustrine/palustrine silty clays. The latter are interbedded toward the margins with fan-delta gravels and sands. The third allostratigraphic unit is composed of alluvial-fan and Xuvial gravels, sands and silts referred to the Lower–Middle Pleistocene.

363

Palaeoenvironments: non-mammalian evidence

364

[Figure 17.6] (a) Panoramic view of the aeolian sands in the Upper Valdarno overlain by lacustrine clayey silts. Note the horizontal bedding of the sands. (b) Detailed view of the aeolian sands showing well developed adhesion ripples; the outcrop is 1.5 m thick (photographs courtesy of M. Sagri).

Sedimentary facies analysis

Facies analysis of the basal, sand-rich, portion (Fig. 17.6a) of the second allostratigraphic unit revealed it to be mostly composed of whitish mediumWne grained well sorted, sands in tabular beds up to 2 m thick. Planar lowand subordinate high-angle laminations and characteristic ripple-marks are common structures. The ripple-marks are interpreted to have been adhesion ripples (Fig. 17.6b) in origin. The latter are common features in aeolian setting and form when wind ripples are Wxed by moisture and/or salt coatings. Poorly sorted sands of Xuvial origin are interbedded within the aeolian deposits testifying a possible seasonal alternation of dry and moister conditions. This is further indicated by laterally continuous reddish and mottled horizons possibly indicating Xuctuations of the water-table. On the whole the basal sands of the second allostratigraphic unit are interpreted to be accumulated in ephemeral Xuvial systems (terminal fans; Mukerji, 1976) which in the distal portions were dominated by sheet Xooding and aeolian reworking during dry periods, to form aeolian sand sheets, small dunes and adhesion ripples (Magi & Sagri, 1994b, 1996). Palynologic analysis of thin pelitic levels interbedded with sands (Bertini, 1994; Albianelli et al., 1995) record expansion of the herbaceous cover indicating dry and cold conditions and support the palaeoenvironmental picture of an ephemeral Xuvial system developed under periodically arid conditions. These deposits are referred to a cold–arid pulse, tentatively related to the 2.6 Ma arctic glaciation (Bertini, 1994) whose precursor stage is recorded in the underlying, oldest, Xuvio-lacustrine allounit by pollen data (Bertini & Roiron, 1997).

Mio-Pliocene lignites of Tuscany Lignite is a common facies in the Neogene deposits of the Tuscan basins and Upper Miocene–Pliocene lignites are generally present in the lowermost deposits Wlling these basins. In the Baccinello-Cinigiano Basin in Southern Tuscany (Fig. 17.1a,b) a 3 m thick lignite bed is present at the base of Tortonian–Messinian lacustrine and alluvial deposits up to 500 m thick (Benvenuti et al., 1994). From a chronological point of view such a lignite can be reasonably referred to a Late Tortonian on the base of a MN 13 mammal fauna found just below the exposed seam (Benvenuti et al., 1994). The lignite itself bears a quite rich mammal fauna including the hominoid Oreophitecus bambolii. Facies analysis of the visible portion of the lignite seam reveals four main constituting facies (Fig. 17.7): 1) whitish massive calcareous marl; 2) thinly laminated lignite and marl with abundant fresh water molluscs;

365

Palaeoenvironments: non-mammalian evidence

366

[Figure 17.7] Lithologic log of the exposed lignite seam in the Baccinello-Cinigiano Basin.

3) massive lignite; 4) massive organic clay rich in Bithinia opercula and other fresh water molluscs. Facies 1, 2 and 3 are stacked vertically in two sequences, whereas facies 4 is observed only on top of the second facies sequence and is abruptly overlain by open lake muds. In the Upper Valdarno Basin (Fig. 17.1a,b) two lignite seams, the lower one 15 m thick and the upper one 5 m thick, occur in the lower–medium part of the oldest Middle Pliocene allostratigraphic unit. A magnetostratigraphic calibration (Albianelli et al., 1995) refers both these deposits to the late Gauss chron. The lower seam, bearing scanty early Villafranchian mammal remains (Triversa faunal unit; Albianelli et al., 1997) can be more precisely referred to the Kaena subchron. The two lignite seams, are encased in rhythmic silty-clayey deposits interpreted to be deposited in quite deep and stratiWed lakes (Fialdini, 1988). The two lignite seams show similar, irregularly alternating, facies represented by (Fig. 17.8): a) brownish tree trunks in life position; b) massive blackish lignite; c) whitish medium-Wne sands in lenticular beds.

Sedimentary facies analysis

[Figure 17.8] Detailed view of the strongly deformed lower lignite seam in the Upper Valdarno Basin, note tree trunks in life position (facies a). The outcrop is 2 m thick (photograph courtesy of M. Fialdini).

In the Baccinello-Cinigiano Basin the development of a peatland occurred in two distinct episodes through progradation of a reed peat into a shallow marly lacustrine area. Such a process, generally known as ‘terrestrialisation’ (Dachnowsky, 1912; Tallis, 1973; Shotyk, 1992), is a progressive migration of palustrine vegetational belts towards the central part of shallow lakes and ponds. In the general case, marginal palustrine vegetation prevents the arrival of siliciclastic input allowing authigenic or biogenic carbonate deposition in the central parts of the lakes. In the Upper Valdarno the development of swamps occurred twice, possibly as an abrupt transition from an open and stratiWed-water lake, suggesting a lake-level lowering followed by the establishment of a swampy soil. Then the swamp development was periodically aVected by Xuvial Xoods with deposition of the sandy interlayers. Palynologic data indicate for both the Baccinello-Cinigiano (Harrison & Harrison, 1989; Benvenuti et al., 1994) and Upper Valdarno lignite seams (Albianelli et al., 1995; Bertini & Roiron, 1997) a sub-tropical/warm-moist temperate climate during their development. Was a moist climate really the dominant factor for the development of peatland in the Tuscan basins during the Late Miocene and the Middle Pliocene? Generally speaking, well developed and stable humid conditions in the

Palaeoenvironments: non-mammalian evidence

368

geologic record should not be automatically inferred from coals alone (McCabe, 1984). Considering the present day distribution of wetlands and swamplands we can see, in fact, how they can form in a wide range of climatic settings not excluding hot and seasonally dry areas (Moore, 1987). Although precipitation continuity throughout the year, a feature presently limited to the equatorial belt, is a favouring factor for the terrestrial productivity, a major control on peatland development and preservation of peat is exerted by the stability of an adequate groundwater level through the growing season (Ziegler et al., 1987; Lottes & Ziegler, 1994). In some important cases the occurrence of wide peatland in seasonally dry areas such as those in east and south Africa (Sudd and Okavango swamps) is allowed by low topographic gradients and tectonic subsidence, which on the whole cause the water table to be relatively high for long periods. Nevertheless, the present day worldwide peat distribution reXects in large part the rainfall pattern, representing therefore a suitable uniformitarian tool for the interpretation of the distribution and changes of climate in the geologic record (Parrish et al., 1982; Ziegler et al., 1987). With these considerations in mind we interpret the Neogene lignites of the Tuscan basins to record local favourable conditions for development and preservation of peats mainly controlled by the tectonic dynamics of the basins. The balance between the peat growth and both the relative lake levels and water table in the soils was in most part controlled by the low rate tectonic subsidence in the early stages of the tectono-sedimentary evolution of both the Baccinello-Cinigiano and the Upper Valdarno basins. The abrupt Xooding of swamps, decay of peats and subsequent burial with open lake muds was in both cases due to rapid increase of tectonic subsidence. Similar considerations on the various occurrence of lignites in the Neogene–Quaternary basins of Tuscany have been made by Martini & Sagri (1993), who found the formation of peatland during the Late Miocene to have been common in the supposed tectonically controlled lower part of the basins.

Neogene flood-dominated alluvial successions A common facies component in the Tuscan Upper Miocene–Pliocene successions is represented by alluvial coarse-grained deposits (tens to hundreds of metres thick). These coarse deposits mark in most cases the early stages of the basin Wll and record major modiWcations of the basin shoulders induced by tectonism. Alluvial deposits show proximal and distal facies assemblages. The former are generally represented by irregular alternation of poorly sorted gravels in thick (up to 10 m) often amalgamated beds,

Sedimentary facies analysis

sands, and locally, silts. Intervening palaeosols are also frequent. These proximal deposits are generally interpreted as formed in alluvial fans or in Xood dominated, high-gradient rivers which distributed coarse grained materials in the basins, through highly concentrated Xows (Benvenuti & Martini, 1997). Depending on the basinal conditions, that is whether lacustrine–marine areas were present or not, such systems graded distally into the subaqueous portion of fan deltas or into Xoodplains respectively. An example of a Xoodplain facies assemblage occurs in the Middle Pliocene alluvial to shallow marine succession of the Valdelsa Basin (Fig. 17.1a,b). This basin is Wlled with about 2000 m of Neogene–Quaternary deposits (Ghelardoni et al., 1968). The Pliocene succession cropping out in the central portion of the basin can be broadly subdivided into three main units which from the bottom are: 1) Xuvio-deltaic sand and alluvial gravel, respectively on the western and eastern margins; 2) late Lower–Middle Pliocene (Conti, 1993) marine shelf deposits on the west, and coastal plain sediments on the northeast; 3) alternances of paralic and shallow marine sand and mud and Xuvio-deltaic gravel, sand and mud the latter on the eastern margin; these units bear Early–Middle Villafranchian vertebrates referred to the Triversa and Montopoli faunal units (Benvenuti et al., 1995a) which on the whole are correlated to the interval 3.3–2.6 Ma. The Xoodplain facies assemblage of the basal portion of unit 3 is represented by channellised bodies of sand and gravel up to 10 m thick (Fig. 17.9), resting erosively on massive to thinly laminated greyish mud locally rich in plant remains, land snails and evidences of pedogenic modiWcations (calcareous nodules, roots, mottles). The channellised bodies are massive in the basal part, passing upward into clinostratiWed beds (Fig. 17.9) of coarse– medium sand and gravel up to 50 cm thick, outlining large macroforms up to 3 m high and a few tens of metres long, interpreted as Xood bars and lobes. Palaeocurrent analysis shows that growth of the bars occurred both through lateral and frontal accretion. On the whole these deposits record the development of a Xood-dominated Xuvial system (Mutti et al., 1996) characterised by massive deposition of sand and gravel in the early stage of the channel Wll, followed by growth of bars due to sudden Xood-Xows expansion. The Wlling of such channels was followed by the development of a Xoodplain characterised by poorly drained conditions with settlement of Wne-grained deposits and pedogenic modiWcatons. The alternation of Xood channels and Xood plain suggests some cyclic control on the activation/ deactivation of these depositional systems. Such cyclicity can be explained

369

[Figure 17.9] Floodplain deposits in the Valdelsa Basin. A flood bar composed of inclined gravelly beds (1) is erosively overlain by sands filling a shallow channel (2). This in turn is capped by floodplain mudstones (3).

Sedimentary facies analysis

as controlled by Xood events without signiWcant change of base level (Mutti et al., 1996) or as the combined eVect of high frequency base-level oscillations and Xood events (Wright & Marriot, 1993). Beside these possibilities, the climatic imprint of these deposits points to an episodic and Xashy regime of discharges, reXecting highly variable hydrology of the alluvial system, and hence variability of the rainfall pattern. Palynologic data for equivalent paralic and shallow marine deposits in the central part of the basin (Valleri et al., 1990; Benvenuti et al., 1995b) indicate a subtropical/warm–temperate climate during their deposition. Facies analysis of the coeval alluvial plain areas suggests that such a climatic regime was prone to generate Xoods. This is a feature that could Wnd explanation in Xood–climatologic considerations (Hayden, 1988). The Mediterranean region is presently within the range of seasonal Xuctuations of the low-latitude (barotropic) and high latitude (baroclynic) atmospheric circulation modes, characterised by diVerent patterns of storm generation. In barotropic regions storms are generally concentrated in the summer (monsoons) and related to the Intertropical Convergence Zone and to tropical cyclons. In the baroclynic regions, meteoric events are due to frontal storms, more eYcient in generating high magnitude Xoods at middle latitudes. The western Mediterranean region, including the Tyrrhenian coast of Italy, is presently characterised by a Xood climatology that seasonally shares features with both the barotropic and baroclynic circulations, being therefore located in a critical area for Xood generation. Moreover, Xoods in this region are (and have been in the past) enhanced by heavy orographic rainfall, favoured by the presence of high mountain chains. Here we make the preliminarly suggestion that the Neogene alluvial successions from Tuscany possibly represent a palaeoclimatic archive which through integrated analysis eventually will furnish valuable data on the Xood pattern during the Middle Pliocene.

Conclusion Case histories from the Neogene successions of Tuscany indicate that facies analysis can contribute to palaeoclimatic and palaeoenvironmental reconstructions. Evaporites are generally regarded as indicators of an arid climate. This is not always true, as shown by the case presented above, where a detailed facies analysis, together with geochemical and palaeontological data, indicates arid conditions only for the Messinian upper evaporites of the Fine and Volterra basins (central Tuscany, Italy). Periods of aridity cyclically

371

Palaeoenvironments: non-mammalian evidence

372

alternated with more humid ones, probably as a consequence of precessional forcing, as suggested for other Messinian upper evaporites sections. In fact, we envision conditions that are not particularly arid for the deposition of early Messinian Radicondoli Gypsum and for late Messinian lower evaporites. Aeolian and ephemeral stream sediments in the Upper Pliocene continental succession of the Upper Valdarno Basin record periodically arid conditions, reXecting a climatic deterioration which on the basis of palynological data is tentatively correlated with the 2.6 Ma arctic glaciation. Peat formation testiWes to an equilibrium between local subsidence and peat growth, and is therefore a good indicator of stasis or low-rate tectonism in the early stages of basin development. The evolution of peatlands occurred in the Tortonian of the Baccinello-Cinigiano Basin as a progressive migration of the palustral vegetation fringe toward the centre of shallow lakes (terrestrialisation). In the Middle Pliocene of the Upper Valdarno Basin, swamp development possibly occurred following lake-level lowering. In both cases, common also to other examples from Tuscany, the peatlands were abruptly Xooded and replaced with open lacustrine conditions. Although lake levels can be inXuenced also by climatic change such as variation in rainfall, we consider the generalised deepening upward recorded in the Wrst stage of the basin Wlls as the eVect of abrupt increase of tectonic subsidence in the basins (cf. Lambiase, 1989). Alluvial deposits form in many cases the bulk of the Neogene basin Wlls and although poorly investigated for their palaeoclimatic meaning, bear some information on past climates. In the Middle Pliocene deposits of the Valdelsa Basin, both proximal and distal alluvial facies associations, point to depositional mechanisms involving both high energy and high sediment concentration of Xoodwaters. Relevance of Xoods in ancient Xuvial and deltaic systems has recently been discussed by Mutti et al. (1996) but little is known on their palaeoclimatic and palaeohydrologic implications. Future integrated research should be directed to decipher this record, taking into account the dynamics of Xood generation due to the atmospheric circulation (cf. Hayden, 1988) and its possible reactions to past climatic change.

References Albianelli, A., Azzaroli, A., Bertini A., Ficcarelli, G., Napoleone, G. & Torre, D. 1997. Paleomagnetic and palynologic investigations in the Upper Valdarno Basin (Central Italy): calibration of an Early Villafranchian fauna. Riv. It. Pal. Strat., 103, 111–18.

Sedimentary facies analysis

Albianelli, A., Bertini, A., Magi, M., Napoleone, G. & Sagri, M. 1995. Il bacino plio-pleistocenico del Valdarno Superiore: eventi deposizionali, paleomagnetici e paleoclimatici. Il Quaternario, 8, 11–18. Benvenuti, M. & Martini, I. P. 1997. Sedimentological facies analysis of alluvial, highly-concentrated-Xow deposits. In Debris-Xow hazards mitigation: mechanics, prediction, and assessment, Cheng-lung Chen (ed.), ASCE, Proceedings, 496–505. Benvenuti, M., Bertini, A. & Rook, L. 1994. Facies analysis, vertebrate paleontology and palynology in the Late Miocene, Baccinello-Cinigiano Basin (Southern Tuscany). Mem. Soc. Geol. It., 48, 415–23. Benvenuti, M., Dominici, S. & Rook, L. 1995a. Inquadramento stratigraWco-deposizionale delle faune a mammiferi villafranchiane (unita` faunistiche Triversa e Montopoli) del Valdarno Inferiore nella zona a sud dell’Arno (Toscana). Il Quaternario, 8, 457–64. Benvenuti, M., Bertini, A., Conti, C., Dominici, S. & Falcone D. 1995b. Analisi stratigraWca e paleoambientale integrata del Pliocene dei dintorni di S. Miniato. Quad. Mus. St. Nat. Livorno, 14 (suppl.1), 29–49. Bernini, M., Boccaletti, M., Moratti, G., Papani, G., Sani, F. & Torelli, L. 1990. Episodi compressivi neogenico-quaternari nell’area estensionale tirrenica nord-orientale. Dati in mare e a terra. Mem. Soc. Geol. It., 45, 577–89. Bertini, A. 1994. Palynological investigations on Upper Neogene and Lower Pleistocene sections in central and northern Italy. Mem. Soc. Geol. It., 48, 431–43. Bertini, A. & Roiron, P. 1997. E ´ volution de la ve´ge´tation et du climat pendant le Plioce`ne moyen, en Italie centrale: apport de la palynologie et de la macroXore a` l’e´tude du bassin du Valdarno supe´rieur (coupe de Santa Barbara). C. R. Acad. Sci. Paris, 324 (se´rie IIa), 763–71. Boccaletti, M., Bonini, M., Moratti, G. & Sani, F. 1995. Nuove ipotesi sulla genesi e l’evoluzione dei bacini post-nappe in relazione alle fasi compressive neogenico-quaternarie dell’Appennino Settentrionale. Acc. Naz. Sci. detta dei XL, Scritti e Documenti, 14, 229–62. Boccaletti, M., Cerrina Feroni, A., Martinelli, P., Moratti, G., Plesi, G. & Sani, F. 1991. L’area Tosco-Laziale come dominio di transizione tra il bacino tirrenico e i thrusts esterni: rassegna di dati mesostrutturali e possibili relazioni con le discontinuita` del ‘Ciclo Neoautoctono’. Mem. Descr. Carta Geol. It., 49, 9–22. Bossio, A., Cerri, A., Mazzei, R., Salvatorini, G. & Sandrelli, F. 1996. Geologia dell’area Spicchiaiola-Pignano (settore orientale del bacino di Volterra). Boll. Soc. Geol. It., 115, 393–422. Bossio, A., Costantini, A., Foresi, L., Mazzanti, R., Monteforti, B., Salvatorini, G., Sandrelli, F. & Testa, G. 1994. Note preliminari sul neoautoctono dell’area di Sassa (settore SO del bacino di Volterra) provincie di Livorno e Pisa. Studi Geologici Camerti, vol. spec., 1994/1, 33–43. Bossio, A., Costantini, A., Lazzarotto, A., Liotta, D., Mazzanti, R., Mazzei, R., Salvatorini, G. & Sandrelli, F. 1993. Rassegna delle conoscenze sulla stratigraWa del neoautoctono toscano. Mem. Soc. Geol., It., 49, 17–98. Bossio, A., Esteban, M., Giannelli, L., Longinelli, A., Mazzanti, R., Mazzei, R., Ricci Lucchi, F. & Salvatorini, G. 1978. Some aspects of the upper Miocene in Tuscany. Messinian Seminar, 4, Pacini, Pisa, pp. 88.

373

Palaeoenvironments: non-mammalian evidence

374

Bossio, A., Giannelli, L., Mazzanti, R., Mazzei, R. & Salvatorini, G. 1981. Il passaggio dalla facies lacustre alla evaporitica e le ‘Argille a Pycnodonta’ presso Radicondoli (Siena). IX Conv. Soc. Pal. It., Weld trip guidebook, Pacini Pisa, pp. 161–74. Bossio, A., Mazzanti, R., Mazzei, R. & Salvatorini, G. 1985. Analisi micropaleontologiche delle formazioni mioceniche, plioceniche e pleistoceniche dell’area del Comune di Rosignano. Quad. Mus. St. Nat. Livorno, 6, (suppl. 1), 129–70. Carmignani, L., Decandia, F. A., Fantozzi, P. L., Lazzarotto, A., Liotta, D. & Meccheri, M. 1994. Tertiary extensional tectonics in Tuscany (northern Apennines, Italy). Tectonophysics, 238, 295–315. Cita, M. B., Colalongo, M. L., D’onofrio, S., Iaccarino, S. & Salvatorini, G. 1978. Biostratigraphy of Miocene deep-sea sediments with special reference to the Messinian/pre-Messinian interval. Init. Rep. Deep Sea Drill. Proj., 42, 671–85. Claypool, G. E., Holser, W. T., Kaplan, I. R., Sakai, H. & Zak, I. 1980. The age curves of sulfur and oxygen isotopes in marine sulfate and their mutual interpretation. Chemical Geology, 28, 199–260. Colalongo, M. L., Di Grande, A., D’Onofrio, S., Giannelli, L., Iaccarino, S., Mazzei, R., Poppi Brigatti, M. F., Romeo, M., Rossi, A. & Salvatorini, G. 1979. A proposal for the Tortonian-Messinian Boundary. Ann. Geol. Pays Hellen.T.h.s., f. 1, VIIth International Congress on Mediterranean Neogene, Athens. Conti, C. 1993. Studio biostratigraWco e sedimentologico dei sedimenti pliocenici dell’area nord-orientale del Bacino dell’Elsa (Certaldo). Dipartimento di Scienze della Terra, Universita` di Firenze. Unpublished Laurea Thesis, pp. 193. D’Onofrio, S., Giannelli, L., Iaccarino, S., Morlotti, E., Romeo, M., Salvatorini, G., Sampo`, M. & Sprovieri, R. 1975. Planktonik foraminifera of the Miocene from some Italian Sections and the Problem of the Lower Boundary of the Messinian. Boll. Soc. Paleot. It., 14, 177–96. Dachnowski, A. 1912. Peat deposits of Ohio. Their origin, formation, and uses. Geol. Surv. Ohio Bull., 16, 424. Decima, A. & Wezel, F. 1973. Late Miocene evaporites of the Central Sicilian Basin, Italy. Init. Rep. Deep Sea Drill. Proj., 13, 1234–44. Dinelli, E., Testa, G., Cortecci, G. & Barbieri, M. (in press) Stratigraphic and Petrographic constraints to trace element and isotope geochemistry of Messinian sulfates of Tuscany. Mem. Soc. Geol. It. Dinelli, E., Lugli, S., Testa, G., Barbieri, M. & Cortecci, G. 1997. Petrographic constraints to trace elements and isotope geochemistry of Messinian sufates of Tuscany. In Neogene Mediterranean Palaeoceanography, ‘Ettore Majorana’ Centre for ScientiWc Culture Erice, Italy, abstract volume. Fialdini, M. 1988. I depositi Xuvio-lacustri villafranchiani inferiori del gruppo di Castelnuovo dei Sabbioni (Valdarno Superiore): indagini sedimentologiche e geochimiche. Dipartimento di Scienze della Terra, Universita` di Firenze. Unpublished Laurea Thesis, pp. 136. Ghelardoni, R., Giannini, E. & Nardi, R. 1968. Ricostruzione paleogeograWca dei bacini neogenici e quaternari nella bassa valle dell’Arno sulla base dei sondaggi e dei rilievi sismici. Mem. Soc. Geol. It., 7, 91–106. Gliozzi, E., Abbazzi, L., Argenti, P., Azzaroli A., Caloi, L., Capasso Barbato, L., Di Stefano, G., Esu, D., Ficcarelli, G., Girotti, O., Kotsakis, T., Masini, F., Mazza, P., Mezzabotta, C., Palombo, M. R., Petronio, C., Rook, L., Sala, B., Sardella, R.,

Sedimentary facies analysis

Zanalda, E. & Torre, D. 1997. Biochronology of selected large mammals, molluscs and ostracods from the Middle Pliocene to the Late Pleistocene in Italy. The state of the art. Riv. It. Pal. Strat., 103, 369–88. Hardie, L. A. & Eugster, H. P. 1971. The depositional environments of marine evaporites: a case for shallow siliciclastic accumulation. Sedimentology, 16, 187–220. Harrison, T. S. & Harrison, T. 1989. Palynology of the Late Miocene Oreophitecus-bearing lignite from Baccinello, Italy. Pal. Pal. Pal., 76, 45–65. Hayden, B. P. 1988. Flood climates. In Flood Geomorphology, Baker, V. R., Kochel, R. C. & Patton, P. C. (eds.), pp. 13–26. Wiley. Hsu¨, K. J., Montadert, L., Bernoulli, D., Cita, M. B., Erickson, A., Garrison, R. E., Kidd, R. B., Melieres, F., Mu ¨ ller, C. & Wright, R. 1978. History of the Mediterranean salinity crisis. Init. Rep. Deep Sea Drill. Proj., 42, 1053–78. Iaccarino, S. 1985. Mediterranean Miocene and Pliocene planktic foraminifera. In Plankton Stratigraphy, Bolli, H. M., Saunders, J. B. & Perch Nielsen, K. (eds.), pp. 283–314. Cambridge University Press. Iaccarino, S. & Salvatorini, G. 1982. A framework of planktonic foraminiferal biostratigraphy for Early Miocene to Late Pliocene Mediterranean area. Paleont. stratigr. ed evoluz. Quad., 2, 115–25. Kinsman, D. J. J. 1966. Gypsum and anhydrite of Recent age, Trucial Coast, Persian Gulf. In Proc. Second Salt Symposium, Rau, J. L. (ed.), Northern Ohio Geol. Soc., 1, 302–26. Krijgsman, W. 1996. Miocene magnetostratigraphy and cyclostratigraphy in the Mediterranean: extension of the Astronomical Polarity Time Scale. Geol. Ultraiectina, 141, 207. Lambiase, J. J. 1989. A model for tectonic control of lacustrine stratigraphic sequences in continental rift basins. Am. Ass. Petr. Geol. Mem., 50, 265–76. Longinelli, A. 1979. Isotope geochemistry of some Messinian evaporites: palaeoenvironmental implications. Pal. Pal. Pal., 29, 95–123. Lottes, A. L. & Ziegler, A. M. 1994. World peat occurrence and the seasonality of climate and vegetation. Pal. Pal. Pal., 106, 23–37. Lugli, S. 1993. Studio petrograWco e geochimico sulle origini e sulla evoluzione post-deposizionale della Formazione Evaporitica di Burano (Trias Sup.). Dipartimento di Scienze della Terra di Modena. Unpublished PhD. Thesis, pp. 150. Lugli, S. & Testa, G. 1993. The origin of the gypsum alabaster spheroids in the Messinian evaporites from Castellina Marittima (Pisa, Italy): preliminary observations. Giorn. Geologia 55, 51–68. Lugli, S. & Testa, G. 1996. The Messinian halite rocks from the Volterra basin (central Tuscany, Italy). 17th IAS Reg. Meeting, Sfax (Tunisia), abstract volume. Magi, M. & Sagri, M. 1994a. The Valdarno Basin. 15th IAS Reg. Meeting, Ischia (Italy), Excursion B4 Guidebook, 54–61. Magi, M. & Sagri, M. 1994b. Terminal fan deposits in the Plio-Pleistocene Valdarno Basin (Northern Apennines, Italy). 15th IAS Reg. Meeting, Ischia (Italy), abstract volume, 261–2. Magi, M. & Sagri, M. 1996. Aeolian sand sheets and terminal fans: sedimentological evidences of arid climatic conditions in the Plio-Pleistocene Valdarno Basin (Northern Apennines, Italy). 17th IAS Reg. Meeting, Sfax (Tunisia), abstract volume, 172–3.

375

Palaeoenvironments: non-mammalian evidence

376

Martini, I. P. & Sagri, M. 1993. Tectono-sedimentary characteristics of Late Miocene-Quaternary extensional basins of the Northern Apennines, Italy. Earth Sci. Rev., 34, 197–233. Mazzanti, R. 1966. Geologia della zona di Pomarance – Larderello (prov. di Pisa). Mem. Soc. Geol. It., 5, 105–38. McCabe, P. J. 1984. Depositional environment of coal and coal-bearing strata. Int. Ass. Sediment., Spec. Publ., 7, 13–42. Moore, P. D. 1987. Ecological and hydrological aspects of peat formation. Geol. Soc. Spec. Publ., 32, 7–15. Mukerji, A. B. 1976. Terminal fans of inland streams in Sutlej-Yamuna plain, India. Z. Geomorph., 20, 190–204. Mutti, E., Davoli, G., Tinterri, R. & Zavala, C. 1996. The importance of ancient Xuvio-deltaic systems dominated by catastrophic Xooding in tectonically active basins. Mem. Sci. Geol. Padova, 48, 233–91. Parrish, J. T., Ziegler, A. M. & Scotese, C. R. 1982. Rainfall patterns and the distribution of coals and evaporites in the Mesozoic and Cenozoic. Pal. Pal. Pal., 40, 67–101. Ruggieri, G. 1967. The Miocene and later evolution of the Mediterranean Sea. In Aspects of Tethyan Biogeography: Syst. SSOC. Publ. No. 7, Adams, C. G. & Ager, D. V. (eds.), pp. 238. Ryan, W. B. F., Hsu ¨ , K. J. et al. 1973. Initial Reports of Deep Sea Drilling Project, Leg XIII, pp. 1447. Sarti, G. 1995. I depositi neogenici della Val di Fine (Toscana a sud dell’Arno) nel quadro dell’evoluzione del margine tirrenico della catena appenninica tra il Tortoniano superiore ed il Pliocene basale., Dipartimento di Scienze della Terra Pisa. Unpublished PhD Thesis, pp. 236. Sarti, G. & Testa, G. 1994. Studio geologico stratigraWco della successione tortoniana superiore - pliocenica del margine occidentale del bacino di Volterra (Pisa). Mem. Descr. Carta Geol. It., 49, 189–228. Schreiber, B. C. 1982. Environments of subaqueous gypsum deposition. SEPM publication, 4, 43–73. Schreiber, B. C., Friedman, G. M., Decima, A. & Schreiber, E. 1976. The depositional environment of the upper Miocene (Messinian) deposits of the Sicilian Basin. Sedimentology, 23, 729–60. Schreiber, B. C. & Kinsman, D. J. J. 1975. New observations on the Pleistocene evaporites of Montallegro, Sicily and a modern analog. Jour. Sed. Petr., 45, 469–79. Shearman, D. J. 1982. Evaporites of coastal sabkhas. SEPM publication, 4, 6–42. Shotyk, W. 1992. Organic soils. In Weathering, Soils and Paleosols, Martini, I. P. & Chesworth, W. (eds.), pp. 203–24. Elsevier. Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Augustı´, J. 1996. Circum-mediterranean Neogene (Miocene and Pliocene) marine-continental chronologic correlations of european mammal units. In The Evolution of Western Eurasian Neogene Mammal Faunas, R. L. Bernor et al. (eds.), pp. 7–46. Columbia University Press. Tallis, J. H. 1973. The terrestrialization of lakes basins in North Cheshire, with special reference to the development of a ‘Schwingmoor’ structure. Journ. Ecol., 61, 537–67. Testa, G. 1995. Upper Miocene extensional tectonics and syn-rift sedimentation in the western sector of the Volterra basin (Tuscany, Italy). Studi Geologici Camerti,

Sedimentary facies analysis

vol. spec. 1995/1, 617–30. Testa, G. 1996a. Il bacino evaporitico messiniano di Volterra: evoluzione sedimentaria, tettonica e diagenetica. Dipartimento di Scienze della Terra Pisa. Unpublished PhD Thesis, pp. 172. Testa, G. 1996b. The Messinian evaporites from Tuscany. Petrology and diagenetic evolution. Plinius, 15, 168–72. Torre, D., Albianelli, A., Bertini, A., Ficcarelli, G., Masini, F. & Napoleone, G. 1996. Paleomagnetic calibration of Plio-Pleistocene mammal localities in central Italy. Acta Zool. Cracov., 39, 559–70. Vai, G. B. 1989. A Weld-trip guide to the Romagna Apennine geology. The Lamone valley. In Fossil vertebrates in the Lamone valley, Romagna Apennine. Field trip guidebook, De Giuli, C & Vai, G. B. (eds.), pp. 7–37. Valleri, G., Bertoldi, R. & Bertini, A. 1990. Studio delle associazioni a foraminiferi e a pollini del Pliocene di Ponte a Elsa (Valdarno Inferiore, Toscana). Boll. Soc. Paleont. It., 29, 321–33. Wright, V. P. & Marriot, S. B. 1993. The sequence stratigraphy of Xuvial depositional systems: the role of Xoodplain sediment storage. Sediment. Geol., 86, 203–10. Ziegler, A. M., Raymond, A. L., Gierlowski, T. C., Horrell, M. A., Rowley, D. B. & Lottes, A. L. 1987. Coal, climate and terrestrial productivity: the present and early Cretaceous compared. Geol. Soc. Spec. Publ., 32, 25–49.

377

18 Neogene vegetation changes in West European and West circumMediterranean areas Jean-Pierre Suc, Se´ verine Fauquette, Mostefa Bessedik, Adele Bertini, Zhuo Zheng, Georges Clauzon, Danica Suballyova, Filomena Diniz, Pierre Que´ zel, Najat Feddi, Martine Clet, the late Ezzedine Bessais, Naima Bachiri Taoufiq, Henriette Meon and Nathalie Combourieu-Nebout

Introduction The value of pollen records for vegetation and climate reconstructions is now well-established. Above all, the West Mediterranean region is highly documented, especially for what concerns the Lower Pliocene (Suc et al., 1995a) and oVers a favourite Weld for climatic quantiWcations (Fauquette et al., 1998a, 1998b and in press). The next step is to construct vegetation maps in order to (1) emphasize new problems for a better understanding of past and modern ecosystems, (2) to estimate vegetation changes on a large geographic scale, and (3) to provide data (on the continental carbon mass for example) to climate modellers (Chandler et al., 1994; Sloan et al., 1996). The Wrst aim of this paper is to lay the foundations of such a project for the Mediterranean and Western European regions. Then, a review of vegetation changes throughout the whole Neogene is presented concerning a region rich in data (Southwestern Europe). Most of the pollen localities used in this paper are chronologically well-calibrated according to foraminifers and/or nannoplankton and, for some of them, palaeomagnetism and rodents (for more details, see Bessedik, 1984; Suc, 1989; Suc et al., 1995a).

Vegetation maps Validation of pollen records Modern pollen spectra (lacustrine, lagoonal or marine surface samples) are compared to present-day vegetation on a map which also takes into account cultivated surfaces (Fig. 18.1). We now make some comments on the value of pollen data. Warm-temperate deciduous trees are mostly predominant in the Northwestern Mediterranean area and northward. Pines and non-identiWed Pinaceae (poorly preserved pollen grains) are often overrepresented because of their proliWc production and the advantage they have in air and water transport. They will be not considered in our com-

Neogene vegetation changes

parisons because they inhabit several vegetation belts. They are particularly abundant in some localities, mostly marine (localities 7, 9, 13, 14, 17, 19). Herbaceous pollen grains are numerous in the South Mediterranean area, in addition to some edaphic conditions (pond and/or lagoon environments: localities 2 and 8). Subtropical elements have only been recorded in some meridional localities (11, 12, 14). Thermophilous Mediterranean evergreen plants (Olea, Pistacia, Cistus, Rhamnus) characterize South Mediterranean landscapes, often in association with Artemisia steppe (localities 10, 11, 12, 13, 14, 15, 16). Some details are also signiWcant: (1) cedar has been recorded only around North Africa (localities 10, 11, 12, 13) where it is living (Rif Massif and Tellian mountains, southward of Algiers); (2) Wr and spruce, high altitude or high latitude trees, have been recorded near high mountains (Pyrenees: sites 5, 6; Alps: sites 3, 4, 7, 8) or northward (site 1). So, at the level of general vegetation assemblages, pollen spectra (including those obtained on marine sediments) can be considered as representative of vegetation, as has been demonstrated for the Rhoˆne delta by Cambon et al. (1997).

Method for drawing palaeovegetation maps A Wrst attempt for world-wide vegetation mapping has been made by Thompson & Fleming (1996), based on an empirical use of pollen records combined with the General Circulation Model simulation of the mid-Pliocene climate (Chandler et al., 1994). The Mediterranean and Western European regions are at most represented by three colours in such a map, that appears inadequate by reference to the vegetation variety expressed by pollen data (Suc et al., 1995a,b). Our aim was to map palaeovegetation using up to 11 colours, that correspond to the main vegetation assemblages identiWed in the pollen diagrams (7 forest environments, 4 open environments; see legend of Fig. 18.2). After a very detailed botanical inventory of pollen Xora, the pollen diagram is turned into a synthetic pollen diagram (Suc, 1984) in which taxa grouping is representative of vegetation organizing with the exception of pines and the ecologically insigniWcant elements (the two white coloured pollen groups, respectively to the left and to the right in the pollen diagrams) which concern several ecosystems and/or vegetation belts; each pollen group corresponds to a vegetation type. The forest vegetation types run according to the thermic gradient from the tropical assemblage to the high altitude/latitude forest. In addition, some taxa such as Cathaya and Cupressaceae which take a prominent part in some areas have been individualized. The edaphic coastal ecosystems

379

Palaeoenvironments: non-mammalian evidence

380

(such as the Avicennia mangrove or the Taxodium swamps) have been respectively included within the two Wrst thermic groups. The open environments are also considered in this classiWcation, the subdivision of which is mostly controlled by variation in moisture. The vegetation maps of which we have experience are drawn in the palaeogeographic context (littoral outline for instance) of the concerned areas (Ro¨gl & Steininger, 1983; Dercourt et al., 1985; Boccaletti et al., 1990; Clauzon et al., 1990; Suc et al., 1992). Relief elevation is deduced from the literature but also from Xora composition (Fauquette et al., accepted). Vegetation is mapped for areas which enclose a signiWcative density of coeval pollen diagrams. Mapping is interrupted when the density of pollen records appears inadequate. Percentages of pollen groups constituting the synthetic pollen diagram are more or less directly transferred in vegetation surfaces, taking into account their altitudinal distribution and their generally known respective under-/over-representativityfor the concerned vegetation types (Muller, 1959; Koreneva, 1971; Tissot, 1980; Heusser, 1983, 1988; Sun & Wu, 1988; Chmura & Liu, 1990; Jackson, 1994; Streel & Richelot, 1994; Cambon et al., 1997). It is probable that the Wrst attempt is highly empiric but the mapping will be shortly developed on a quantitative background, stimulated by the biome method (Prentice et al., 1992) but with increasing details as in Wgure 3 of Fauquette et al. (1998a). Vegetation of three short periods has been mapped corresponding to the above-mentioned criteria: early Aquitanian (c. 24 Ma) for Southeastern France; Langhian (c. 15 Ma) for Southeastern France and central Catalonya (Northeastern Spain); early Zanclean (5.3 to 5.0 Ma) for the entire West Mediterranean and West European regions.

The early Aquitanian vegetation Southern France is rich in coastal pollen localities (Bessedik, 1984), among which three have been selected as being very representative (Fig. 18.2). They allow a conclusive comparison between land bordering the Vence Gulf that opened towards the Mediterranean Sea and those bordering the recently formed Gulf of Lions thanks to the Corsica–Sardinia rifting. The eastern part of the Corsica–Sardinia massif was covered by coniferous forests varying according to altitudinal organization. Coastal open vegetation was very reduced and Sequoia forest developed immediately from the shoreline. The western part of the study area was characterized by more open and drier

Neogene vegetation changes

landscapes, such as also supported by sedimentary environments (evaporitic basins of Aix-en-Provence and Portel; Chamley & Nury, 1973). The Avicennia mangrove scarcely existed. At mid-altitude, the forest was rather diVerent, being mainly constituted by evergreen Juglandaceae (Engelhardia, Platycarya) and, above, by deciduous trees (Ulmus and Zelkova among the most common).

The Langhian vegetation Two areas can be compared (Bessedik, 1984; Bessedik & Cabrera, 1985) (Fig. 18.3). The Avicennia mangrove was more important northward, as well as the tropical vegetation (riparian forest?). Open dry environments were more developed on the eastern side of the peri-Alpine sea and around the Valle`s Penede`s basin (Catalonya). Nevertheless, composition of the evergreen and mixed-mesophytic forest was almost the same as during the previous period.

The early Zanclean vegetation Early Zanclean is the most documented period in West Europe and Northwest Africa (Fig. 18.4a). The palaeogeography is very close to the modern one, except for the deep rias which penetrated lands as a result of the preceding Messinian salinity crisis (cutting of canyons by rivers because of the desiccation of the Mediterranean Sea; Clauzon et al., 1990, 1996). In addition to the pollen diagrams presented on Fig. 18.4a (Bertini, 1994; Bessais & Cravatte, 1988; Clet-Pellerin, 1983; Feddi, unpublished; Menke, 1975; Me´on et al., 1990; Suc, 1989 and unpublished; Suc & Bessais, 1990; Suc & Cravatte, 1982; Suc et al., 1986, 1995b; Zagwijn, 1960; Zheng, 1990), we have used the data published by Sittler & Geissert (1993) over the AlsaceVosges area and by Me´on-Vilain (1970) over the Rhoˆne Valley. Forest cover prevailed to the North of the Mediterranean region whereas open environment prevailed to the South (Fig. 18.4b). It has been possible to subdivide the studied area into three main vegetation domains (Suc et al., 1995a,b; Fig. 18.4b): West Europe, North Mediterranean, South Mediterranean. West Europe shows an almost homogenous vegetation organization from Germany to Portugal, mostly constituted by Taxodium swamps at low altitude and deciduous forests at mid-altitude. Ericaceae moors already took an important place in this domain. In the North Mediterranean domain, vegetation organization was mostly controlled by proximity of high mountains (Pyrenees, Alps, Apennines).

381

Palaeoenvironments: non-mammalian evidence

382

In some places, the coastal plain was inhabited by Taxodium swamps (Barcelona area and Po Valley for instance). Near mountains, forest environments are dominated by Sequoia and a strong altitudinal gradient can be evidenced (Sequoia forest successively replaced by deciduous forest, then by Cathaya, Tsuga and Cedrus forest, at last by Abies and Picea belt). Mediterranean xerophytic vegetation (Quercus ilex, Phillyrea, etc.) has developed in some sunny and rocky slopes. In detail, pollen records show variations in coastal vegetation such as today in the Mississippi delta and in Florida (Taxodium swamps and Cyperaceae marshes; Munaut, 1976). In the South Mediterranean domain, open environments predominated while forest was restricted to altitudinal belts. Such subdesertic assemblages are rich in Asteraceae, Poaceae, Convolvulus, Geraniaceae, Nitraria, Neurada, Calligonum, Lygeum, etc. A transitional latitudinal zone spanned from Tarragona to Sicily, characterized by elements belonging today to the thermo-Mediterranean formation (Olea, Pistacia, Ceratonia, Ziziphus, etc.). This phytogeographical subdivision is also supported by the results of the climatic transfer function established on pollen data (Fauquette et al., 1998a, 1998b and in press), the mean annual temperature and the annual precipitation being indicated (on average) in Fig. 18.4b. It is obvious that a strong break in rainfall existed between the latitudes of Barcelona and Tarragona, whereas temperature progressively increased from the North to the South. As far as vegetation and climate thresholds are concerned, another interesting aspect is to be discussed. In the earliest Pliocene, localities from the Northern edge of the Pyrenees mountains (Roussillon ria: Suc, 1976; Cravatte et al., 1984; Suc et al., 1992) are a little bit less rich in Mediterranean xerophytes than the localities of the Southern Pyrenean edge (Ampurdan and Llobregat rias: Suc & Cravatte, 1982) (see the diVerence between localities 11 and 12 on Fig. 18.4a). Their content of taxa living under humid conditions (as Taxodiaceae) is almost the same. This might indicate more or less similar climatic conditions on both sides of the Pyrenees mountains as pointed out by the climatic transfer function (Fig. 18.4b). Then, Taxodiaceae disappeared from the Roussillon ria (Suc, 1976; Cravatte et al., 1984) at a stratigraphic level belonging to the normal chron C3n.2n (Aguilar et al., 1998), i.e. at about 4.5 Ma. An increase in Mediterranean xerophytes preceded this event in the Languedoc and Roussillon pollen diagrams (see localities 10 and 11 on Fig. 18.4a). In the area of Barcelona, Taxodiaceae disappeared after the onset of the North Hemisphere glacial–interglacial cycles (Suc & Cravatte, 1982; Suc, 1984), i.e. after 2.6 Ma. Nowadays, the forest vegetation is more developed on the meridional Pyrenean slope than on the

Neogene vegetation changes

Northern one which appears drier, probably because of an intense wind action (Tramontane and Mistral winds). Maybe the appearance of the boreal winds might have favoured the early disappearance of Taxodiaceae on the Northern edge of Pyrenees whereas decreases in temperature and especially in moisture (Fauquette et al., 1998b) could have forced their extinction in Catalonya.

Neogene vegetation changes in Southwestern Europe Only Southeastern France and Southwestern Po Valley show a more or less continuous pollen record in the studied area throughout the whole Neogene (Fig. 18.5). Two important breaks occurred respectively in the Upper Serravallian–Lower Tortonian and in the earliest Gelasian. The Wrst break is characterized by: (1) the disappearance of tropical elements; (2) the replacement of Avicennia by Taxodium within the coastal edaphic environments; (3) the strong decrease of Engelhardia, Platycarya, Sapotaceae, etc. among the subtropical elements; (4) the decrease of some Mediterranean xerophytes (Olea mostly: still represented by tropical–subtropical species?); (5) the development of altitudinal trees (Cathaya, Cedrus, Tsuga, Abies and Picea). This break corresponds to a decrease in temperature. It may have inXuenced the life of animals such as the hominoid primates. Indeed, before the break, trees belonged to various climatic assemblages and forests were able to provide fruits all the year long. Then, fruit production was reduced to some months. Pollen data support the hypothesis of Andrews (1992), and may constitute an argument to explain the extinction of most hominoid primates from Western Europe at about 9 Ma (Late Vallesian–Early Turolian; Andrews et al., 1996). The second break is marked by the development of the Artemisia steppe during glacials. In parallel, forests evolved towards their modern status (progressive extinction of subtropical species).

Conclusion The interest in mapping Neogene vegetation is obvious. Pollen records are now in suYcient quantity and of good-enough quality to shortly develop such reconstructions. This present attempt is empirical, and will be improved upon by a statistical approach. In this way, a narrow statistical connection is to be sought between modern pollen spectra and present-day

383

Palaeoenvironments: non-mammalian evidence

384

vegetation. Then, extrapolation methods will be applied in order to cover areas devoid of any pollen record. West European and West circum-Mediterranean vegetation has changed greatly during the past 25 Ma, not only because of climatic forcing but also of relief uplift. Biotopes have been modiWed that may contribute towards explaining changes in mammal fauna.

Acknowledgements We acknowledge the scientists who provided to us modern surface samples: M. B. Cita, P. Cochonat, R. Dalongeville, C. Pierre, B. Savoye, C. Vergnaud Grazzini, Institut Franc¸ais de Recherche pour l’Exploitation de la Mer, Institut Franc¸ais de la Recherche et de la Technologie Polaire, Centre des Faibles Radioactivite´s (CEA/CNRS). P. Mein is acknowledged for providing helpful information.

References Aguilar, J.-P., Legendre, S., Michaux, J. & Montuire, S. 1998. Pliocene mammals and climatic reconstruction in the Western Mediterranean area. In The Pliocene: Time of Change, Wrenn, J. H., Suc, J.-P. & Leroy, S. A. G. (eds.), pp. 109–20. American Association of Stratigaphic Palynologists Foundation. Andrews, P. 1992. Evolution and environment in the Hominoidea. Nature, 360, 641–6. Andrews, P., Harrison, T., Delson, E., Bernor, R. L. & Martin, L. 1996. Distribution and Biochronology of European and Southwest Asian Miocene Catarrhines. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 168–207. Columbia University Press. Bertini, A. 1994. Messinian-Zanclean vegetation and climate in North-Central Italy. Historical Biology, 9, 3–10. Bessais, E. & Cravatte, J. 1988. Les e´cosyste`mes ve´ge´taux plioce`nes en Catalogne me´ridionale. Variations latitudinales dans le domaine nord-ouest me´diterrane´en. Geobios, 21, 1, 49–63. Bessedik, M. 1984. The Early Aquitanian and Upper Langhian-Lower Serravallian environments in the Northwestern Mediterranean region. Pale´obiologie Continentale, 14, 2, 153–79. Bessedik, M. 1985. Reconstitution des environnements mioce`nes des re´gions nord-ouest me´diterrane´ennes a` partir de la palynologie. Thesis, University Montpellier 2, pp. 162. Bessedik, M. & Cabrera, L. 1985. Le couple re´cif-mangrove a` Sant Pau d’Ordal (Valle`s-Penede`s, Espagne), te´moin du maximum transgressif en Me´diterrrane´e nord-occidentale (Burdigalien supe´ rieur-Langhien infe´rieur). Newsletters Stratigraphy, 14, 1, 20–35.

Neogene vegetation changes

Boccaletti, M., CiaranW, N., Cosentino, D., Deiana, G., Gelati, R., Lentini, F., Massari, F., Moratti, G., Pescatore, T., Ricci Lucchi, F. & Tortorici, L. 1990. Palinspastic restoration and paleogeographic reconstruction of the peri-Tyrrhenian area during the Neogene. Palaeogeography, Palaeoclimatology, Palaeoecology, 77, 41–50. Brun, A. 1979. Recherches palynologiques sur le Golfe de Gabe`s: re´sultats pre´liminaires. Ge´ologie Me´diterrane´enne, 6, 1, 247–64. Cambon, G., Suc, J.-P., Aloı¨si, J.-C., Giresse, P., Monaco, A., Touzani, A., Duzer, D. & Ferrier, J. 1997. Modern pollen deposition in the Rhoˆne delta area (lagoonal and marine sediments), France. Grana, 36, 105–13. Chamley, H. & Nury, D. 1973. Pale´oge´ographie du bassin de se´dimentation d’Aix-en-Provence a` l’Oligoce`ne moyen: donne´es nouvelles des mine´raux argileux. Comptes Rendus Sommaires de la Socie´te´ Ge´ ologique de France, 15, 1, 31–2. Chandler, M., Rind, D. & Thompson, R. 1994. Joint investigations of the middle Pliocene climate II: GISS GCM Northern Hemisphere results. Global and Planetary Change, 9, 197–219. Chmura, G. L. & Liu, K.-B. 1990. Pollen in the lower Mississippi River. Review of Palaeobotany and Palynology, 64, 253–61. Clauzon, G., Suc, J.-P., Aguilar, J.-P., Ambert, P., Cappetta, H., Cravatte, J., Drivaliari, A., Dome´nech, R., Dubar, M., Leroy, S., Martinell, J., Michaux, J., Roiron, P., Rubino, J.-L., Savoye, B. & Vernet, J.-L. 1990. Pliocene geodynamic and climatic evolution in the French Mediterranean region. Paleontologia i Evolucio´, special issue, 2, 131–86. Clauzon, G., Suc, J.-P., Gautier, F., Berger, A. & Loutre, M.-F. 1996. Alternate interpretation of the Messinian salinity crisis: Controversy resolved? Geology, 24, 4, 363–6. Clet-Pellerin, M. 1983. Le Plio-Ple´istoce`ne en Normandie. Apports de la palynologie. Thesis, University Caen, 135 pp. Cravatte, J. & Suc, J.-P. 1981. Climatic evolution of north-western Mediterranean area during Pliocene and early Pleistocene by pollen-analysis and forams of drill Autan 1. Chronostratigraphic correlations. Pollen et Spores, 23, 2, 247–58. Cravatte, J., Matias, I. & Suc, J.-P. 1984. Nouvelles recherches biostratigraphiques sur le Plioce`ne du Roussillon. Ge´ologie de la France, 1–2, 149–63. Dercourt, J., Zoneshain, L. P., Ricou, L.-E., Kazmin, V.G., Le Pichon, X. L., Knipper, A. L., Grandjacquet, C., Sborshchikov, I. M., Boulin, J., Sorokhtin, O., Geyssant, J., Lepvrier, C., Biju-Duval, B., Sibuet, J.-C., Savostin, L. A., Westphal, M. & Lauer, J.-P. 1985. Pre´sentation de neuf cartes pale´oge´ographiques au 1/20 000 000 s’e´tendant de l’Atlantique au Pamir pour la pe´riode du Lias a` l’Actuel. Bulletin de la Socie´te´ ge´ologique de France, ser. 8, 1, 5, 637–52. Diniz, F. 1984. Etude palynologique du bassin plioce` ne de Rio Maior. Pale´obiologie continentale, 14, 259–67. Fauquette, S., Clauzon, G., Suc, J.-P., Zheng, Z. & Guiot, J. (accepted for publication). A new approach for paleoaltitude estimates based on pollen records: example of the Mercantour massif (Southeastern France) at the earliest Pliocene. Earth and Planetary and Science Letters. Fauquette, S., Que´zel, P., Guiot, J. & Suc, J.-P. 1998a. SigniWcation bioclimatique de taxons-guides du Plioce`ne me´diterrane´en. Geobios, 31, 2, 151–69.

385

Palaeoenvironments: non-mammalian evidence

386

Fauquette, S., Guiot, J. & Suc, J.-P. 1998b. A method for climatic reconstruction of the Mediterranean Pliocene using pollen data. Palaeogeography, Palaeoclimatology, Palaeoecology, 144, 1–2, 183–201. Fauquette, S., Suc, J.-P., Guiot, J., Diniz, F., Feddi, N., Zheng, Z., Bessais, E. & Drivaliari, A. (in press). Climate and biomes in the West Mediterranean area during the Pliocene. Palaeogeography, Palaeoclimatology, Palaeoecology. Heusser, L. E. 1983. Contemporary pollen distribution in coastal California and Oregon. Palynology, 7, 19–42. Heusser, L. E. 1988. Pollen distribution in marine sediments on the continental margin oV northern California. Marine Geology, 80, 131–47. Jackson, S. T. 1994. Pollen and spores in Quaternary lake sediments as sensors of vegetation composition: theoretical models and empirical evidence. In Sedimentation of Organic Particles, Traverse, A. (ed.), pp. 253–86. Cambridge University Press. Koreneva, E. V. 1971. Spores and pollen in Mediterranean bottom sediments. In The Micropaleontology of Oceans, Funnell, B. M. & Riedel, W. R. (eds.), pp. 361–71. Cambridge University Press. Leroy, S. 1989. Pale´oclimats plio-ple´istoce`nes en Catalogne et Languedoc d’apre`s la palynologie de formations lacustres. Thesis, University Louvain-la-Neuve, pp. 522. Menke, B. 1975. Vegetationsgeschichte und Florenstratigraphie Nordwest-Deutschlands im Plioza¨n und Fru¨hquarta¨r. Mit einem Beitrag zur Biostratigraphie des Weichselfru¨hglazials. Geologische Jahrbuch, ser. A, 26, 3–151. Me´on-Vilain, H. 1970. Palynologie des formations mioce` nes supe´rieures et plioce`nes du bassin du Rhoˆne (France). Documents des Laboratoires de Ge´ologie de la Faculte´ des Sciences de Lyon, 38, 1–167. Me´on, H., Bachiri, N. & Puisse´gur, J.-J. 1990. Analyse sporopollinique du sondage de Beaune (NW de la Bresse, France). Stratigraphie et restitution climatique. Revue de Micropale´ontologie, 32, 4, 277–90. Muller, J. 1959. Palynology of Recent Orinoco delta and shelf sediments. Micropaleontology, 5, 1–32. Munaut, A. V. 1976. Paysages ve´ge´taux de la Floride me´ridionale. Les Naturalistes Belges, 57, 73–99. Naud, G. & Suc, J.-P. 1975. Contribution a` l’e´tude pale´oXoristique des Coirons (Arde`che): premie`res analyses polliniques dans les alluvions sous-basaltiques et interbasaltiques de Mirabel (Mioce`ne supe´rieur). Bulletin de la Socie´te´ Ge´ologique de France, ser. 7, 5, 820–7. Pirazzoli, P., Planchais, N., Rosset-Moulinier, M. & Thommeret, J. 1981. Interpre´tation pale´oge´ographique d’une tourbe de Torson di Sotto (Lagune de Venise, Italie). Ge´ologie Me´diterrane´enne, 7, 3, 121–8. Planchais, N. 1985. Analyses polliniques du remplissage holoce`ne de la lagune de Canet (plaine du Roussillon, de´partement des Pyre´ne´es-orientales). Ecologia Mediterranea, 11, 1, 117–27. Prentice, I. C., Cramer, W., Harrison, S. P., Leemans, R., Monserud, R. A. & Solomon, A. M. 1992. A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography, 19, 117–34.

Neogene vegetation changes

Ro ¨ gl, F. & Steininger, F. F. 1983. Vom Zerfall der Tethys zu Mediterran und Paratethys. Die neogene Pala¨ogeographie und Palinspastik des zirkum-mediterranen Raumes. Annales Naturhistorische Museum Wien, 85, A, 135–63. Sittler, C. & Geissert, F. 1993. Flore palynologique et Xore carpologique, un te´moignage pale´obioge´ographique conjugue´, au Mioce`ne et au Plioce`ne dans la valle´e du Rhin. Palynosciences, 2, 15–37. Sloan, L. C., Crowley, T. J. & Pollard, D. 1996. Modeling of middle Pliocene climate with the NCAR GENESIS general circulation model. Marine Micropaleontology, 27, 51–61. Streel, M. & Richelot, C. 1994. Wind and water transport and sedimentation of miospores along two rivers subject to major Xoods and entering the Mediterranean Sea at Calvi (Corsica, France). In Sedimentation of Organic Particles, Traverse, A. (ed)., pp. 59–67. Cambridge University Press. Suc, J.-P. 1976. Apports de la palynologie a` la connaissance du Plioce`ne du Roussillon (sud de la France). Geobios, 9, 6, 741–71. Suc, J.-P. 1984. Origin and evolution of the Mediterranean vegetation and climate in Europe. Nature, 307, 429–32. Suc, J.-P. 1989. Distribution latitudinale et e´tagement des associations ve´ge´tales au Ce´nozoı¨que supe´rieur dans l’aire ouest-me´diterrane´enne. Bulletin de la Socie´te´ Ge´ologique de France, ser. 8, 5, 3, 541–50. Suc, J.-P., Bertini, A., Combourieu-Nebout, N., Diniz, F., Leroy, S., Russo-Ermolli, E., Zheng, Z., Bessais, E. & Ferrier, J. 1995a. Structure of West Mediterranean vegetation and climate since 5.3 Ma. Acta Zoologica Cracoviense, 38, 1, 3–16. Suc, J.-P. & Bessais, E. 1990. Pe´rennite´ d’un climat thermo-xe´rique en Sicile avant, pendant, apre`s la crise de salinite´ messinienne. Comptes Rendus de l’Acade´mie des Sciences de Paris, ser. III, 310, 1701–7. Suc, J.-P., Clauzon, G., Bessedik, M., Leroy, S., Zheng, Z., Drivaliari, A., Roiron, P., Ambert, P., Martinell, J., Dome´nech, R., Matias, I., Julia`, R. & Anglada, R. 1992. Neogene and Lower Pleistocene in Southern France and Northwestern Spain. Mediterranean environments and climate. Cahiers de Micropale´ontologie, 7, 1–2, 165–86. Suc, J.-P. & Cravatte, J. 1982. Etude palynologique du Plioce`ne de Catalogne (nord-est de l’Espagne). Pale´obiologie Continentale, 13, 1, 1–31. Suc, J.-P., Diniz, F., Leroy, S., Poumot, C., Bertini, A., Dupont, L., Clet, M., Bessais, E., Zheng, Z., Fauquette, S. & Ferrier, J. 1995b. Zanclean ( ~ Brunssumian) to early Piacenzian ( ~ early-middle Reuverian) climate from 4° to 54° north latitude (West Africa, West Europe and West Mediterranean areas). Mededelingen Rijks Geologische Dienst, 52, 43–56. Suc, J.-P., Legigan, P. & Diniz, F. 1986. Analyse pollinique de lignites ne´oge`nes des Landes: Arjuzanx et Hostens (France). Bulletin de l’Institut de Ge´ologie du Bassin d’Aquitaine, 40, 53–65. Sun, X. & Wu, Y. 1988. The distribution of pollen and algae in surface sediments of Dianchi, Yunnan Province, China. Review of Palaeobotany and Palynology, 55, 193–206. Thompson, R. S. & Fleming, R. F. 1996. Middle Pliocene vegetation: reconstructions, paleoclimatic inferences, and boundary conditions for climate modeling. Marine Micropaleontology, 27, 27–49.

387

Palaeoenvironments: non-mammalian evidence

388

Tissot, C. 1980. Palynologie et e´volution re´cente de deux mangroves du Tamil Nadu (Inde). Travaux et Documents de Ge´ographique Tropicale, 39, 109–214. Zagwijn, W. H. 1960. Aspects of the Pliocene and early Pleistocene vegetation in The Netherlands. Mededelingen van de Geologische Stichting, ser. C, 3, 5, 1–78. Zheng, Z. 1990. Ve´ge´tations et climats ne´oge`nes des Alpes maritimes franco-italiennes d’apre`s les donne´es de l’analyse palynologique. Pale´obiologie Continentale, 17, 217–44.

PART IV

Palaeoenvironments: mammalian evidence

19 Shrews (Mammalia, Insectivora, Soricidae) as paleoclimatic indicators in the European Neogene Jelle W. F. Reumer

The present paper is based on a more extensive study published a few years ago in a chapter in the book Paleoclimate and Evolution with Emphasis on Human Origins (Vrba et al., 1995). It is thought appropriate both by the author and the editors to republish the most important aspects of that article in the framework of the present volume, even though no new points of view are presented here.

Introduction The study of the paleoecological relevance of shrews is ruled by an extremely simple mathematical principle. Imagine a cube with sides of 4 cm each. The total surface of the cube (six squares of 4 ; 4 = 16 cm2 each) is 6 ; 16 = 96 cm2. The volume or content of the cube is 4 ; 4 ; 4 = 64 cm3. A second cube has sides of 2 cm each. Its surface is 6 ; 2 ; 2 = 24 cm2. Its volume is 2 ; 2 ; 2 = 8 cm3. A third cube has sides of 1 cm each, a surface of 6 cm2 and a volume of 1 cm3. The Wrst cube has a surface-to-volume (S/V) ratio of 96: 64 = 1.5. The second cube has an S/V ratio of 24: 8 = 3. The third cube has an S/V ratio of 6: 1 = 6. This simple sequence shows that when a body acquires dimensions that are smaller by a factor of two, its S/V ratio doubles. Evidently such simple calculations not only apply to cubes, but to all three-dimensional objects, including mammals. Thus, if a mammal becomes smaller, its surface becomes larger in relation to its volume. The surface of a mammal (or of any object for that matter) is the location of temperature exchanges with the surrounding environment. To use another example: when a cup of boiling water and a bucket of boiling water are placed in a freezer, the cup will become frozen more rapidly than the bucket, because it has a relatively larger surface through which to lose its heat.

Shrews and paleoclimate Shrews are among the smallest living homoiothermic animals. This means that – in comparison to other mammals – they have an extremely high surface-to-volume ratio. As mammals lose considerable amounts of heat through their skin, a high S/V ratio is unfavourable for heat retention.

Shrews in the European Neogene

Shrews therefore have a tendency to cool oV rapidly. This tendency is counterbalanced by maintaining a very high metabolic rate, a relatively high oxygen consumption, and by having a nearly constant need for food intake. Shrews need to live in environments that are not too cold, and that provide an uninterrupted food source. Given the fact that the diet of Soricidae consists as a rule of small invertebrates, this latter provision implies that the environment may not be too arid either. Ecological studies showed that environmental moisture may be the ultimate determinant of within-habitat diversity and numerical abundance of soricids (Feldhamer et al., 1993; see also Vogel, 1980; Reumer, 1985, 1989; ChurchWeld, 1990, for details). It has been shown that the composition and diversity of soricid faunas can be taken as a measure of ecological adequacy of the environment for shrews. This conclusion can be used in both directions. In the Wrst place, if an association of fossil mammals contains an abundance of Soricidae, this can be taken as evidence of a relatively warm and humid paleoclimate. In the second place, if a certain area is known to have had a relatively warm and humid paleoclimate at a given period of time, it should come as no surprise that the biodiversityof Soricidae in a fossil mammal associationis ratherhigh. Therefore, changes in the composition of an association of fossil Soricidae can be interpreted in a paleoclimatic sense. Observed increasing biodiversity should have been caused by a paleoclimatic change towards warmer and/or more humid situations. A drop in soricid diversity should have been caused by increasing aridity and/or decreasing temperatures. This general conclusion implies that Soricidae can be used in paleoclimatic interpretations as one of the biotic parameters.

Taxonomical framework When considering the European Neogene, nine groups (subfamilies and tribes) must be considered: the Crocidosoricinae, the Allosoricinae, the Crocidurinae, and within the subfamily Soricinae the tribes Soricini, Blarinellini, Anourosoricini, Blarinini, Neomyini, and Beremendiini. (For a detailed picture of the taxonomic relationships within the family Soricidae see Reumer, 1998.) They are here brieXy discussed (data on strictly American groups are omitted for the sake of briefness).

Crocidosoricinae This group of plesiomorphic shrews originated in Asia, where representatives are found in strata of Early Oligocene age. Europe was reached after

391

Palaeoenvironments: mammalian evidence

392

the ‘Grande Coupure’ event, which roughly corresponds to the MP 20/ MP 21 boundary (MP = Mammalian Paleogene), about 33 Ma ago, although apparently not immediately after that event. The European Wrst appearance date is in an MP 23 locality (Saint-Martin-de-Castillon, Southern France). The Crocidosoricinae witnessed their maximum abundance both in terms of biodiversity and geographical distribution in the Early Miocene (Ramblian–Aragonian; biozones MN 3 and MN 4 (MN = Mammalian Neogene)). Then they began to decrease, to disappear during the early Late Miocene (early Vallesian, MN 9, c. 11 Ma ago). By that time they lived only in Spain, after having been spread all over Europe during Early and Middle Miocene times. The subfamily Crocidosoricinae is at the base of shrew phylogeny; all other subfamilies are thought derived from it.

Allosoricinae The most important genus of this bigeneric subfamily, Paenelimnoecus Baudelot, 1972, witnessed a range from the Ramblian/Orleanian (MN 3, Wintershof-West) up till the early Villa´nyian (MN 16, Mala Cave, Poland and BetWa 13, Rumania). By the time of MN 4, the genus Paenelimnoecus was well established at least in Central Europe. The extinction came after the important soricid extinction wave around the Pliocene/Pleistocene boundary.

Soricinae: Soricini The tribe may have originated in North America around the Arikareean/ Hemingfordian boundary, that is c. 20 Ma ago. A problem is that from that time onwards, the record of the Soricini is obscure, both in Eurasia and in North America. True and clearly identiWable Soricini do not occur until the latest Miocene or Early Pliocene. In Europe, the late Turolian witnesses the Wrst occurrence of the genera Sorex L., 1758 and DeinsdorWa Heller, 1963 (Maramena, Greece, situated around the MN 13/MN 14 boundary). Representatives of this tribe are still living in Europe.

Soricinae: Blarinellini Hemisorex sp. from the German locality of Stubersheim 3 (MN 3–4) may represent the earliest occurrence of this group of shrews. It has been concluded that this tribe originated around the MN 3/MN 4 boundary (the Ramblian/Aragonian boundary), roughly 18–19 Ma ago, and probably in

Shrews in the European Neogene

MN 3, some 20 Ma ago. Representatives of the Blarinellini thrived in Europe during the late Miocene and the Pliocene; one extant genus (Blarinella Thomas, 1911) is found in Central Asia. Soricinae: Anourosoricini This tribe came into existence during Late Miocene times. The tribe, of which one Asian representative (Anourosorex squamipes Milne-Edwards, 1872) is still extant, is Wrst encountered in localities in Spain in the form of Crusafontina endemica Gibert, 1975. The type-locality for this species is Can Llobateres, which is also the stratotype for mammal zone MN 9 (Early Vallesian). This means that this lineage started c. 11 Ma ago. Several genera lived in Eurasia during Late Miocene and Pliocene times, of which Amblycoptus Kormos, 1926, was the most successful one. Soricinae: Blarinini The Blarinini came into existence in America during the Early Hemphillian, c. 9 Ma ago. The early Ruscinian of Europe witnessed the appearance of Blarinini of the genera Blarinoides Sulimski, 1959, and MaWa Reumer, 1984, followed in the Late Ruscinian by Sulimskia Reumer, 1984. Given the early presence of Blarinini in America c. 9 Ma ago and the Wrst appearance in Europe c. 4.5 Ma ago, it can be assumed that the migration took place from America into Eurasia. It is, however, unknown how and when the Blarinini crossed the Beringian barrier between America and Europe. Presently, Blarinini live only in America, and they are the only shrews to have reached South America. Soricinae: Neomyini This tribe is exclusively Eurasian. Many forms are still extant, especially in Asia, which could suggest an Asiatic origin of this tribe. The earliest record of a Neomyini is from the latest Turolian of Greece: Asoriculus gibberodon (Petenyi, 1864) from Maramena. This makes this lineage one of a series of developments that appeared around the Miocene/Pliocene boundary. Within the European realm, the Neomyini never were very diverse; the genus Asoriculus developed one major species (A. gibberodon) and gave rise to several late Neogene and Pleistocene island endemics of the genus Nesiotites Bate, 1944.

393

Palaeoenvironments: mammalian evidence

394

Soricinae: Beremendiini The last distinct lineage within the Soricinae is formed by the tribe Beremendiini. Its only European genus is Beremendia Kormos, 1934, with four described species of Eurasiatic origin. The Early Ruscinian fauna from Osztramos 1 (Hungary) can be considered the earliest record of Beremendiini: both species B. minor Rzebik, 1976, and B. Wssidens (Petenyi, 1864) are found in this fauna. Especially B. Wssidens proved to be an extremely successful species, which is to be found in many Ruscinian and Villa´nyian localities, and which continued well into the Pleistocene. The tribe is now extinct.

Crocidurinae Finally, the subfamily Crocidurinae comes into the European fossil record during the Late Ruscinian. This late appearance comes as a great surprise, as at present the crocidurines are by far the most diverse group of shrews, both in terms of biogeographic distribution and of morphology. The surprise becomes even greater if one considers the rather plesiomorphic nature of several characters of the dentition and mandibular osteology. The Crocidurinae must form an old group, but the fossil record does not conWrm this idea. The oldest Wnd that came to my attention is a Crocidura sp. from the Aegean island of Rhodes, in the Apolakkia Local Fauna, dated to MN 15 (Middle Ruscinian).

Results In Reumer (1995) a range chart of these groups of shrews, and of American groups as well, is given. When studying the ranges of these diVerent groups of shrews and the taxa of which they are consituted, it becomes evident that three periods of time can be distinguished in which groups Wrst appeared and that are interpreted as periods of enhanced speciation (Reumer, 1995). The Wrst such period is found in the Early Miocene, some 19–20 Ma ago. We notice the Wrst appearance of the Allosoricinae, of the tribe Blarinellini, and possibly also the Wrst appearance (in America) of the tribe Soricini. Similar developments are described from North Ameria (viz. the Wrst appearacne of the subfamily Limnoecinae). Seen on a global scale, two, and possibly three, of the four subfamilies that arose from crocidosoricine stock Wrst appeared in this time-span. The second period of speciation occurred in the early Late Miocene, more or less coinciding with the Vallesian, roughly between 9 and 11 Ma

Shrews in the European Neogene

ago. This period witnessed the disappearance of the Crocidosoricinae from continental Europe, and the Wrst appearance of the tribe Anourosoricini in Europe; again, parallel developments are described from the New World (viz. the Wrst appearance of the soricine tribes Blarinini and Notiosoricini). The third period coincides more or less with the Pliocene (Ruscinian, somewhat between 6.0 and 2.4 Ma ago). The tribes Soricini, Neomyini and Beremendiini appear and the Wrst Crocidurinae is noted in Europe. Around or just beyond the end of the Ruscinian the last Allosoricinae disappear, perhaps somewhat later than the 2.4 Ma extinction wave, but certainly as a part of the Pliocene/Pleistocene boundary shrew crisis (Reumer, 1984, 1985). It is interesting to see what these three periods of time have in common and what circumstances prevailed that might have stimulated shrew evolution. For this, we have to make comparisons with published climatic curves and other relevant paleoclimatic data (Daams et al., 1988; Van der Meulen & Daams, 1992; Calvo et al., 1993; Reumer, 1995, and references therein). Comparison of the three mentioned time-spans with the overwhelming amount of published paleoclimatic data shows that periods of high relative humidity coincide with the periods during which the Soricidae proliferate (Reumer, 1995). The conclusion therefore seems justiWed that proliferation of shrews is stimulated by humid paleoclimates. Climatic deterioration around the Pliocene/Pleistocene boundary (c. 2.7–2.3 Ma ago) caused the European shrew fauna to severely impoverish, thus showing an eVect of temperature as well. The conclusion that humid paleoclimates per se stimulated shrew proliferation is interesting, but needs to be looked at with some caution. It can not be excluded that while such humid conditions coincide with general abundance and geographic spreading of shrews, the actual speciation events took place before that. More arid and cooler periods preceding the warmer and humid conditions could have forced the shrews into smaller, allopatric populations that were under climatic pressure. This would have been a situation that is generally considered to be conducive to speciation.

Summary Shrews (Soricidae) are very small mammals that are maladapted to extremely cold and/or extremely arid environments. When the ranges of the various distinct groups of shrews are plotted against a paleoecological background, it is concluded that most events (Wrst appearances and last appearances) took place in periods of time when humid paleoclimates prevailed. En-

395

Palaeoenvironments: mammalian evidence

396

hanced proliferation is shown to have occurred around 19–20 Ma ago, between 11–9 Ma ago and between 6–2.5 Ma ago.

References Calvo, J.P., Daams, R., Morales, J. and 23 other authors. 1993. An update of the Spanish continental Neogene synthesis and a paleoclimatic interpretation. Revista de la Sociedad Geologica Espanola, 6 (3–4), 29–40. ChurchWeld, S. 1990. The natural history of shrews. Ithaca, New York, Cornell University Press. Daams, R., Freudenthal, M. & Van der Meulen, A. J. 1988. Ecostratigraphy of micromammal faunas from the Neogene of Spain. In Biostratigraphy and Paleoecology of the Neogene Micromammalian Faunas from the Calatayud-Teruel Basin (Spain). Scripta Geologica, Special Issue 1, Freudenthal, M. (ed.), pp. 287–302. Leiden. Feldhamer, G. A., Klann, R. S., Gerard, A. S. & Driskell, A. C. 1993. Habitat partitioning, body size, and timing of parturition in pygmy shrews and associated soricids. Journal of Mammalogy, 74 (2), 403–11. Reumer, J. W. F. 1984. Ruscinian and early Pleistocene Soricidae (Insectivora, Mammalia) from Tegelen (the Netherlands) and Hungary. Scripta Geologica, 73, 1–173. Reumer, J. W. F. 1985. The paleoecology of Soricidae (Insectivora, Mammalia) and its application to the debate on the Plio-Pleistocene boundary. Revue de Pale´obiologie, 4 (2), 211–14. Reumer, J. W. F. 1989. Speciation and evolution in the Soricidae (Mammalia: Insectivora) in relation with the paleoclimate. Revue suisse de Zoologie, 96 (1), 81–90. Reumer, J. W. F. 1995. The eVect of palaeoclimate on the evolution of the Soricidae (Mammalia, Insectivora). In Paleoclimate and Evolution with Emphasis on Human Origins, Vrba, E. S. et al. (eds.), pp. 135–47. New Haven, Yale University Press. Reumer, J. W. F. 1998. ClassiWcation of the Fossil and Recent Shrews. In Evolution of Shrews, Wo´jcik, J. M. & Wolsan, M. (eds.), pp. 5–22. Bialowieza, Polish Academy of Sciences. Van der Meulen, A.J. & Daams, R. 1992. Evolution of Early-Middle Miocene rodent faunas in relation to long term palaeoenvironmental changes. Palaeogeography, Palaeoclimatology, Palaeoecology, 93, 227–53. Vogel, P. 1980. Metabolic levels and biological strategies in shrews. In Comparative Physiology: Primitive Mammals, Schmidt-Nielsen, K., Bolis, L. & Taylor, C. R. (eds.), pp. 170–80. Cambridge, Cambridge University Press. Vrba, E. S., Denton, G. H., Partridge, T. C. & Burckle, Ll. H (eds.) 1995. Paleoclimate and Evolution with Emphasis on Human Origins. New Haven, Yale University Press.

20 Mammal turnover and global climate change in the late Miocene terrestrial record of the Valle` s-Penede` s Basin (NE Spain) Jorge Agustı´ , Lluı´ s Cabrera, Miguel Garce´ s and Manel Llenas

Introduction The global climatic events detected from 12 to 7 Ma include changes in the atmospheric and oceanic circulation, and increasing latitudinal gradients of temperature, starting with the extensive Antarctic glaciation (Miller et al., 1991; Wright et al., 1992; Keller & Barron, 1983) and the beginning of minor glacial processes in the Arctic regions (Zubakov & Borzenkova, 1990). Late Miocene mammal turnovers have often been regarded as being logically induced by this set of changes in the global climate (Steininger et al., 1985; Barry et al., 1985; Meulen & Daams, 1992). This is especially true in the case of hominid evolution, in which extension of dry environmental conditions over large areas of Eurasia and Africa has been regarded as a major feature (Coppens, 1983; Vrba et al., 1997). However, in several cases this is not based on long terrestrial stratigraphic sections with good chronological control but rather on assumed biochronologic correlations not directly supported by an accurate, reliable stratigraphic framework (with the noticeable exception of the work developed in the the Potwar Plateau by the ‘Siwaliks team’; see Barry et al., 1985; Barry & Flynn, 1990; Morgan et al., 1994; Pilbeam et al., 1996). The recently developed magnetostratigraphic and biostratigraphic framework in the Valle`s-Penede`s Basin (NE Spain) allows, for the Wrst time, a close comparison between the middle–late Miocene mammalian record of Western Mediterranean and Central Asia areas (Figs. 20.1 and 20.2) and its long distance correlation with the global events detected in the oceanic record (Miller et al., 1991; Wright et al., 1992; Flower & Kennett, 1994). This comparison enables one to establish the isochronous or time transgressive pattern of the faunal changes in the context of the evolving late Miocene Earth system.

Geological setting and chronostratigraphic framework The Valle`s-Penede`s Basin has become famous among vertebrate paleontologists because of its high density of mammalian fossiliferous localities, covering most of the Miocene (Crusafont et al., 1955; Agustı´ et al., 1985, 1997). This basin was formed as a consequence of rifting processes and later

Palaeoenvironments: mammalian evidence

398

[Figure 20.1] Main middle to late Miocene Old World paleomastological sequences discussed in this paper.

extensional tectonics related to the NE Spain continental margin evolution (Cabrera et al., 1991; Bartrina et al., 1992; Roca & Guimera`, 1992). The Valle`s-Penede`s half graben is up to 100 km long and 12–14 km wide, and the Valle`s-Penede`s master fault, with a vertical slip larger than 3000 m bounds its NW margin. Activity along this fault took place at least from earliest Miocene to late Miocene (Gallart, 1981; Amigo´, 1986). Minor faults with up to a few hundred meters of slip make up the present southern boundary of the depression, although in most of the southern structural highs of the Garraf-Montnegre Horst, these faults were overlapped by latest Burdigalian–Langhian sediments. The Valle`s-Penede`s Basin was inWlled mainly by thick and continuous early to late Miocene alluvial sequences (Agustı´ et al., 1985; Cabrera et al., 1991). During the middle Miocene, the alluvial fans became fan deltas when they were aVected by late Burdigalian–Langhian and Lower Serravallian marine transgressions. Moreover, beyond the fan delta-inXuenced zones, the marine episodes resulted in a variety of environments (coastal sebkhas, terrigenous bay-shelves, coralgal carbonate platforms and mixed terrigenous carbonate shelf systems). Early Burdigalian to Langhian palynological assemblages (Bessedik & Cabrera, 1985), fossil plant leaf assemblages (Sanz de Siria, 1981) and reptile remains (Crusafont et al., 1955; Cabrera, 1979) suggest that Burdigalian–Langhian paleoclimatic conditions were warm, seasonal and arid tropical–subtropical. This is conWrmed by the widespread occurrence of coral reef complexes (linked to the occur-

Mammal turnover and global climate change

rence of an impoverished mangrove assemblage; Bessedik & Cabrera, 1985), as well as by isotopic paleotemperatures, ranging from 24.6 to 15.5 °C (Va´zquez et al., 1991). However, the marine paleobiota recorded in the later Lower Serravallian carbonate–siliciclastic shelf successions suggest that these deposits were already being deposited under more temperate conditions. Late Miocene sequences in the Valle`s-Penede`s Basin were deposited on proximal to distal-marginal alluvial zones of several alluvial fan systems. This fact resulted in the deposition of the so called Upper Continental Complex (Agustı´ et al., 1985; Cabrera et al., 1991). Proximal conglomerate and breccia dominated alluvial successions which developed along the northern basin margin, grade laterally into sandstone to mudstone dominated sequences. These sequences correspond to middle to distal-marginal alluvial plain and mud Xat facies assemblages and are especially favourable for delivering rich mammalian associations, especially where palustrine conditions developed. The chronostratigraphy of the outcropping Neogene successions of the Valle`s-Penede`s depression has been well established thanks to the many fossil mammal assemblages (Crusafont et al., 1955; Agustı´ et al., 1985) ranging from late Ramblian to early Turolian (i.e. Burdigalian to Tortonian) and Ruscinian (i.e. Pliocene). The marine successions have been attributed to the late Burdigalian, Langhian and Lower Serravallian using the calcareous nannoplankton and the planktonic microforaminifera asssemblages (Magne, 1979; Permanyer et al., 1983; Porta & Civis, 1990; MacPherson, 1992). The highest concentration of mammal localities in the Valle`sPenede`s terrestrial sequence is found around the late Aragonian (late middle Miocene) and Vallesian (late Miocene) stages. The main interest of the late Aragonian–Vallesian succession in the Valle`s-Penede`s is that this sequence records in detail the starting of the environmental changes that led from tropical–subtropical warm forest dominated conditions to the Wnal installation of more open conditions characterizing the latest Miocene in Western Eurasia. From late Aragonian to latest Vallesian–early Turolian a number of punctuated events recorded the faunal turnovers associated with this change.

Holarctic dispersals at the beginning of the late Miocene The oceanic record of the middle to late Miocene transition shows a conspicuous major glacioeustatic lowering at about 11.3 and 11.4 Ma, in the transition between N14 and N15 Blow biozones. According to the eustatic cycle charts (Haq et al., 1987), this event involved a major sea level lowering

399

Chronology of the middle to late Miocene main mammal events in diverse Eurasian regions and NE Africa and correlation with oceanic events. Data sources for Anatolia: Sen (1995); Siwaliks: Barry & Flynn (1990), and Barry et al. (1991); East Africa: Thomas (1979), Thomas & Petter (1986), and Kalb et al. (1982); Oceanic events: Miller et al. (1991), Oslick et al. (1994), and Keller & Barron (1983).

[Figure 20.2]

Mammal turnover and global climate change

401

[Figure 20.3] Quantitative rodent (muroid) succession in the Siwaliks (Potwar Plateau, Pakistan) and Valle`s-Penede`s (NE Spain) sequences. Data from Siwaliks after Barry et al. (1991).

of 140 m (i.e. − 90 m below present sea level). It correlates with the signiWcant isotopic shift Mi5 (middle part of chron C5r, 11.5 Ma) and the oceanic hiatus NH4 (covering the N15 planktonic foraminifera zone, between 10.8 and 11.4 Ma; Keller & Barron, 1983; Berggren et al., 1995). In the Old World terrestrial realm, the main consequence of this event was the quick dispersal of the Wrst grazer equids of the genus Hippotherium (‘Hipparion’) from North America into Eurasia after the probable re-establishment of a landbridge in the Bering area (Woodburne, 1996). Until recently, the isochrony or diachroneity of this dispersal has been the subject of an intensive discussion, but recent paleomagnetic calibrations in the Valle`s-Penede`s Basin established an age of 11.1 Ma for the entry of this equid in Europe (Garce´s et al., 1997; Agustı´ et al., 1997). This European datum indicates that the spread of Hippotherium was a very quick event, since it is coincident with or even older than the ages obtained from East Africa (Kalb et al., 1982; Bernor et al., 1987, 1988) and South Central Asia (Fig. 20.2). Surprisingly, the entry of the hipparionine horses, despite being coeval with major global oceanic and high latitude climatic changes, was a single taxon bioevent and did not involve any signiWcant change in the previously existing mammal association (Agustı´ et al., 1997).

Palaeoenvironments: mammalian evidence

402

Early Vallesian mammal turnover The Wrst signiWcant faunal event in South-Western Eurasia after the entry of Hippotherium is recorded in the Valle`s-Penede`s at 10.4 Ma (i.e. middle part of chron C5N). This was the abrupt decline of the megacricetodontine hamsters within the rodent faunas, coinciding with the spread of the Wrst modern hamsters of the genus Cricetulodon (Garce´s et al., 1996; Agustı´ et al., 1997; Figs. 20.2 and 20.3). Megacricetodon was the dominant cricetid genus in the early and middle Miocene faunas of Europe, with several vicariant species. The record of this genus can be extended from Western Europe to the middle Miocene of the Potwar Plateau (Pakistan; Lindsay, 1988) and to China (Qiu et al., 1981). The megacricetodontines also colonized North America in the late Hemphillian (Bensonomys gileyi from White Cone; Jacobs & Lindsay, 1984). However, in the Potwar Plateau, the megacricetodontines declined abruptly and Wnally became locally extinct at 11 Ma, i.e. much earlier than in Europe (Barry et al., 1991; see Figs. 20.2 and 20.3). In this case, this precocious decline can be attributed to the spread of the murids, which appeared in Southern Asia at about 14 Ma (FAD of Antemus; Jacobs et al., 1984) and became dominant by 12.3 Ma (Barry et al., 1991). This interpretation is suported by the absence of major changes in the remaining faunal elements. In Western Europe, as evidenced in the Valle`s-Penede`s Basin, the decline of the megacricetodontines occurred well before the FAD (‘First Appearence Datum’) of the murids, being replaced in this case by the Wrst ‘modern hamsters’ (Cricetinae) of the genus Cricetulodon (Agustı´ et al., 1997; Fig. 20.3). However, in contrast with Siwaliks, the evidence suggests that this was not a case of competition among rodents, but rather the result of a change in the general ecological conditions. Therefore, the last megacricetodontine faunas in the Valle`s-Penede`s (which were similar to those associated with the FAD of Hippotherium) appear within a mammal assemblage indicating relatively drier conditions, as shown by the absence of humid, forest indicators. On the contrary, the later Cricetulodon bearing paleofaunas in SW Europe include a signiWcant set of wet, forest indicators: castorids, Xying squirrels, glirids, tragulids, tapirids and hominoids. Thus, the replacement of the megacricetodontines by Cricetulodon took place in the frame of a general change to more forested and humid conditions. The spread of Cricetulodon is an event recorded in other Western and Eastern peri-Mediterranean Basins (Calatayud-Daroca and Duero Basins in Spain, Greece and Anatolia; Garcı´a-Moreno, 1988; Unay & Bruijn, 1984). This event is close to the Mi6 isotopic shift recorded within chron C5n at around 10.3 Ma (Miller et al., 1991; Oslick et al., 1994). This isotopic event has been interpreted

Mammal turnover and global climate change

either as a consequence of the ice sheet development in Antarctica or of an oceanic water temperature lowering.

The Mid Vallesian Crisis (MVC) Described for the Wrst time in the Valle`s-Penede`s Basin (Agustı´ & Moya`-Sola`, 1987, 1990), the Mid Vallesian Crisis (MVC) involved the disappearance of most of the middle Miocene elements adapted to the dominant warm and wet-subtropical conditions of that time in Western Europe. This was the case for several rodents, carnivores, suids, ceratomorphs and, of course, primates. Among the large mammals, the crisis aVected especially the carnivores (amphicyonids, nimravids and primitive hyaenids), artiodactyls (the suids Listriodon and Albanohyus) and perissodactyls (tapirids and the rhinoceroses Lartetotherium sansaniense and ‘Dicerorhinus’ steinheimensis). Specially signiWcant was the extinction of the European hominoids of the genus Dryopithecus, very abundantly represented in the early Vallesian levels of the Valle`s-Penede`s (Agustı´ et al., 1996; Moya`-Sola` & Kohler, 1993). Among the rodents, most of the glirids (dormice) which characterized the middle and early late Miocene faunas, such as Miodyromys, Eomuscardinus, Myoglis, Bransatoglis and Paraglirulus, disappeared. For the rest of the Neogene, this family remained restricted to the present genera (basically Eliomys, Muscardinus and Myomimus). The castorids of the genera Chalicomys and Euroxenomys also disappeared, as well as several Xying-squirrels. In the Valle`s-Penede`s Basin, this set of rodent extinctions coincided with the Wrst noticeable spread of the murids of the genus Progonomys, raising again the question of a possible replacement by competition among species. However, the faunal elements which appear more aVected and sensitive to the Vallesian Crisis were again those species adapted to a forest environment, while the more opportunistic elements, such as Cricetulodon, remained unaVected by the entry of Progonomys (Fig. 20.3). The MVC corresponds to a deep, overall restructuring of the Neogene mammalian faunas, coinciding with the noticeable decay of the subtropical warm evergreen forest and the spread in Western Europe of deciduous dominated forests and dry-woodland biomass that characterize the late Miocene all around Western Eurasia (van der Burgh et al., 1993). The fact that this faunal turnover aVected especially those taxa with forest aYnities and the latitudinal character of these set of extinctions (a number of forest adapted taxa survived until the late Vallesian of Central Europe; see Franzen & Storch, 1975, and this volume) strongly suggests a climate forcing for this event. This change was also close to the Mi7 isotopic shift at 9.3–9.6 Ma (chron

403

Palaeoenvironments: mammalian evidence

404

C4Ar.1n; Oslick et al., 1994), which is coincident with the age of 9.7 Ma obtained for the early–late Vallesian boundary in the Valle`s-Penede`s stratigraphic sections, including Can Llobateres (Garce´s et al., 1996; Agustı´ et al., 1997).

Latest Vallesian bioevents During the latest Vallesian times, the trends previously observed in the early–late Vallesian boundary persisted in the Valle`s-Penede`s succession. The early Vallesian high mammal diversity disappears forever, and no recovery is seen in the remainder of the Cenozoic. Although no major extinction events are recorded in this time, the disappearance of middle Miocene taxa continued (Parachleuastochoerus, Miotragocerus), this loss being compensated by the entry of new ‘Turolian’ elements. Among the rodents, the murids diversiWed (Occitanomys hispanicus, Progonomys woelferi). However, the most signiWcant event among the small mammal faunas was not related to a new dispersal or to an extinction event, but to the local evolution of the Cricetulodon lineage which developed a peculiar dental design, the so-called ‘sigmodont pattern’. Rodents displaying a ‘sigmodont’ pattern present a dental design based on a unique sinusoidal ridge and a Xat wear surface. This pattern is present in several extant rodents and is usually associated with a grazing diet (for instance, in the Holarctic arvicolids or the Nearctic ‘cotton rats’). The quick evolution of Cricetulodon from bunodontbrachydont molars into the sigmodont-semihypsodont molars of the genus Rotundomys is a highly signiWcant evolutionary event probably indicating the spread of grasses and a more sclerophylous vegetation. In fact, the transition from the early to the late Vallesian is characterized by the appearence of sigmodont patterns in several cricetid lineages in a widespread region. This is the case of Microtocricetus from several early and late Vallesian localities from Central Europe (Fahlbusch & Mayr, 1975), or Ischymomys from Eastern Europe and Western Asia (Topachevski et al., 1996). This ‘sigmodont event’ (Cricetulodon–Rotundomys transition) is recorded in the Valle`s-Penede`s Basin at around 9.2–9.3 Ma, at the upper part of chron c4Ar. In the Siwaliks, a faunal association including archaic carnivores and rhinoceroses (Brachypotherium), proboscideans (Deinotherium), dormice, shrews and hominoids (Sivapithecus) persisted until chron 4r, at 8.3 Ma. Between 8.3 and 7.8 Ma, a major ecological change associated with a set of extinctions similar to those of the LVC took place (Pilbeam et al., 1996). This faunal turnover led to the Wnal disappearance of several cricetid,

Mammal turnover and global climate change

bovid and tragulid species, while Sivapithecus was replaced by cercopithecid monkeys. For the Wrst time, the Siwaliks fauna became more closely similar to those of Western Eurasia. The delay of more than a million years between Western Europe (LVC of the Valle`s-Penede`s Basin) and Siwaliks (Potwar Plateau) in the onset of late Miocene faunal turnovers is probably explained on the basis of the settling of monsoon atmospheric dynamics in the latter region related to the uplift of the Tibetan Plateau, which could have maintained there the forested subtropical conditions until 8.3 Ma. This faunal change coincides with a shift in the 13C isotopic composition of the paleosoil and dental carbonates, indicating a change from a dominant C3 to dominant C4 (grasses) vegetation, which has been interpreted either as a quick and sharp global change (Quade et al., 1989) or as a more complex and long lasting process (see Morgan et al., 1994, and below).

Late Miocene mammal changes in Southern Eurasia and worldwide environmental change The comparison of the Western European and SW Central Asian middle to late Miocene paleomastological records shows that the patterns of appearance, spreading and dominance of the diverse fossil mammal taxa as well as the changes of the onland ecological structure in the Southern Eurasian regions were sometimes highly time transgressive (Figs. 20.2 and 20.3). Similar paleofaunal changes to those observed in the Western Mediterranean area took place at a diVerent time, probably because of diverse regional features, in the SW Asian region. While some changes are recorded at an earlier time in this region (decline of megacricetodontines, decline of middle Miocene suoids), others (disappearance of the subtropical, warm forest fauna) were delayed until the middle Turolian (or beginning of the Messinian). The decline of the megacricetodontine hamsters and their precocious middle Miocene local extinction in that area seems more related to ecological competition with the expanding murids, than to a general environmental change. The equivalent to the Late Vallesian Crisis in the Siwaliks is reXected in a coeval signiWcant decrease in the tragulid representation and its replacement by bovids, but again in this case no general faunal restructuring is recognized in the Potwar Plateau area (as it was the case of the LVC in the Western Mediterranean area). Finally, the most important trend of changes in the Siwaliks took place between 8.3 and 7.1 Ma, coincident with the appearance and development of the most typical Turolian paleofaunal assemblages in the peri-Mediterranean regions.

405

Palaeoenvironments: mammalian evidence

406

Cerling et al. (1997) have provided impressive isotopic evidence for a global vegetational change through the late Miocene to early Pliocene times. The period between the late Miocene and Pliocene is identiWed as one involving not only a signiWcant change in diet, but also a period of worldwide faunal change, thus demonstrating the signiWcant eVects of the vegetational change upon the composition of the land vertebrate faunas. It is assumed by Cerling et al. (1997) that this faunal turnover has been also the consequence of the vegetational change involved by the transition from a C3 dominant diet to a C4 dominant one and that an important global ecological change aVecting large mammals was underway at that time. Several examples are given of transition from a woodland adapted fauna to more open-habitat assemblages: Pakistan, North America, East Africa. However, as an exception, the authors are not able to Wnd such a transition from C3 to C4 dominant vegetation in any sample from Spain, France and the Eastern Mediterranean. The absence of a C3–C4 transition in any sample of any time in the Mediterranean region is explained by Cerling et al. (1997) assuming a latitudinal eVect: since the threshold for C3 photosynthesis is higher at warmer temperatures, the change occurred earlier at lower latitudes. But the question is not that the change occurred later in higher latitudes, but that, in fact, it never occurred in Western Europe (or, better, the Mediterranean region). We Wnd here a signiWcant contradiction which cannot be explained by assuming a simple latitudinal eVect. In fact, as shown, most of the mammalian changes described for Pakistan and North America (that is, the transition from a woodland-forest fauna to an open-land one) occurred in Spain at 9.6 Ma during the Late Vallesian Crisis, well before the 8–7 Ma age given for the transition from a C3 to C4 vegetation. Moreover, no special change in the mammal fauna is detected in Spain in the period between 8 and 7 Ma. Only during the late Turolian did a signiWcant faunal change occur, associated to the Messinian crisis in the Mediterranean. Some of the new elements that had entered Spain by that time came from Asia and Africa, being already adapted, therefore, to a C4 vegetation. We have to assume, according to Cerling et al. (1997), that they re-adapted to a C3 regime after they settled on the Iberian peninsula. Therefore, the association of this vegetational change to the faunal turnover as recorded, for instance, in Pakistan, seems not to be so conclusive as suggested by Cerling et al. (1997). Instead, mammal turnovers such as those described by these authors took place before and after that event, independently of that vegetational change.

Mammal turnover and global climate change

Concluding remarks: the role of regional physiographic conditions Western European and SW Central Asian middle to late Miocene paleofaunal records show that dispersal, diversiWcation and declining of mammal taxa were sometimes highly time transgressive; for instance the westward spread of murids (Garce´s et al., 1996; Agustı´ et al., 1997) and the decline of the forest fauna (Figs. 20.1 and 20.2). The diachroneity of the faunal changes recorded in middle to low latitude Old World regions suggests that the climatically forced changes, although signiWcant, were delayed and modiWed in some regions by other features (i.e. short term physical disturbances, biological interactions and paleogeographic features enhancing dispersal or triggering extinction). The understanding of Old World middle to late Miocene bioevents is positioned in the frame of deep changes resulting from the collision of the African and Indian Plates with the southern margin of Eurasia (Biju Duval et al., 1977). The northward drift of Africa and diverse microplates in the Mediterranean regions caused important modiWcations in the Tethyan region, leading to the progressive isolation of the Mediterranean sea from the Indian and Atlantic oceans. Closure of the eastern Mediterranean gateway in late Serravallian–early Tortonian prevented longitudinal oceanic circulation and heat transport from the Indian ocean towards the western Mediterranean. The uplift of the Himalaya and Tibetan Plateau (Amano & Taira, 1992; Rea, 1993; Coleman & Hodges, 1995), as well as several other major to medium ranges in the Mediterranean region (Dercourt et al., 1993), caused important changes in surface orography. The appearance of high mountains and plateaus at mid-latitudes likely aVected the atmospheric circulation over very extensive regions of the northern hemisphere. Modeling the inXuence of topography on the atmospheric circulation indicates that late Miocene Earth’s climate in the Old World could be strongly dependent on the development of these elevations (Kutzbach et al., 1993). The modeling of the raising of the Tibetan Plateau predicts: 1) the enhancement of summer drying, desertiWcation and reduction of wet-adapted vegetation in West Asia and the peri-Mediterranean regions; 2) cooling and disappearance of subtropical thermophilous vegetation in other neighbouring extensive Eurasian regions. The climatic changes predicted by this elegant model coincide with overall evolution of the middle to late Miocene mammalian faunas in the peri-Mediterranean regions, North-Central Europe and Northern Africa.

407

Palaeoenvironments: mammalian evidence

408

They provide a plausible, at least partial explanation for the recorded heterochronous late Miocene mammal changes in extensive Asian regions. The delay of more than a million years in the beginning of the faunal change between Western Europe (MVC of the Valle`s-Penede`s Basin) and Siwaliks (Potwar Plateau) is probably explained on the basis of the relatively later setting in of a remarkably dry–wet, strongly seasonal climatic regime in that area, related to the reinforcement of the monsoonal circulation in Southern Asia between 8.8 and 8.3 Ma, linked to the uplift of the Tibetan Plateau (Prell et al., 1992). This fact suggests that the global climatic changes which initially should induce the mammal turnovers were strongly modiWed by regional features. As a conclusion it can be established that the time transgressive pattern of these mammalian changes was highly dependent upon the regional location and paleoclimatic conditions in each area (i.e. degree of continental paleogeography, importance of monsoonal circulation, dry– wet seasonality, temperature gradient).

Acknowledgments Funded by the Spanish Ministry of Education and Science (research projects PB94-1265 and GEO89-0831) and the Comissionat per Universitats i Recerca de la Generalitat de Catalunya, Grup de Qualitat GRQ94-104.

References Agustı´, J. & Moya`-Sola`, S. 1987. Mammal extinctions in the Vallesian (upper Miocene). In Int. Conf. Paleont. & Evol.: Extinction Events, Lamolda, M. (ed.). Leioa. Agustı´, J. & Moya`-Sola`, S. 1990. Mammal extinctions in the Vallesian (Upper Miocene). In Extinction Events in Earth History, KauVman, E.G. & Walliser, O.H. (eds.), pp. 425–32. Agustı´, J., Cabrera, Ll. & Moya`-Sola`, S. 1985. Sinopsis estratigra´Wca del Neo´geno de la fosa del Valle`s-Penede`s. Paleontologia i Evolucio´, 18, 57–81. Agustı´, J., Cabrera, L., Moya`-Sola`, S., Ko ¨ hler, M., Garce´s, M. & Pare´s, J. M. 1996. Can Llobateres: The pattern and timing of the vallesian hominoid radiation reconsidered. Journal of Human Evolution, 31, 143–55. Agustı´, J., Cabrera, L., Garce´s, M. & Pare´s, J. M. 1997. The Vallesian mammal succession in the Valle`s-Penede`s basin (NE Spain): Paleomagnetic calibration and correlation with global events. Palaeogeography, Palaeoclimatology, Palaeoecology, 133, (3–4). Amano, K. & Taira, A. 1992. Two-phase uplift of Higher Himalaya since 17 Ma. Geology, 20, 391–4. Amigo´, J. 1986. Estructura del massı´s del Gaia`. Relacions estructurals amb les fosses del Penede`s i del Camp de Tarragona, Ph.D. Thesis, Univ. Barcelona, pp. 253.

Mammal turnover and global climate change

Barry, J. C. & Flynn, L. J. 1990. Key biostratigraphic events in the Siwalik sequence. In European Neogene Mammal Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 557–72. Barry, J. C., Johnson, N. M., Mahmood-Raza, S. & Jacobs, L. L. 1985. Neogene mammalian faunal change in southern Asia: Correlations with climatic, tectonic, and eustatic events. Geology, 13, 637–40. Barry, J. C., Morgan, M. E., Winkler, A. J., Flynn, L. J., Lindsay, E. H., Jacobs, L. L. & Pilbeam, D. 1991. Faunal interchange and Miocene terrestrial vertebrates of southern Asia. Paleobiology, 17, 231–45. Bartrina, M. T., Cabrera, L., Jurado, M. J., Guimera`, J. & Roca, E. 1992. Evolution of the central Catalan margin of the Valencia Trough. Tectonophysics, 203, 1–4, 219–47. Berggren, R. H., Kent, D. V., Swisher, C. C. I. & Aubry, M. P. 1995. A revised Cenozoic geochronology and chronostratigraphy. In Geochronology, Time Scales and Global Stratigraphic Correlations: A UniWed Temporal Framework for an Historical Geology: SEPM Spec. Publ., Berggren, W. A., Kent, D. V. & Hardenbol, J. (eds.), pp. 129–212. Bernor, R. L., Brunet, M., Ginsburg, L. et al. 1987. A consideration of some major topics concerning Old World Miocene mammalian chronology, migrations and paleogeography. Geobios, 20, 431–9. Bernor, R. L., Kovar-Eder, J., Lipscomb, D., Ro¨gl, F., Sen, S. & Tobien, H. 1988. Systematic, stratigraphic, and paleoenvironmental contexts of Wrst-appearing Hipparion in the Vienna basin, Austria. Journal of Vertebrate Paleontology, 8, 427–52. Bessedik, M. & Cabrera, Ll. 1985. Le couple re´cif-mangrove a` Sant Pau d’Ordal (Valle`s-Penede`s), te´moin du maximum transgressive en Me´diterrane´e nord occidentale (Burdigalien supe´rieur-Langhien inferieur). Newsl. Stratigr., 14, 20–35. Biju Duval, B. & Montadert, L. 1977. Introduction to the structural history of the Mediterranean basins. In Structural History of Mediterranean Basins, Biju Duval, L. & Montadert, L. (eds.), pp. 1–12. Technip, Paris. Cabrera, Ll. 1979. Estudio estratigra´Wco y sedimentolo´gico de los depo´sitos contenentales basales de la depresio´n del Valle`s-Penede`s. Masters Degree. Univ. of Barcelona, pp. 361. Cabrera, L., Calvert, F., Guimera`, J. & Permanyer, A. 1991. El registro sedimentario mioce´nico en los semigrabens del Valle`s-Penede`s y de El Camp: Organizacio ´n secuencial y relaciones tecto´nica sedimentacio´n. In Field Guide Book no 4 of the I Congreso del Grupo Espan ˜ ol del Terciario, Vic, Colombo, F. (ed.), pp. 132. Cerling, T. E., Harris, J. R., MacFadden, B. J., Leakey, M. G., Quade, J., Eisenmann, V. & Ehleringer, J. R. 1997. Global vegetation change through the Miocene/Pliocene boundary. Nature, 389, 153–8. Coleman, M. & Hodges, K. 1995. Evidence for the Tibetan plateau uplift before 14 Myr ago from new minimum age for east-west extension. Nature, 374, 49–52. Coppens, Y. 1983. Le Singe, l’Afrique et l’homme, Fayard (ed.), p. 148. Paris. Crusafont, M., Villalta, J. E. de & Truyols, J. 1955. El Burdigaliense continental de la cuenca del Valle`s-Penede`s. Mem. y Com. Inst. Geol. Dip. Prov. de Barcelona, 12, 260. Dercourt, J., Ricou, L. E. & Vrielynk, B. (eds.) 1993. Atlas Tethys Paleoenvironmental Maps. Gauthier Villars, pp. 307.

409

Palaeoenvironments: mammalian evidence

410

Fahlbusch, V. & Mayr, H. 1975. Microtoide Cricetiden (Mammalia Rodentia) aus der Oberen Su¨sswasser-Molasse Bayerns. Pala`ont. Z., 49 (1–2), 78–93. Flower, B. P. & Kennet, J. P. 1994. The middle Miocene climatic transition: east Antarctic ice sheet development, deep ocean circulation and global carbon cycling. Palaeogeography, Palaeoclimatology, Palaeoecology, 108, 537–55. Franzen, J. L. & Storch, G. 1975. Die Unterplio¨zane Wirbeltierfauna von Do ¨ rn-Durkheim, Rheinhessen (SW Deutschland). 1: Carnivora, Proboscidea, Rodentia. Senckenbergiana Lethaea, 56, 233–303. Gallart, F. 1981. Neo´geno superior y Cuaternario del Penede`s (Catalunya, Espan ˜ a). Acta Geol. Hisp., 16, 151–7. Garce´s, M., Agustı´, J., Cabrera, L. & Pare´s, J. M. 1996. Magnetostratigraphy of the Vallesian (late Miocene) in the Valle`s-Penede`s Basin (northeast Spain). Earth and Planetary Science Letters, 142, 381–96. Garce´s, M., Cabrera, L., Agustı´, J. & Pare´s, J. M. 1997. Old World Wrst appearance datum of ‘Hipparion’ horses: Late Miocene large-mammal dispersal and global events. Geology, 25, 19–22. Garcı´a-Moreno, E. G. 1988. The Miocene rodent biostratigraphy of the Duero Basin (Spain); a proposition for a new Aragonian/Vallesian limit. Paleontologia y Evolucion, 22, 103–12. Haq, B. U., Hardenbol, J. & Vail, P. R. 1987. Chronology of Xuctuating sea levels since the Triassic. Science, 235, 1156–67. Jacobs, L. L. & Lindsay, E. H. 1984. Holarctic radiation of Neogene Muroid rodents and the origin of South American cricetids. Journal of Vertebrate Paleontology, 4, 265–72. Kalb, J. E., Jolly, C. J., Tebedge, S. et al. 1982. Vertebrate faunas from Awash group, middle Awash valley, Afar, Ethiopia. Journal of Vertebrate Paleontology, 2, 237–58. Keller, G. & Barron, J. A. 1983. Paleoceanographic implications of Miocene deep-sea hiatuses. Geological Society of America Bulletin, 94, 590–613. Kutzbach, J. E., Prell, L. & Ruddiman, W. F. 1993. Sensitivity of Eurasian climate to surface uplift of Tibetan Plateau. The Journal of Geology, 101, 177–90. Lindsay, E. H. 1988. Cricetid rodents from Siwaliks deposits near Chinji village. Part 1: Megacricetodontinae, Myocricetodontinae and Dendromuridae. Palaeovertebrata, 18, 94–154. MacPherson, I. 1992. Paleoecologı´a de los foraminı´feros en el Mioceno Medio de la Cuenca del Penede`s. Tesis doctoral Universidad de Barcelona. Deptos. de Ecologı´a y Geologı´a Dina´mica, Geofı´sica y Paleontologı´a. pp. 434. Magne´, J. 1979. Etudes microstratigraphiques sur le Ne´oge`ne de la Mediterrane´e nordoccidentale. Vol. I. Les Basins Ne´oge`nes Catalans. Univ. Paul Sabatier. Toulouse, pp. 260. Meulen, A. van der & Daams, R. 1992. Evolution of Early–Middle Miocene rodent faunas in relation to long-term palaeoenvironmental changes. Palaeogeography, Palaeoclimtology, Palaeoecology, 93, 227–53. Miller, K. G., Wright, J. D. & Fairbanks, R. G. 1991. Unlocking the icehouse: Oligocene–Miocene oxygen isotopes, eustary and margin erosion. J. Geophys Res, 6829–48. Morgan, M. E., Kingston, J. D. & Marino, B. D. 1994. Carbon isotopic evidence for the

Mammal turnover and global climate change

emergence of C4 plants in the Neogene from Pakistan and Kenya. Nature, 367, 162–5. Moya`-Sola`, S. & Kohler, M. 1993. Recent discoveries of Dryopithecus shed new light on evolution of great apes. Nature, 365, 543–5. Oslick, J., Miller, K. G., Feigenson, M. D. & Wright, J. D. 1994. Oligocene–Miocene strontium isotopes: Stratigraphic revisions and correlations to an inferred glacioeustatic record. Paleoceanography, 9, 427–43. Permanyer, A., Salvatorini, G. & Mazzei, R. 1983. New precision about marine Miocene age of the Penede`s basin (Catalonia, NE Spain). Acta Geol. Hisp., 14, 293–310. Pilbeam, D., Barry, J., Ibrahim, S. M., Pickford, M. H. L., Bishop, W. W., Thomas, H. & Jacobs, L. L. 1977. Geology and palaeontology of Neogene strata of Pakistan. Nature, 270. Pilbeam, D., Morgan, M., Barry, J. C. & Flynn, L. J. 1996. European MN Units and the Siwalik faunal sequence of Pakistan. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H. W. (eds.). Columbia University Press. Porta, J. de & Civis, J. 1990. Events and correlation in the Neogene of Prelitoral Catalonian depression. In The Valencia Trough: Geology and Geophysics. Terra Abstracts, 2 (2), 116–17. Prell, W. L., Murray, D. W. & Clemens, S. C. 1992. Evolution and variability of the Indian ocean summer monsoon: Evidence from the western Arabian sea drilling program. In Synthesis of Results from ScientiWc Drilling in the Indian Ocean, Duncan, R. A., Rea, D. K., Kidd, R. B., von Rad, U. & Weissel, J. K. (eds.), pp. 447–69. Washington, American Geophysical Union. Qiu, Z., Li, C. & Wang, S. 1981. Miocene mammalian fossils from Xining basin. Vertebreta Paliasiatica, 19, 156–73. Quade, J., Cerling, T. E. & Bowman, J. R. 1989. Development of Asian monsoon revealed by marked ecological shift during the latest Miocene in northern Pakistan. Nature, 342, 163–4. Rea, D. K. 1993. Geologic records in Deep Sea Muds. GSA Today, 3. Roca, E. & Guimera`, J. 1992. The Neogene structuration of the Eastern Iberian Margin: structural constraints to the crustal evolution of the Valencia Trough (Western Mediterranean). Tectonophysics, 203, 203–18. Sanz de Siria, A. 1981. La Xora Burdigaliense de los alrededores de Martorell (Barcelona). Paleont. i Evol., Sabadell, 16, 3–13. Sen, S. 1995. Magnetostratigraphie des formations continentales. Ge´ochronique, 56, 12. Steininger, F. F., Rabeder, G. & Ro¨gl, F. 1985. Land mammal distribution in the Mediterranean Neogene. A consequence of geokinematic and climatic events. In Geological Evolution of the Mediterranean Basin, Stanley, D. J. & Wezel, F. C. (eds.), pp. 559–71. Springer Verlag. Thomas, H. 1979. Miotragocerus cyrenaicus sp. nov. (Bovidae, Artiodactila, Mammalia) du Mioce`ne superior de Sahabi (Libye) et ses rapports avec autres Miotragocerus. Geobios, 12, 267–80. Thomas, H. & Petter, G. 1986. Re´vision de la faune de Mammife`res du Mioce`ne de Menacer (ex-Marceau), Alge´rie: discussion sur l’age du gisement. Geobios, 19, 357–73.

411

Palaeoenvironments: mammalian evidence

412

Topachevsky, V. A., Nesin, V. A., Topachevsky, I. V. & Semenov, Yu. A. 1996. The most ancient locality of the Middle Sarmatian small mammals fauna (Insectivora, Lagomorpha, Rodentia) in the East Europe. Dopovidi NAN Ukraine, N2, 107–10. Unay, E. & De Bruijn, H. 1984. On some Neogene rodent assemblages from both sides of the Dardanelles, Turkey. Newsl Strat, 13, 119–32. Van der Burgh, J., Vesscher, H., Dilcher, D. L. & Ku ¨ rschner, W. M. 1993. Paleoatmospheric signatures in Neogene Fossil Leaves. Science, 260, 1788–90. Va´zquez, A., Zamarren ˜ o, de Porta, J. & Plana, F. 1991. La composicio´n isoto´pica y los elementos traza de Amussiopecten baranensis (Pectinidae) como indicadores paleoambientales, en el Langhiense catala´n. Rev. Soc. Geol. Espan ˜ a, 4 (3–4), 215–27. Vrba, E., Denton, G. H., Partridge, T. C. & Burckle, L. H. (eds.) 1997. Paleoclimate Evolution, with Emphasis on Human Origins. Yale University Press. Woodburne, M. O. 1996. Precision and resolution in mammalian chronostratigraphy: Principles, practices, examples. Journal of Vertebrate Paleontology, 16, 531–55. Wright, J. D., Miller, K. G. & Fairbanks, R. G. 1992. Early and middle Miocene stable isotopes; implications for deepwater circulation and climate. Paleoceanography, 7, 357–89. Zubakov, V. A. & Borzenkova, I. I. (eds.) 1990. Global Paleoclimate of the Late Cenozoic. Elsevier, p. 456.

21 Palaeoenvironments of late Miocene primate localities in Macedonia, Greece Louis de Bonis, Genevieve Bouvrain and George D. Koufos

The late Miocene deposits of northern Greece have been known since the beginning of the twentieth century (Andrews, 1918; Arambourg & Piveteau, 1929). They have yielded several fossil mammalian localities which have been dated to Vallesian and Turolian. Arambourg & Piveteau (1929) published faunal lists and fossil description of several localities found by Arambourg during the First World War. The Arambourg collection is housed in the Museum national d’Histoire naturelle, Paris, and it is labelled ‘Salonique’ in most of the following publications, but the fossils have been recovered from several sites and it is diYcult to know exactly where they have come from. The age of the faunas was given as ‘Pontian’. Andrews’s material is in the Natural History Museum, London, and was found in an unknown locality near the village of Diavata (Koufos, 1985). Field researches were carried out again in the beginning of the 1970s. They allowed the discovery of numerous new fossiliferous sites and they lead to the conclusion that the ages of the faunas, although late Miocene, were quite diVerent from each other. Indeed the oldest ones have been dated to Vallesian (10 to 9 Ma) and the more recent ones to late Turolian (6 Ma). These localities have a great importance insofar as they have, in a quite small area, hominoid primates and cercopithecoids. Hominoids are recovered from Vallesian sites and the cercopithecoids from the Turolian ones. The locality of Pyrgos near Athens has a faunal list which sounds Turolian although this locality has yielded a mandible of hominoid. But the fossils were identiWed from poor specimens and these identiWcations need to be reappraised (see Bonis & Koufos, this volume). Hominoids and cercopithecoids are never found together into the same locality (Bonis et al., 1987 ). This replacement of a superfamily of primates by another one is linked with great changes in the mammalian fauna and probably with changes of the palaeoenvironment. Studies of recent mammalian faunas lead to the conclusion that the composition of a fauna depends on the ecology and on the environment. Actually it is relatively easy to recognize a forest fauna from an open environment fauna by several methods (Andrews et al., 1979; Legendre, 1986; Bonis et al., 1992; Andrews, 1995). There are several methods of studying a fauna in this respect. We have chosen to use the taxonomic composition to compare the faunas, but if the taxonomy is used at the species or even the genus level, it will show up geographical or stratigraphical variations more than ecological diVerences because two similar envi-

Palaeoenvironments: mammalian evidence

414

ronments could have diVerent speciWc or generic taxa. For example, a Malaysian tropical rain forest diVers greatly from an African one if we look at the species level. However, both share several similarities if we consider the number of species in higher taxonomic units (Bonis et al., 1992), and so that is the way we proceeded to compare the late Miocene mammalian faunas from northern Greece with each other and with some European or Asiatic faunas. Andrews et al. (1979) have also found that taxonomic comparisons by order produced statistically signiWcant diVerences between faunas. Alternatively it would be possible to choose ecological or functional groups (Andrews et al., 1979; Andrews, 1995). Insofar as there are very few micromammals in the fossiliferous sites of northern Greece, due to a taphonomic bias, we only took into account the large mammals. Beyond that, among large mammals, the carnivorans were excluded for two reasons. First, many carnivorans are ubiquitous animals. Nowadays tigers live from southern Asia to the polar circle, until recently lions inhabited Europe as well as south Africa, and jaguars are known from sea level to the higher Andes plateau. So, although a complete guild of carnivoran mammals can give good indications on the environment, it is not the case if we have only some of them. Second, carnivoran remains are more numerous and far more diversiWed in some fossiliferous sites, for example those that correspond to ancient swamps or small lakes as is the case for the Miocene localities of Paulhiac (MN 1), Laugnac (MN 2), Sansan (MN 6) or Rudabanya (MN 9). It could be the same for some Wssure Wllings where the carnivorans can be over represented. But carnivorans are generally far less frequent in river deposits such as the late Miocene localities of northern Greece, and it would be impossible to get a complete guild of carnivorans from such small samples. In the following we shall try to show that the faunal compositions of the Greek and neighbouring (Greco-Iranian faunal province) late Miocene mammalian localities are clearly diVerent from those of the rest of Europe and that there was a signiWcant faunal change between Vallesian and Turolian layers under the diVerent conditions pertaining to those regions. The Vallesian localities of northern Greece are Ravin de la Pluie (RPL), Ravin des Zouaves no. 1 (RZ 1), Xirochori 1 (XIR), Pentalophos (PNT) and Nikiti 1 (NKT). Three of them (RPL, XIR and NKT) have yielded remains of the hominoid primate Ouranopithecus. The Turolian ones are Ravin des Zouaves no. 5 (RZO), the Vathylakkos group (VAT, VTK and VLO), Prochoma (PXM), the Dytiko group (DTK, DIT and DKO) (Bonis et al., 1992) and Maramena (MARA) (Schmidt-Kittler et al., 1995). Most of them have yielded remains of the cercopithecoid Mesopithecus.

Late Miocene primate localities

Chalicoteriidae as indicators of palaeoecology Some large mammals could be by themselves indicators of a quite precise palaeoenvironment. Two subfamilies of Chalicoteriidae are present in the Eurasian late Miocene, the Chalicotheriinae with the genera Macrotherium and Chalicotherium, and the Schizotheriinae with Ancylotherium. They have not been found together in the same locality except in the old collection of Pikermi, but it is possible that the fossils of the old Pikermi collections come from more than one locality. A look at the faunal lists of several localities (Table 21.1) shows that the chalicotherines are generally associated with several suids, tragulids, several cervids and tapirids. On the other hand, the schizotherines are never associated with tragulids or tapirids but with many bovids or giraYds. The equids and elephantids have not been taken into account insofar as they occur in every locality. Some other groups (Hyracoidea or Tubulidentata) have also been excluded because they are too scarce and uncommon in the studied localities. Two histograms (Fig. 21.1) illustrating the percentage of localities with diVerent kinds of mammals associated with one or the other subfamily of chalicotherids are clearly diVerent. So it seems likely that schizotherines lived in diVerent habitats than chalicotherines. The tragulids and tapirids are principally forest dwellers today and are probably indicators of quite closed environments. They have never been found with a schizothere. The latter could be associated with few cervids but always with many bovids and giraYds, and until now a schizotherine has never been found with a hominoid. For example, Ancylotherium is present in Pentalophos which is a Macedonian Vallesian locality without hominoids. If, until now, the presence of a schizotherine is exclusive of some taxa (tragulids, tapirids or hominoids), schizotherines or chalicotherines could be associated to the other ones. But the diVerences are in the number of localities in which one of these taxa is present in a geographic area. The percentage of localities indicates a trend of both sub-families to dwell in open or more forested environment but one species or another one cannot indicate Wrmly by themselves the type of environment they were dwelling in.

The faunal changes between the Vallesian and Turolian Everywhere, the transition between the Vallesian and Turolian is marked by changes in the faunas. In this article we did not take into account the mid-Vallesian crisis because we have few Vallesian localities in Greece and we cannot see if this crisis is present or not. The diVerences between the two

415

Palaeoenvironments: mammalian evidence

Table 21.1. Percentage of association of some taxa with Chalicotheriinae and Schizotheriinae in some late Miocene European mammalian faunas

71%

11

36%

9%

0%

46%

[Figure 21.1] Histograms summarizing the differences of mammalian assemblages between late Miocene localities with Chalicotheriinae or Schizotheriinae.

9

With Tapir

58%

8 With Hominoids

50%

7 + Than 1 Rhinoth.

54%

6

With Giraffes

24

5

+ Than 1 Bovid

With Cervids

4

With Tragulia

3

+ Than 1 Suid

Localities with Chalicotheres Localities with Schizotheres

2

With Dinotheres

1

n

416

38%

54% 67%

33%

53%

100% 100% 55%

0%

0%

Late Miocene primate localities

Table 21.2. Percentage of localities of the late Miocene Greco-Iranian faunal province bearing some taxa of high taxonomical categories during the Vallesian and Turolian ages of mammals respectively Percentage of localities

With Deinotheres More than 1 Suid With Tragulid More than 1 Cervid More than 1 Bovid With more than 1 Giraffe Small primates Large primates More than 1 Rhino With Tapir More than 1 equid With Chalicothere With Schizothere

MN 9–MN 10

MN 11–MN 13

20 40 0 0 100 100 0 60 60 0 80 20 20

20 13 33 0 100 53 53 0 33 0 100 20 27

groups of localities, Vallesian and Turolian, have been noted in three diVerent regions. The Wrst one is called the Greco-Iranian province or the sub-Paratethyan province (Bernor, 1978; Bonis et al., 1979) and it extends from the former Yugoslavia (Titov Veles) to Iran and maybe Afghanistan (Bonis et al., 1993). In this province (Table 21.2), the diVerences between the two sets are summarized by the following characteristics from the Vallesian to the Turolian: same low percentge of localities with deinotheres, less with suids and giraYds, presence of some tragulids, replacement of large primates (hominoids) by small primates (cercopithecoids). We do not see any signiWcant diVerence in the percentages of localities with perissodactyls. Rhinocerotids, tapirids, chalicotherids or equids are similarly represented in Vallesian and Turolian sites. The second region is central Europe, on the shores of the Paratethys sea (Table 21.3). Here the pattern of change is quite diVerent. The percentage of localities with deinotheres remains almost the same but the number of localities with more than one suid falls to zero, as is the case for tragulids, tapirids and large primates; likewise, the number of localities with more than one cervid decreases also. In contrast there is an increasing number of localities with more than one bovid, one giraYd or one equid while at the same time there was appearance of the schizotheres. The Vallesian primates of central Europe are large (Dryopithecus) or small (Pliopithecidae). They

417

Palaeoenvironments: mammalian evidence

418

Table 21.3. Percentage of localities of the late Miocene central European faunal province bearing some taxa of high taxonomical categories during the Vallesian and Turolian ages of mammals respectively Percentage of localities MN 9–MN 10 With Deinotheres More than 1 Suid With Tragulid More than 1 Cervid More than 1 Bovid With more than 1 Giraffe Small primates Large primates More than 1 Rhino With Tapir More than 1 equid With Chalicothere With Schizothere

67 50 17 33 33 17 17 17 83 33 17 17 0

MN 11–MN 13 67 0 0 17 83 58 33 0 83 0 66 8 17

are replaced in the Turolian by small primates (Mesopithecus). The general trend could correspond to a drier (and/or cooler?) climate with more localities having mammals which generally inhabit open environments. In western Europe, Spain, France or Germany, the pattern is diVerent again (Table 21.4). The number of localities with deinotheres is lower in the Turolian than in the Vallesian, and the number of sites with more than one suid, more than one cervid, one giraYd, one rhinocerotid or with tragulids, tapirids and chalicotheres show a small decrease or it remains almost the same. There are more localities with more than one bovid and one equid. There were no schizotheres either in the Vallesian or in the Turolian. Like central Europe the palaeoenvironment of western Europe became more open during the Turolian but probably with more remains of forested area than in central or southeastern Europe. A general comparison of the three regions (Fig. 21.2) shows that changes between the Vallesian and Turolian are more signiWcant in central or western Europe than in the eastern Mediterranean, where the more marked change is on the primates. Marked diVerences exist in western Europe on deinotheres, suids, tragulids, bovids, primates and chalicothines while in central Europe they exist on the bovids, giraYds and the number of equids. It is possible to consider these changes diVerently if distinct steps are considered in the Vallesian (Fortelius et al., 1996). The dramatic ‘crisis of the middle Vallesian’ corresponds to ‘. . . abrupt change from closed to open

Late Miocene primate localities

Table 21.4. Percentage of localities of the late Miocene western European faunal province bearing some taxa of high taxonomical categories during the Vallesian and Turolian ages of mammals respectively Percentage of localities MN 9–MN 10 With Deinotheres More than 1 Suid With Tragulid More than 1 Cervid More than 1 Bovid With more than 1 Giraffe Small primates Large primates More than 1 Rhino With Tapir More than 1 equid With Chalicothere With Schizothere

80 70 60 70 20 10 20 50 90 50 20 80 0

MN 11–MN 13 40 17 25 67 58 0 8 0 25 3 58 25 0

environments’ in Western Europe while ‘. . . its central changed much less’ and in any case ‘open woodland mammal faunas of East . . . transgressed in the newly altered parts of West during the Turolian’ and ‘. . . with the opening of its habitats West (except Central Europe) became more like East in many respects . . .’.

Multivariate (factor) analysis Multivariate analysis of the faunas can also be used to obtain such results. The variates used are the number of species in some high taxonomic categories. Most of the high taxonomic categories correspond more or less to adaptative zones in which every member of the category occupies a peculiar ecological niche. So the proportion of the species of diVerent categories in diVerent faunas could indicate the general ecological conditions in which the faunas are dwelling. The matrix used for the analysis consists of a table in which each line represents a local fauna (Table 21.5) and each column a taxonomic category (Table 21.6). The cells of the table contain the number of species corresponding to a category in a locality. The method used is Correspondence Factor Analysis (EscoWer & Pages, 1990; Lebart et al., 1984 ). First it is used for the recent faunas of which ecological conditions are well known, then for the fossil faunas.

419

Palaeoenvironments: mammalian evidence

420

[Figure 21.2] Diagram summarizing the differences of mammalian assemblages during the late Miocene (Vallesian and Turolian) in western Europe, central Europe and Greco-Iranian faunal province.

Late Miocene primate localities

Table 21.5. List of the recent local faunas used in the analysis of Fig. 21.3 Sites

Countries

Environments

References Geerling & Bokdam, 1973; Lartiges & Poilecot, 1997 Poulet, 1972

Comoe´

COMO

Ivory Coast

Wooded savanna, flood plain, forests

North Ferlo Umfolozi

NFER UMFO

Se´ne´gal South Africa

Hluhlueve Jebel Marra Chobe Lamto savanna Lamto forest Taı¨

HLUH MARR CHOB LAMS LAMF TAIZ

South Africa Sudan Botswana Ivory Coast Ivory Coast Ivory Coast

Kafue

KAFU

Zambia

Bushed savanna, open savanna Savanna with acacias, gallery forests Decidual humid forest Montane savanna Flood plain Dry savanna Montane forest Evergreen tropical forest Flood plain, grass, bushes, forests

Makokou Irangi Mount Kivu

MAKO IRAN KIVU

Gabon Congo Zaı¨re

Equatorial forest Equatorial forest Montane forest, savanna and marsh

La Maboke´ Tarangire Zinave Transvaal 1 Transvaal 2 Transvaal 4 Transvaal 5 Transvaal 6 Transvaal 8 Transvaal 9 Baoule´ W. National Park Pendjari Sangmelina Willpattu

LAMA TARA ZINA TRA1 TRA2 TRA4 TRA5 TRA6 TRA8 TRA9 BAOU WNPK PEND SANG WILP

Central African Rep. Tanzania Mozambique South Africa South Africa South Africa South Africa South Africa South Africa South Africa Mali Be´nin Be´nin Cameroon Sri Lanka

Equatorial forest Acacia savanna Bushed savanna Savanna-grassland Savanna-grassland Savanna-woodland Savanna-bushland Savanna-bushland Savanna-bushland Savanna-bushland Dry savanna Wooded savanna Flood plain Evergreen tropical forest Forest in flood plain, meadow

Gunong Benom Kanha Gir Gunung Mulu

GUNO KANH GIRZ GUNU

Malaysia India India Malaysia

Montane tropical rain forest Forest, meadow Open forest Open forest

Kuala Lumpat

KUAL

Malaysia

Tropical rain forest

Mentis, 1970 Bourquin et al., 1971 Happold, 1969 Scheppe & Haas, 1976 Bourlie`re et al., 1974 Bourlie`re et al., 1974 Bousquet, 1992 Scheppe & Osborne, 1971 Happold, 1996 Rham, 1996 Rham & Christiaensen, 1963 Petter & Pujol, 1963 Lamprey, 1964 Dalquest, 1965, 1968 Rautenbach, 1978 Rautenbach, 1978 Rautenbach, 1978 Rautenbach, 1978 Rautenbach, 1978 Rautenbach, 1978 Rautenbach, 1978 Bousquet, 1992 Bousquet, 1992 Bousquet, 1992 Bousquet, 1992 Eisenberg & Lockhart, 1972 Medway, 1972 Schaller, 1967 Berwick, 1974 Anderson & Jermy, 1982 Medway & Wells, 1971

Palaeoenvironments: mammalian evidence

422

Table 21.6. List of the taxonomic categories used in the analysis of recent and fossil faunas Variates used for the study of the recent faunas V1 – Proboscidea V2 – Suidae V3 – Hippopotamidae V4 – Tragulidae V5 – Bovidae V6 – Cervidae V7 – Giraffidae V8 – Primates Variates used for the study of the fossil faunas V1 – Deinotheriidae V2 – Elephantidae V3 – Suidae V4 – Tragulidae V5 – Cervidae V6 – Bovidae V7 – Giraffidae V8 – Rhinocerotidae V9 – Tapiridae

V9 – Equidae V10 – Rhinocerotidae V11 – Tapiridae V12 – Hyracoidea V13 – Lagomorpha V14 – Hystricidae V15 – Tubulidentata V16 – Pholidota

V10 – Equidae V11 – Chalicotheriinae V12 – Schizotheriinae V13 – Tubulidentata V14 – Hystricidae V15 – Hyracoidea V16 – Castoridae V17 – Primates

Recent faunas Several recent faunas of Africa and Asia have been listed from literature data (Table 21.5). Some are evergreen tropical forest faunas, others come from Xood plains or savannas, others from wooded grasslands or mixed faunas. The aim of this analysis is to know if the environmental conditions are more signiWcant than geographic ones; if for example an African forest fauna plots near another African fauna whatever the environment or near an Asian forest fauna. Ecological similarities between African and Asian faunas have been shown already (Andrews, 1992). Here, the plotting on the plan of both Wrst and second axis (Fig. 21.3a) displays a grouping of the African forest faunas on the positive side of the Wrst axis and another grouping of the African savanna faunas on the negative side. The Asian faunas which come from quite forested areas have positive coordinates like the African forested ones and the Asian forested grassland faunas as well. The most signiWcant variates on the Wrst axis are V 4 (Tragulidae), V 6 (Cervidae), V 8 (Primates) and V 11 (Tapiridae) to identify the forest faunas and V 5 (Bovidae), V 7 (GiraYdae), V 9 (Equidae), V 10 (Rhinocerotidae) and V 15 (Tubulidentata) for the more open environments. The African forested sites plot very close together but the Asian localities are quite scattered along the second axis. The reason is probably that in the latter the number of species is quite low in

Late Miocene primate localities

423

[Figure 21.3] Correspondence factor analysis (CFA) of recent faunas listed in Table 21.5. (a) Plotting on the plan of the first and second axes. (b) Plotting on the plan of the first and third axes (for legend see Tables 21.5 and 21.6).

Palaeoenvironments: mammalian evidence

424

each locality, so small diVerences in some variates such as V 8 (Primates), V 13 (Lagomorpha) or V 14 (Hystricidae) can separate the plots of these localities. The grouping is far better on the plan of the Wrst and the third axis (Fig. 21.3b) when the weight of these variates is weak. It is interesting to note that the African locality of Comoe (COMO) whose fauna is mixed between savanna and forest plots exactly between the set of forest localities and that of savanna localities. On the other hand, both localities of Lamto (Lamto savanna and Lamto forest) are close together in the Weld but are clearly separated on the diagram, plotting with savannas and forests respectively. So the replacement of the recent faunas by the fossil ones using the same methodology could give signiWcant results for the palaeoenvironment. We think that the ecological trends of high taxonomic categories during the late Miocene were close to those observed in the recent faunas. Thus the same variates will have the same eVect in the analysis together with the eVect of the variates corresponding to taxonomic categories which have disappeared nowadays.

Fossil faunas A Wrst analysis considers late Miocene primate bearing localities of Europe and the Middle East (see Table 21.7) with few other elements: late middle Miocene European sites (Can Mata, Castel de Barbara, Montrejeau and La Grive) and Chinese localities (Lufeng and Longiakou). The latter six localities have been added to see if the diVerences could be due to geographical position or to an age diVerence. The plotting of each locality on the plan of Wrst and second axes (Fig. 21.4) shows a clear cut diVerence between western Europe and the localities of the Greco-Iranian province. Remembering what is mentioned above, we can assume that this pattern expresses an ecological diVerence. The variates linked to the western European localities, V 3 (Suidae), V 4 (Tragulidae) or V 5 (Cervidae), are the same as those that indicate wooded environments in the study of the recent faunas. It is certainly the same in this analysis. It is also noted that Chalicotheriinae (V 11) are linked to quite wooded biotopes even if it is possible to Wnd some of them in open environments. Castoridae (V 16) indicate probably more humid conditions. On the other hand, Bovidae (V 6), GiraYdae (V 7), Tubulidentata (V 13) and Hyracoidea (V 15) are more characteristic of open environments. The diVerences between the two sets are clear and they correspond to diVerent climatic conditions and diVerent environments. During the late Miocene, Vallesian and probably Turolian, western Europe was more humid and more forested than the Greco-Iranian faunal province even if the diVerence is less marked during the Turolian. This result is not

Late Miocene primate localities

Table 21.7. List of the fossil mammal bearing localities used for comparison with the late Miocene Macedonian sites Localities

Countries

References

Can Mata

CANM

Spain

MN 7-8

Castel de Barbara Can Llobateres El Firal Can Ponsich La Tarumba Torrent de Fibulinas Crevillente 2

CAST CANL ELFI CANP LATR TORR CREV

Spain Spain Spain Spain Spain Spain Spain

MN 7-8 MN 9 MN 9 MN 9 MN 10 MN 10-11 MN 11

Puente Minero Piera Cerro de la Garita Los Mansuetos El Arquillo Las Casiones La Grive L7 Montrejeau Montredon Soblay Aubignas 2 Luberon Baccinello Wissberg Eppelsheim Rudabanya Csakvar Baltavar Polgardi Kalimanci 2 Grebeniki Tirov Veles Kemiklitepe D Kemiklitepe A-B Maragheh M Molayan Pikermi Pentalophos Halmyropotamos Maramena Nikiti 1 Dytiko 1 Dytiko 2 Dytiko 3 Prochoma Vathylakkos 2 Vathylakkos 3 Ravin de la Pluie Ravin des Zouaves 5 Lufeng Longiakou

PUEN PIER CERR LOSM ELAR LASC LAGR MONJ MOND SOBL AUBI LUBE BACC WISS EPPE RUDA CSAK BALT POLG KALI GREB TITO KTPD KTPA MARM MOLA PIKE PNTL HALM MANE NIKI DTKO DITO DKOO PROC VTK2 VAT3 RAPL RAZO LUFE LONG

Spain Spain Spain Spain Spain Spain France France France France France France Italy Germany Germany Hungary Hungary Hungary Hungary Bulgaria Ukraine FYROM Turkey Turkey Iran Afghanistan Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece China China

MN 11 MN 11 MN 12 MN 12 MN 13 MN 13 MN 7-8 MN 7-8 MN 10 MN 10 MN 12 MN 13 MN 13 MN 9 MN 9 MN 9 MN 11 MN 13 MN 13 MN 12-13 MN 11 MN 12? MN 11 MN 12 MN 11 MN 12 MN 12 MN 9? MN 12 MN 13 MN 10 MN 13 MN 13 MN 13 MN 11/12 MN 11/12 MN 11/12 MN 10 MN 11 MN 12? MN 12?

Agustı´ et al., 1983–4; Moya`-Sola` et al., 1989–90 Moya`-Sola` et al., 1989–90 Agustı´ et al., 1996 Golpe, 1074 Moya`-Sola` et al., 1989–90 ESF Alcala, 1994 Alcala & Montoya, 1989–90; Azanza & Montoya, 1995 Alcala, 1994 Alcala & Montoya, 1989–90 Alcala, 1994 Alcala, 1994 Alcala, 1994 Alcala, 1994 Escuille´ (unpublished) Ginsburg, 1974 Michaux, 1988 Gue´rin & Mein, 1971 Azanza et al., 1993 Gaudry, 1873 Rook et al., 1996 Wagner, 1946 Thenius, 1959 Kordos, 1991 Kretzoi, 1954 Kormos, 1913, Circ, 1957 Ciric, 1957 Kojumdkieva et al., 1982 Gabunia, 1981 Ciric, 1957 Bonis et al., 1994 Bonis et al., 1994 Bernor et al., 1996 Brunet et al., 1984 Bonsi et al., 1992 Bouvrain, 1997 Melentis, 1967 Schmidt-Kittler et al., 1995 Kostopoulos & Koufos, 1996 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Bonis et al., 1992 Qi, 1993 Xue & Coombs, 1985

Palaeoenvironments: mammalian evidence

426

[Figure 21.4] Correspondence factor analysis (CFA) of late Miocene primate bearing localities of the Greco-Iranian Faunal Province compared to western European localities (two Chinese late Miocene localities and four European late middle Miocene localities have been added to the analysis) (for legend see Tables 21.6 and 21.7).

very diVerent from that of Fortelius et al. (1996) who Wnd that there are ‘more closed habitats in West, with the trend toward opening of the tree cover never reaching the extent seen in East’. The position of central European localities, Polgardi and Baltavar, is intermediate and it could correspond to an intermediate ecosystem. The plotting of Chinese localities close to the Spanish ones indicates that despite the geographical diVerence, the ecological conditions were probably similar. In the same way, Can Mata, Castel de Barbera, Montrejeau and La Grive are grouped together with the other European sites despite their older ages. The comparison of the Vallesian localities, with or without primates, from Greece and western Europe (Spain, France or Germany) (Fig. 21.5) indicates that the distance on the graph between both sets is very large. It means a large ecological diVerence, during the Vallesian time, between the uppermost ends of Europe, the west being far more forested than the southeastern part. DiVerences between Turolian localities, with or without primates, in the same areas (Fig. 21.6) exist, but they are far weaker. Both areas are well separated on the whole but some localities of western Europe plot with some localities from the Greco-Iranian Faunal Province. This can lead to the conclusion that climatic conditions were less diVerent in both areas during the end (Turolian) than during the beginning (Vallesian) of the late Miocene.

Late Miocene primate localities

[Figure 21.5] Correspondence factor analysis (CFA) of Vallesian mammalian faunas from the GIFP compared to western Europe (for legend see Tables 21.6 and 21.7).

[Figure 21.6] Correspondence factor analysis (CFA) of Turolian mammalian faunas from the GIFP compared to western Europe (for legend see Tables 21.6 and 21.7).

Another analysis takes into account the Vallesian and Turolian localities of Western Europe (Fig. 21.7). It indicates that most of the Vallesian sites are eparatedfrom most of the Turolian ones but there is not a clear cut separation between both groups. There was certainly an ecological shifting between Vallesian and Turolian but it was not a dramatic one. The Turolian environment was more open or less forested than the Vallesian one on the whole but some localities of each group could have a quite similar environment.

427

Palaeoenvironments: mammalian evidence

428

[Figure 21.7] Correspondence factor analysis (CFA) of Vallesian and Turolian faunas from western Europe (for legend see Tables 21.6 and 21.7).

[Figure 21.8] Correspondence factor analysis (CFA) of Vallesian and Turolian faunas from the GIFP (for legend see Tables 21.6 and 21.7).

The same kind of analysis (Vallesian/Turolian) for the Greco-Iranian Faunal Province localities reaches a diVerent result (Fig. 21.8). Here it is impossible to distinguish Vallesian from Turolian fossil mammal bearing sites. The conclusion is that the ecological conditions were similar (open environment) in this area during the whole of the late Miocene. But in fact, another study using the genus level as the operational taxonomic unit gives another aspect. A cluster analysis on the basis of presence and absence of the diVerent genera in the studied localities separates clearly the Vallesian

Late Miocene primate localities

429

[Figure 21.9] Cluster (hierarchic) analysis of the Macedonian late Miocene mammalian assemblages (for legend see Table 21.7).

localities from the Turolian ones (Fig. 21.9). The number of species in the higher categories (bovids, giraYds, tragulids . . .) could be the same but the genera and of course the species are diVerent. Looking at the bovids for example, Mesembriacerus, Helladorcas and Ouzocerus disappear in the Turolian sites (nevertheless Ouzocerus has been identiWed, on fragmentary remains, in the late Turolian locality Maramena), while Samotragus, Prostrepsiceros or Oioceros are represented through diVerent species. The genus Hipparion is present both in the Vallesian and Turolian localities but with diVerent species, number of species and morphology. It is represented in the Vallesian sites by two species while in the Turolian ones there are always three species. Moreover the Vallesian species are characterized by robust and brachyodont forms with richly plicated enamel while the Turolian ones are slender, hypsodont, and with fewer enamel plications. So there is a signiWcant faunal turnover during the late Miocene which corresponds also to the replacement of the hominoid primates by cercopithecoids, which was probably due to change in the ecological conditions (a little colder or a little dryer?). This change was not exactly similar to that which occured in western Europe but it probably corresponded to the same climatic shifting.

Conclusions For this study, we have used published (and when possible recently revised) faunal lists as well as the faunas we have directly studied. We have noted a signiWcant general turnover between the Vallesian and Turolian ages of mammals, everywhere they have been studied. The intensity and signiWcance of the changes are not exactly the same, and they depend on the place they are studied, western Europe, central Europe or southeastern Europe and the proximal area. The general comparison of the taxonomic conditions

Palaeoenvironments: mammalian evidence

430

of the faunas can give good results when compared with recent faunas. On the other hand, it seems clear that an isolated specimen cannot give by itself signiWcant information on the palaeoenvironment. The fossil mammal assemblages are not unequivocal. They are the results of diVerent processes in which diVerent kinds of mammal could be fossilized. A forest animal could have been buried by chance with open environment dwellers especially in river deposits. The method that we are using has been criticized (Eisenmann & Mein, 1996) for the use of the number of species of equids as a variate in a multivariate analysis. For these authors ‘at any rate, the contribution of Equidae to one or the other factorial axis will depend on whether the number of hipparions adapted to arid or humid conditions was predominant in the samples used for that analysis’. It is true that the number of species of equids is not weighty in our analysis. So we must conclude that the number of species of equids could be the same in diVerent environments. But, as a lonely swallow cannot make a summer, we are not sure that a lonely species of Hipparion can determine by itself a peculiar palaeoenvironment (see above for the problem of chalicotherids). To admit that a species was adapted to a particular environment and to conclude that a locality was representative of this palaeoenvironment because this species is present is a circular reasoning which comes back on itself like a snake which bites its own tail. It seems better to reach a conclusion on palaeoenvironment from the study of the fauna (rather than just a small part). So, specialized students of hipparions (Bernor et al., 1996) have referred, for example, Hippotherium primigenium in Vallesian localities of Germany and Hungary although the bulk of the faunas of the sites they are coming from are very diVerent and indicate diVerent palaeoenvironments. The same species of Hipparion is also referred by other specialists (Alberdi et al., 1997) in Spanish Vallesian and Turolian sites whose environments were also diVerent. In Europe there were always changes between Vallesian and Turolian mammalian localities. These changes seem to correspond to a climatic shift toward dryer conditions but there were some diVerences depending on the place where these changes were occurring. The most signiWcant changes are found in western Europe and the less signiWcant ones in southeastern Europe. In the latter area, the turnovers in the mammalian assemblages concerned mostly the speciWc or the generic level, but the number of species in the mammalian higher taxonomic categories remained quite the same. More studies will be necessary for a better understanding of these problems. New fossil material from new excavations in newly discovered sites as well as in known localities will be necessary, together with careful revision and comparisons of the existing material of eastern, central and western

Late Miocene primate localities

Europe. For instance, the present study is focused on Northern Greece localities and it does not take into account the ‘mid-Vallesian crisis’ because of the poor number of localities in Greece. Palaeomagnetic data and, when possible, radiochronological data will permit us to access better the timing of these changes, and geochemical analysis of soils, bones and tooth enamel as well as analysis of tooth microwear will contribute signiWcantly to a synthesis. Some of this work has begun and we can be sure that our knowledge on the evolution of the Hominoidea and on the European late Miocene environments will increase greatly in the near future.

Acknowledgements We thank the European Science Foundation which allowed us to attend the workshop of Siena, and L. Rook for its organisation. The last Weld campaigns in Greece were funded by the Singer-Polignac Foundation and the Leakey Foundation for Anthropological Research. This study is also a part of the program ‘Paleoenvironnement – Evolution des Hominide´s’ of the French CNRS. We thank very much our colleague P. Andrews whose remarks improved greatly the manuscript. We thank also G. Florent who prepared the manuscript and S. RiVaud for the graphics.

References Agustı´, J., Cabrera, L. & Moya`-Sola`, S. 1983–4. Sinopsis estratigraWca del Neogeno de la fossa del Valle`s-Penede`s. Paleontologia y Evolucion, 18, 57–81. Agustı´, J., Cabrera, L., Garces, M., Llenas, M., Moya`-Sola`, S. & Paris, J. M. 1996. The Vallesian of the Valle`s Penede`s Basin (NE Spain). Field-trip Guide Book. European Science Foundation, Workshop on the Vallesian, San Feliu de Guixols. Alberdi, M. T., Azanza, B., Cerdeno, A. & Prado, J. L. 1997. Similarity relationship between Mammal faunas and biochronology from latest Miocene to Pleistocene in the Western Mediterranean area. Eclogae Geologicae Helvetiae, 90, 115–32. Alcala Martinez, L. 1994. Macromamiferos Neogenos de la Fosa de Alfambra-Teruel. Instituto de Estudios Turolienses, Museo Nacional de Ciencias Naturales, pp. 1–554. Alcala Martinez, L. & Montoya, P. 1989–90. Los faunas de macromamiferos del Turoliense inferior espanol. Paleontologia y Evolucion, 23, 111–19. Andrews, C. W. 1918. Note on some fossil mammals from Salonica and Imbros. Geological Magazine, London, 5 (6), 540–43. Andrews, P. 1992. Community evolution in forest habitats. Journal of Human Evolution, 22, 423–38.

431

Palaeoenvironments: mammalian evidence

432

Andrews, P. 1995. Mammals as palaeoecological indicators. Acta Zoologica Cracoviensis, 38 (1), 59–72. Andrews, P., Lord, J., & Evans, E. M. N. 1979. Patterns of ecological diversity in fossil and modern mammalian faunas. Biological Journal of the Linnean Society, 11, 177–205. Anderson, J. A. R. & Jermy, A. C. 1982. Gunung Mulu National Park, Sarawak. The Earl of Cranbrook, Royal Geographical Society, London, pp. 1–345. Arambourg, C. & Piveteau, J. 1929. Les verte´bre´s du Pontien de Salonique. Annales de Pale´ontologie, Paris, 18, 59–138. Azanza, B., Broin, F. de, Galoyer, A., Ginsburg, L. & Zouhri, S. 1993. Un nouveau site a` mammife`res dans le Mioce`ne supe´rieur d’Aubignas (Arde`che). Comptes Rendus de l’Acade´mie des Sciences, Paris, II, 317, 1129–34. Azanza, B. & Montoya, P. 1995. A new deer from the lower Turolian of Spain. Journal of Paleontology, 69 (6), 1163–75. Bernor, R. L. 1978. The mammalian systematics, biostratigraphy, and biochronology of Marageh, Iran. Ph.D. thesis, University of California, Los Angeles, pp. 1314. Bernor, R. L., Solounias, N., Swisher III, C. C. & Van Couvering, J. A. 1996. The correlation of the three classical ‘Pikermian’ mammal faunas-Maragheh, Samos and Pikermi with the European MN unit system. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R.L., Fahlbusch, V. & Mittmann, H. W. (eds.), pp. 137–54. Columbia University Press, New York. Berwick, S. 1974. The community of wild Ruminants in the Gir forest ecosystem. India. Ph.D. Yale University, unpublished, pp. 1–248. Bonis, L. de, Bouvrain, G. & Geraads, D. 1979. Artiodactyles du Mioce` ne supe´rieur de Mace´doine. 7 e International Congress of Mediterranean Neogene, Annales Ge´ologiques des Pays Helle´niques, h.s, 1, 167–75. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. 1992. Diversity and paleoecology of Greek late Miocene mammalian faunas. Palaeogeography, Palaeoclimatology, Palaeoecology, 91, 99–121. Bonis, L. de, Bouvrain, G., Geraads, D., Koufos, G., Sen, S. & Tassy, P. 1994. Les gisements de mammife`res du Mioce`ne supe´rieur de Kemiklitepe, Turquie. 11. Biochronologie, pale´oe´cologie et relations pale´oge´ographiques. Bulletin du Muse´um national d’Histoire Naturelle, Paris, 4e se´r., 16, C (1), 225–40. Bonis, L. de, Bouvrain, G. & Koufos, G. 1987. Late Miocene mammal localities of the lower Axios valley (Macedonia, Greece) and their stratigraphic signiWcance. Modern Geology, 13 (2), 141–7. Bonis, L. de, Brunet, M., Heintz, E. & Sen, S. 1993. La province gre´co-irano-afghane et la re´parition des faunes mammaliennes au Mioce`ne supe´rieur. Paleontologia y Evolucion, Barcelone, 96–106. Bourlie`re, F., Minner, E. & Vuattoux, R. 1974. Les grands mammife`res de la re´gion de Lasuto, Coˆte d’Ivoire. Mammalia, 38 (3), 433–47. Bourquin, O., Vincent, J. & Hitchim, P. 1971. The Vertebrates of the Hluhluwe game reserve-corridor (State-land) – Umfolozi game reserve complex. The Lammergeyer, 14, 5–58. Bousquet, B. 1992. Guide des parcs nationaux d’Afrique Afrique du nord-Afrique de l’Ouest. Delachaux et Niestle´, Neuchatel, 1–368. Bouvrain, G. 1997. Les bovide´s du Mioce`ne supe´rieur de Pentalophos (Mace´doine, Gre`ce). Mu ¨ ncher Geowiss. Abhandlungen, A, 34, 5–22.

Late Miocene primate localities

Brunet, M., Heintz, E. & Battail, B. 1984. Molayan (Afghanistan) and the Khaur Siwaliks of Pakistan: an example of biogeographic isolation of late Miocene mammalian faunas. Geologie en Mijnbouw, Amsterdam, 63, 31–8. Ciric, A. 1957. Pikermiska fauna iz okoline Titovog Velesa. Bulletin du Museum d’Histoire Naturelle du Pays Serbe, ser. A, 8, 1–82. Dalquest, W. W. 1965. Mammals from the Save river Mozambique, with descriptions of two new bats. Journal of Mammalogy, 46 (2), 254–64. Dalquest, W. W. 1968. Additional notes on mammals from Mozambique. Journal of Mammalogy, 49 (1), 117–21. Eisenberg, J. & Lockhart, M. 1972. An ecological reconnaissance of Wilpattu National Park, Ceylan. Smithsonian Contributions to Zoology, 101, 1–118. Eisenmann, V. & Mein, P. 1996. Revision of the faunal list and study of Hipparion (Equidae, Perissodactyla, Mammalia) of the Pliocene locality of La Gloria 4 (Spain). Acta Zoologica Cracoviense, Krakow, 39 (1), 121–30. EscoWer, B. & Pages, J. 1990. Analyses Factorielles Simples et Multiples. Dunod, Paris, pp. 1–267. Escuille, F., unpublished. La faune de l’Astaracien (MN7–MN8) en nord-Dauphine´. Fortelius, M., Werdelin, L., Andrews, P., Bernor, R. L., Gentry, A., Humphrey, L., Mittmann, H.-W. & Viratana, S. 1996. Provinciality, diversity, turnover, and palaeoecology in land mammal faunas of the late Miocene of Western Eurasia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Falhbusch, V. & Mittmann, H. W. (eds.), pp. 414–48. Columbia University Press, New York. Gabunia, L. 1981. Traits essentiels de l’e´volution des faunes de mammife`res ne´oge`nes de la re´gion mer Noire-Caspienne. Bulletin du Muse´um National d’Histoire Naturelle, Paris, 4e se´r., 3, C (2), 195–204. Gaudry, A. 1973. Animaux Fossiles du Mont Le´beron. Etudes sur les verte´bre´s. F. Savy, Paris, pp. 1–112. Geerling, C. & Bokdam, J. 1973. Fauna of the Comoe national park, Ivory Coast Biological Conservation, 5 (4), 251–7. Ginsburg, L. 1974. Les faunes de mammife`res bundigaliens et vindoboniens des bassins de la Loire et de la Garonne. Me´moire du BRGM, 78 (1), 153–67. Golpe, J. M. 1974. Faunas de yacimientos con suiformes en el Terciario espanol. Paleontologia y Evolucion, 8, 1–87. Gue´rin, C. & Mein, P. 1971. Les principaux gisements de mammife`res mioce`nes et plioce`nes du domaine rhodanien. Documents du Laboratoire de Ge´ologie de la Faculte´ des Sciences de Lyon, H.S., 131–70. Happold, D. C. D. 1969. The mammalian fauna of some Jebels in the northern Sudan. Journal of Zoology, 157 (2), 144–5. Happold, D. C. D. 1996. Mammals of the Guinea-Congo rain forest. Proceedings of the Royal Society of Edinburgh, B, 104, 243–84. Kojumdkieva, E., Nicolov, I., Nedjalkov, P. & Busev, A. 1982. Stratigraphy of the Neogene in Sandanski graben. Geologica Balkanica, 12 (3), 69–81. Kordos, L. 1991. Le Rudapithecus hungaricus de Rudabanya. L’Anthropologie, 95, 343–62. Kormos, Th. 1913. Uber die resultate meiner Ausgrabungen in Jahr 1913. Jahresbereiht der Ko¨niglich Ungarischen Geologischen Drichsandstalt, Budapest, 559–604. Kostopoulos, D. S. & Koufos, G. D. 1996. Late Miocene bovids (Mammalia,

433

Palaeoenvironments: mammalian evidence

434

Artiodactyla) from the locality ‘Nikiti-1’ (NKT), Macedonia, Greeece. Annales de Pale´ ontologie, 81 (4), 251–300. Koufos, G. D. 1985. Hipparion sp. (Equidae, Perissodactyla) from Diavata (Thessaloniki, northern Greece). Bulletin du Muse´um national d’Histoire Naturelle, (Ge´ol.), 38 (5), 335–45. Koufos, G. D. 1995. The Wrst female maxilla of the hominoid Ouranopithecus macedoniensis from the late Miocene of Macedonia, Greece. Journal of Human Evolution, 29, 385–99. Kretzoi, M. 1954. Rapport Wnal des fouilles pale´ontologiques dans la grotte de Csakvar. Fo¨ldtani Integet evi Jelentesi, 1952, 37–68. Lamprey, H. F. 1964. Estimation of the large mammal densities, biomass and energy exchange in the Tarangire game reserve and the Massai steppe in Tanganyika. East African Wild Life Journal, 2, 1–46. Lartiges, A. & Poilecot, P. 1997. Evolution re´cente de la grande faune dans le parc national de Comoe´ en Co¨te d’Ivoire. Bulletin Mensuel de l’OYce National de la Chasse, 226, 20–31. Lebart, L., Morineau, A. & Warwick, K. M. 1984. Multivariate Descriptive Statistical Analysis. Correspondence Analysis and Related Techniques for Large Matrices. Wiley series in probability and mathematical statistics, John Wiley and Sons, New York, pp. 1–231. Legendre, S. 1986. Analysis of mammalian communities from the late Eocene and Oligocene of southern France. Palaeovertebrata, 16, 191–212. Medway, L. 1972. The distribution and altitudinal zonation of birds and mammals in Gunung Benom. Bulletin of the British Museum Natural History (Zoology), 23, 105–54. Medway, L. & Wells, P. 1971. Diversity and density of birds and mammals at Kuala Lumpur. Malayan Nature Journal, 24, 238–47. Melentis, J. 1967. Die Pikermifauna von Halmyropotamos (Euboa, Griechenland). Annales Ge´ologiques des Pays Helle´niques, 19, 283–411. Mentis, M. T. 1970. Estimates of natural biomasses of large herbivores in the Umfolozi game reserve area. Mammalia, 34 (3), 363–93. Michaux, J. 1988. Contributions a` l’e´tude du gisement mioce`ne supe´rieur de Montredon (He´rault). Les grands mammife`res. 10. Conclusions ge´ne´rales. Palaeovertebrata, Me´moire Extraordinaire, 189–92. Moya`-Sola`, S., Pons Moya, J. & Koelher, M. 1989–90. Primates catharrinos (Mammalia) del Neogeno de la peninsula iberica. Paleontologia y Evolucion, 23, 41–5. Petter, F. & Pujol, R. 1963. Noms vernaculaires lissongo des mammife`res de la re´gion de La Maboke´. Cahiers de la Maboke´, 1 (2), 120–2. Poulet, A. R. 1972. Recherches e´cologiques sur une savane sahe´lienne du Ferlo septentrional, Se´ne´gal: les mammife`res. La Terre et la Vie, n.s., 26 (3), 440–72. Qi, G. 1993. The environment and ecology of the Lufeng hominoids. Journal of Human Evolution, 24 (1), 3–11. Rahm, V. 1966. Les mammife`res de la foreˆt e´quatoriale de l’est du Congo. Annales du Muse´e Royal d’Afrique Centrale, se´r. 8, 149. Rahm, V. & Christiaensen, A. 1963. Les mammife`res de la re´gion occidentale du lac Kivu. Annales du Muse´e Royal d’Afrique Centrale, se´r. 8, 118, 1–83. Rautenbach, I. L. 1978. Ecological distribution of the mammals of the Transvaal. Annals of the Transvaal Museum, 31 (10), 131–56.

Late Miocene primate localities

Rook, L., Harrisson, T. & Engesser, B. 1996. The taxonomic status and biochronological implications of new Wnds of Oreopithecus from Baccinello (Tuscany, Italy). Journal of Human Evolution, 38 (1), 3–27. Schaller, G. B. 1967. The Deer and the Tiger. University of Chicago Press, pp. 1–370. Scheppe, W. & Haas, P. 1976. Large mammal populations of the lower Chobe river, Bostwana. Mammalia, 40 (2), 223–43. Scheppe, W. & Osborne, T. 1971. Patterns of use of a Xood plain by Zambian mammals. Ecological Monographics, 41 (3), 179–205. Schmidt-Kittler, N., de Bruijn, H. & Doukas, C. 1995. The vertebrate locality Maamena (Macedonia, Greece) at the Turolian-Ruscinian boundary (Neogene). 1. General introduction. Mu ¨ nchen Geowissenchaften Abhandlungen, A, 28, 9–18. Thenius, E. 1959. Tertia¨r. 2e Teil: Wirbeltierfauna, Enke, F. (ed.), pp. 1–328. Stuttgart. Wagner, W. 1946. Die unterplioza¨ne Wirbeltierfauna von Wissberg bei Gau-Weinheim in Rhein-hessen. Wissenschaftliche Vero¨Ventlichungen der Technischen Hochschule Darmstadt, 1 (4), 19–28. Xue, X. & Coombs, M. C. 1985. A new species of Chalicotherium from the upper Miocene of Ganou province, China. Journal of Vertebrate Paleontology, 5, 336–44.

435

22 The paleoecology of the Pikermian Biome and the savanna myth Nikos Solounias, J. Michael Plavcan, Jay Quade and Lawrence Witmer

Introduction Many important Eurasian later Miocene localities are characterized by an abundance of species of aardvarks, hyenas, hyraxes, mastodons, hipparions, rhinoceroses, giraVes, and antelopes (Kurte´n, 1952, 1971; Thenius, 1959). The combination of this type of species is only encountered in modern African savannas and therefore such Miocene faunas were originally interpreted to represent savannas (Gaudry, 1862–7; Osborn, 1910; Abel, 1927). More recently, such faunas have also been interpreted to represent savannas (Estes, 1971; Kurte´n, 1952, 1971; Webb, 1983; Janis, 1982, 1989; de Bonis et al., 1992). The single and most inXuential character for interpreting extinct faunas as savannas throughout the Miocene is the presence of tall teeth (hypsodont teeth) in ungulates (Webb, 1983). Among all ungulates, the presence of hypsodont equids has been the most inXuential in determining the presence of savannas in the past (Simpson, 1951; Thenius, 1969; Kurte´n, 1952, 1971; Webb, 1983; Janis, 1982, 1989; de Bonis et al., 1992; MacFadden, 1992). These faunas have also been widely termed ‘hipparion faunas’ because they contain on the average four to seven species of hipparions (Kurte´n, 1952, 1971; Sondaar, 1971; Berggren & Van Couvering, 1974; Woodburne, 1989). Hipparions were widespread and successful species throughout the World (Bernor et al., 1988, 1990; Woodburne, 1989). They were hypsodont equids with three toes and an isolated protocone. Most hipparions have been interpreted to ecologically resemble modern zebras and horses which graze and live in savannas and grasslands (Matthew, 1926; Simpson, 1951; Thenius, 1969; Janis, 1982; Webb, 1983; MacFadden, 1992). Other researchers have interpreted hipparions as species which inhabited woodlands (Bernor et al., 1988, 1990; Hayek et al., 1992). Hipparion faunas have been assumed to represent savannas similar to the savannas of modern Africa, not only because of the hipparion–zebra analogy, but also because of the systematic similarity of extinct carnivores and other ungulates to those of the modern savannas (Thenius, 1959; Kurte´n, 1952, 1971; Estes, 1971; Webb, 1983; de Bonis et al., 1992). The appearance and spread of such faunas approximately 9 million years ago contrasts to older Eurasian faunas which have diVerent dog-like hyenas, no true bone crushing hyenas (i.e. Percorcuta excluded), and very few large ungulates. Kurte´n (1952) provided a map of Eurasia which included the most typical

Paleoecology of the Pikermian Biome

savanna-like localities. In this map a vast oval region of such faunas extends from the west (Greece) through south Russia, Turkey and Iran to the east in China. Kurte´n (1952) listed the following hipparion faunas: Mont Le´beron (South France), Vathylakkos-Thessaloliki, Pikermi, and Samos (Greece), Sinap (Turkey), Sebastopol (Ukraine), Maragheh (Iran), Pawlodar (Russia) and Shanxi (China). Kurte´n’s map could be extended to include later Miocene localities of North Africa and Spain. The hipparion faunas are clearly located where modern dry habitats are. During the later Miocene, north and peripheral to this vast oval area of hipparion faunas, forested faunas have been sampled which had fewer hipparion species (Bernor & Franzen, 1997). The boundaries of this map, however, are an oversimpliWcation as there are many exceptions. Crusafont-Pairo (1950) termed these hipparion faunas ‘Pikermian’ after Pikermi in Greece, the Wrst important locality to be discovered during the beginning of the nineteenth century (Solounias, 1981a,b). More recently, Bernor (1983) identiWed many of these localities as representing a special ecology; the Old World evergreen woodland biome which in turn has been subdivided by Bernor into provinces. We propose the term Pikermian Biome for these provinces. The present study explores two aspects of the Pikermian Biome: (1) such a biome was not a savanna but a sclerophyllous evergreen woodland (a summary of previous research); and (2) that many modern African savanna mammals were to a large extent derived from species of the Pikermian Biome (a new theory). We presently propose this theory as a minor point of this study and leave out the details supporting it. Researchers are invited to either re-aYrm it or attack it. The scenario that the hipparion faunas represent savannas has begun to change in the past 20 years. For example, we have proposed that the paleoecology of these hipparion faunas was not African-like savannas but sclerophyllous evergreen woodlands (Bernor et al., 1979; Solounias & Dawson-Saunders, 1988; Ioakim & Solounias, 1985; Quade et al., 1994; Bernor et al., 1988, 1990; Hayek et al., 1992). Solounias has investigated the masticatory morphology and tooth microwear of several ungulates from the locality Pikermi and especially Samos (a locality faunally similar to Pikermi). These two localities are the most typical of the Pikermian Biome and Samos includes as least seven species of hipparion (Solounias, 1981a,b; Bernor, 1998 personal communication). We also suggest that the signiWcance of the Pikermian Biome extends beyond a simple characterization of the local paleoenvironment. It is believed that extant African savanna faunas, particularly ungulates and carnivores, evolved primarily from endemic African ancestors within Africa. For example, Churcher (1978) derives both the okapi and the giraVe from

437

Palaeoenvironments: mammalian evidence

438

ancestral African GiraYdae and shows that all other Eurasian GiraYdae were also derived from the African stalks and migrated out of Africa. Coppens (1994) shows with panoramic views and a sequence of stages that the entire modern savanna fauna of Africa evolved step by step from the archaic African Miocene taxa. In general, African mammals have been seen as having their evolutionary roots within Africa (Van Couvering & Van Couvering, 1976). We propose that many modern African savanna dwelling large mammals originated not from forest dwelling African Middle Miocene relatives, but rather from taxa of the Pikermian Biome.

Comments on the paleoecology of the Pikermian Biome

Historical The theory that the Pikermian Biome represents savannas or grasslands similar to those of modern central Africa can be found in a few old publications. Gaudry (1862–7) and Abel (1927) have speciWcally stated the similarity of the Pikermi community to the modern Masai Mara of Kenya. Restorations of the Pikermi ecology show strong similarity with the African savanna (e.g. Abel, 1927). Pictorial restorations of hipparions often show them to have been striped like zebras (e.g. Thenius, 1966). However, the majority of the early researchers have expressed the ‘savanna view’ indirectly, in two ways. (1) By naming of several Miocene species of the Pikermian Biome by the same genus with those of modern Africa (e.g. Struthio caratheodoris, Hystrix primigenia, Felis, Hyaena eximia, Hyaena chaeretis, Orycteropus gaudryi, Diceros pachygnathus, GiraVa attica, Gazella capricornis). Other genera diVer but their name contains roots of names indicating similarities and modern African aYnities (e.g. Palaeomanis, Metailurus, Pliohyrax, Postpotamochoerus, Prostrepciceros, Protragelaphus, Prodamaliscus, Protoryx, Palaeoryx). (2) The ‘savanna view’ is explicit whenever ungulate hypsodonty is mentioned in the vast paleontological literature, not only speciWcally for the Pikermi and Samos taxa, but also in similar species found in Russia, Middle East, Iran, China, Africa, and North America. Hypsodonty is an inXuential issue for interpreting certain paleohabitats as savanna (Webb, 1983; Janis, 1982).

Stable isotopes from paleosols The isotopic composition can help determine the presence of woodlands or savannas. Soil carbonates clearly accumulate from the overlying

Paleoecology of the Pikermian Biome

vegetation. Isotopically, there are two types of grassland: C3 and C4. On the contrary, all trees and bushes are isotopically C3. Thus, a C4 signal deWnitely implies the presence of grasslands while a C3 signal implies either forest or a C3 grassland. Clearly, the predominant isotopic message derived from bush undergrowth and forests is C3. C3 grasslands do exist but develop in high altitude or in high latitude and need a wet season and cool temperature. The grasses which grow in forests are C3 because they are the only ones that are able to grow in the shade. On the contrary, C4 grasslands develop in low altitude and latitude and need a hot and dry season and a wet season. Therefore, it is unlikely for C3 grasslands to occur in the later Miocene deposits in Greece or other low latitude and altitude regions especially since the Miocene was milder and wetter than the present. A C3 signal from the Miocene of Greece would therefore imply the presence of more forested habitats. Mio-Pliocene Xuvial rocks containing buried paleosols are common in Greece and Turkey. We used the carbon and oxygen isotopic composition of pedogenic carbonates associated with these paleosols to estimate the proportion of C3 and C4 plants once present at the sites (Quade et al., 1994). Evidence from the paleosols in well-known fossil-bearing formations in the lower Axios Valley in Macedonia (e.g. Vathylakkos), Samos Island, Pikermi near Athens, Pasalar in NW Turkey, and Rhodes, all show that Mio-Pliocene biomasses were dominated by C3 plants, as the entire region is today. In addition, nearly all the sites we visited contained soil carbonate, in association with paleosols, indicating that mean annual precipitation has remained about 1 m in the last 11 Ma. The types of paleosols present are not consistent with the modern soils underlying C3 grasslands of the westernmost United States and southeastern Australia. Given these lines of evidence, we suggest that during the later Miocene, the Greek region was dominated by forest and woodland. The moisture regime could have been either winter- or summer-dominated, or both. Our Wndings imply that the classic fossil-bearing localities of the lower Axios Valley in Macedonia, Samos and Pikermi, Greece, were not savannas, as has been previously thought, but rather, woodlands or forests (Quade et al., 1994).

Isotopes extracted from teeth of Samos herbivores The paleosol reconstruction is supported by carbon isotopic evidence from fossil teeth from Samos (Quade et al., 1994). The d13C values of carbonates built into the structure of biogenic phosphates of selected mammals has been shown to correlate directly with the d13C values of dietary intake. The

439

Palaeoenvironments: mammalian evidence

440

biogenic fraction between diet and structural carbonates in phosphate is about 13. The proportion of C3 to C4 plants in a ruminant’s diet can thereby be estimated from the d13C value of biogenic carbonate. In 1989 we collected several mammalian specimens from the Turolianage beds on Samos. Additional tooth samples from a broader range of species were analyzed from Samos fossils of the American Museum of Natural History. Analysis of these yielded a range of d13C values between − 13.0 and − 9.6‰, with an average value of − 11.4‰. This implies a dietary intake in the range of − 26 to − 22.6‰, for an average of − 24.4‰. This is consistent with a diet composed largely or entirely of C3 plants for the nine ruminant species we analyzed (Table 22.1). We Wnd that most species, even the hipparions (Hipparion dietrichi, Hippotherium giganteum, Cremohipparium proboscideum), gave a C3 signal. This is a remarkable discovery substantiating the hypothesis proposed by the paleosols, that the Samos paleoenvironment was forests and woodlands. Hipparions must have either browsed or grazed on C3 grasses within the woodlands. Of the species we examined, Samotherium major displayed the most enriched values in 13C, indicating that it had a C4 component in its diet. Samotherium major was a grazer.

Isotopic conclusions The d13C values of soil carbonate and tooth enamel indicate that in the sections we sampled all over the Aegean region, C3 biomasses have dominated the ecology of the depositional basins over the past 11 Ma. C4 plants, likely grasses, have also been present, as apparent for example from the paleosol results on Rhodos Island and the younger portions of the lower Axios River deposits near Thessaloniki, but always in settings dominated by C3 plants. The presence of abundant soil carbonate in nearly all our sections suggests annual rainfall was under about 1 m over the period. Drawing upon modern analogs, the paleoecologic setting that best Wts our isotopic and soil morphological evidence is a forest or woodland with seasonal rainfall.

Paleobotany Direct comparisons between fossil plants and animals are rarely possible as they do not commonly fossilize in the same location. Numerous Miocene Xoras from the Mediterranean region have been sampled and studied. The general conclusion is that these Xoras represent subtropical forests and woodlands. The existence of widespread savannas and grasslands has not

Paleoecology of the Pikermian Biome

Table 22.1. Summary of dietary reconstructions for various Pikermian ungulates Masticatory morphologyc

Masticatory morphologyd

Masseter muscles

Premaxillae

Species

Localitya Isotopesb

Equidae Hipparion dietrichi Cremohipparion mediterraneum Cremohipparion matthewi Cremohipparion proboscideum Hippotherium giganteum

S PSX S S S

Cervidae Muntiacus Pliocervus pentelici

S P

browser browser

Giraffidae Helladotherium duvernoyi Palaeotragus rouenii Palaeotagus rouenii

P P S

mixed browser browser

Palaeotragus rouenii Palaeotragus coelophrys Palaeotragus quadricornis Palaeotragus coelophrys Samotherium boissieri

PS M S SM S

Samotherium major Samotherium neumayri

S M

Bohlinia attica Honanotherium schlosseri

PS X

Bovidae Miotragocerus or Tragoportax amalthea

SP

C3

browser or mixed

S SP S

C3 C3

browser or mixed browser or mixed browser

rugosifrons Gazella capricornis-gaudryi Protragelaphus skouzesi Procetrepciceros houtumschindleri Oioceros wegneri Prosinotragus kuhlmanni Palaeoreas lindermayeri Criotherium argalioides Parurmiatherium rugosifrons Palaeoryx pallasi Pachytragus crassicornis Pachytragus laticeps Pseudotragus capricornis Sporadotragus parvidens

S S S P S S SP S S S S

C3 C3

grazer mixed grazer mixed

C3 C3

C3

b

browser-mixed browser-mixed

mixed browsermixed

browser browser mixed C3 and C4

mixed

browser grazer grazer grazer

browser grazer

C3

browser browser browser browser mixed grazer mixed mixed mixed mixed mixed

Localities: P = Pikermi; S = Samos; X = Shanxi; M = Maragheh. Quade et al. (1994). c Solounias & Dawson Saunders (1988); Solounias et al. (1995). d Solounias & Moelleken (1993a). e Solounias & Moelleken (1992); Solounias & Hayek (1993); Hayek et al. (1992). a

Microweare

grazer grazer grazer mixed– grazer grazer mixed– grazer browser grazer

mixed or grazer grazer

mixed mixed

Palaeoenvironments: mammalian evidence

442

been reported from such Xoras (Xoral reviews in: Dorofeyev, 1966; Bernor et al., 1988, 1990). Thus, the interspersed contemporaneous faunas are unlikely to sample savannas either. The paleobotanical literature is extensive and we therefore provide only examples. A seminal study by Axelrod (1975) characterized the presence of sclerophyllous evergreen woodlands with chaparral undergrowth throughout the Miocene of the Mediterranean. Dorofeyev (1966) Wnds numerous woodlands intermixed with most hipparion faunas of the Old World. Several researchers report the presence of woodlands during the Miocene in the peri-Mediterranean regions (Takhajan, 1957; Givulescu & Florei, 1960; Benda, 1971; Benda & Bruijn, 1982; Leopold, 1969; Raven, 1971; Gregor, 1982; Saltilova & Ramishvili, 1984; Kovar-Eder, 1987; Bernor et al., 1990). Paleobotanists who have analyzed Xoras from Greece, presumably closer to Pikermi, also interpret them to represent woodland (Guernet et al., 1976; Sauvage, 1977; Weerd, 1983; Williman, 1980, 1983; Velitzelos & Gregor, 1985, 1986; Velitzelos & Knoboloch, 1986). Orgetta (1979) provides data which show the presence of rich riparian mild moist temperate woodlands at Pikermi in horizons geographically near and of similar age to the classic Turolian bone beds. Ioakim & Solounias (1985) found and described the presence of rich mild moist temperate woodlands on the low lying areas with evergreen conifers in the uplands during the Vallesian of Samos. Note that the Samos Vallesian (Mavradzei Beds) is naturally older (11.86–11.2 Ma) than the classic Samos Turolian bone horizons (7.2 Ma) (Weidmann et al., 1984). In summary, there is no palynological or paleobotanical evidence for widespread savannas in the Pikermian Biome. In conclusion, the Pikermian Biome mammals most likely inhabited sclerophyllous evergreen woodlands. Such woodlands were not an uninterrupted continuum of trees. Many regions were geographically highly heterogeneous and often separated by low and high mountains, lakes, and rivers. Complex coastal plains were also present both from the Paratethys and the Tethys. The central region of the Pikermian Biome occurred approximately where the modern Eurasian and North African dryer regions and deserts are located. That is in Greece, Turkey, Iran, Mongolia, and the Sahara. Fringes of the Pikermian Biome were in Spain, Balkans, the southern part of Russia, and the Shanxi region of China. It appears that the Late Miocene sclerophyllous woodland there was replaced by more open habitats during the Pliocene which eventually became dryer due to climatic change. Within these riparian woodlands there were a few predominantly C3 grasses as in modern Mediterranean environments and in the woodlands of Kanha Park in India. In the Pikermian Biome grasses grew in meadows,

Paleoecology of the Pikermian Biome

clearings, riverine and lake shores, and Xood plains within these woodlands. Importantly such C3 grasses provided grazing for some species. C3 grazing is a novel idea and could explain the observed hypsodonty of later Miocene ungulates (Cerling et al., 1997). These woodlands were similar to the present day undisturbed Madrean of the Americas (peri-mountainous vegetation – area between Navojoa and Guadalajara of Mexico) and other undisturbed Mediterranean climates. However, the later Miocene was warmer and wetter than the modern Mediterranean biotopes. Thus, the botanical physiognomy of the Miocene would diVer from a strict modern Mediterranean Xora, and might also resemble Kansas and Louisiana in some structural ways especially near the rivers and lakes (subtropical – wet woodland – austroriparian: Udvardy, 1978). Although the botanical aspects of the Pikermian Biome can be found today, however not in the original location of the Pikermian Biome, the faunal aspects clearly can not. Most of the Pikermian mammals have become extinct due to climatic change of their ecology; a few evolved into more modern species.

The faunas The Pikermian Biome faunas have been described by numerous researchers (not cited here). The Pikermi fauna of Greece in particular is commonly used in paleontological studies because most of the Pikermi species were found historically early and many of the species are types, and important in comparisons. Presently, there are rich faunas at Samos, numerous Shanxi localities in China and Maragheh in Iran where exquisitely preserved Pikermian faunas have been recovered (Pikermian faunal reviews may be found in Gaudry, 1862–7; Abel, 1927; Kurte´n, 1952, Solounias, 1981a,b; Bernor, 1978, 1983; Bernor et al., 1996 – papers on individual taxa are more than three hundred). The traditional method of evaluating the paleoecology of an extinct species was based on the assumption that their adaptations were similar to those of closely related modern species. African savanna species have been invariably selected as ideal models for interpreting most past communities (Janis, 1982; Webb, 1983). Ungulate and carnivore faunas from Russia, Europe, India and the Indomalayan realm have been largely ignored. The reason for this is the phenomenal diversity and beauty of the African savannas. Comparison of the Miocene to modern Africa savannas is one-sided and has inXuenced the results. Thus, hipparions were originally envisioned to be ecologically similar to horses and zebras and Samotherium to the okapi (Simpson, 1953; Churcher, 1978). Such assumptions have been shown

443

Palaeoenvironments: mammalian evidence

444

to be to some degree incorrect (Solounias et al., 1988; Hayek et al., 1992). Numerous bovids in a locality have also implied the presence of savannas (Thenius, 1959; Kurte´n, 1952, 1971; Estes, 1971; Webb, 1983; de Bonis et al., 1992). Hypsodont ungulates implied savannas (Simpson, 1951; Thenius, 1969; Estes, 1971; Kurte´n, 1952, 1971; Webb, 1983; Janis, 1989; de Bonis et al., 1992; MacFadden, 1992). New methods of paleoecological reconstruction include the subdivision of an extinct species into various morphological components and comparisons to equivalent components in modern species. Hypsodonty and masticatory morphology can be evaluated as isolated species-blind morphologies. Isotopic research has added new information (Cerling et al., 1997). The issue of hypsodonty is complex and will be addressed elsewhere. Grazing in woodlands on C3 vegetation is possible and we have addressed some of our Wndings in publications (summary in Table 22.1).

Masticatory morphology Results of masticatory morphology so far suggest that many of the Pikermian ruminants are best approximated not by savanna species but by those which inhabit some regions of Northern India such as the Park Kanha (Table 22.1). Kanha sustains a subtropical forest with vast glades and swampy lake margins. There, the ungulates are deer, antelopes, and rhinoceros (Schaller, 1967). The Kanha ruminant species are to some extent ecomorphs to many of the Pikermian taxa (Solounias & Dawson-Saunders, 1988; Solounias et al., 1995). A few Pikermian giraYds resemble grazing antelopes but the majority of hipparions and giraYds resemble the mixed feeders sambar and wapiti (Solounias et al., in press). Most of the extinct antelopes resemble modern deer from India or forest grazing and forest browsing antelopes. There are no species similar to modern savanna Equus burchelli, Equus grevyi, Alcelaphini, Hippotragini, Reduncini, and Aepycerotini. If we accept the old ‘savanna view’ as valid, this would imply that species inhabiting savannas 7–9 million years ago showed none of the masticatory adaptations to typical savanna diets seen in extant species. Put another way, if Miocene species were adapted to savanna, why have they changed so much over the past 10 million years? A far simpler explanation is that these Miocene species were adapted to inhabit woodlands not savannas. This conclusion is in agreement with the studies of masticatory morphology where typical Miocene bovids like Tragoportax are browsers or mixedfeeders and resemble modern deer from Kanha. The Pikermian Gazella, Prtotragelaphus, Prostrepciceros, Oioceros, Plalaeoreas, and Prosinotragus have been interpreted to be browsers (Solounias & Dawson-Saunders,

Paleoecology of the Pikermian Biome

1988). Morphologically, the most advanced Pikermian bovids such as Palaeoryx and Pachytragus resemble mixed feeders like the Grant’s gazelle (Gazella granti). Curiously of the most dietarily advanced Pikermian ruminants are giraYds like Palaeotragus rouenii, Palaeotragus coelophrys, and Samotherium boissieri, which resemble forest mixed feeders or grazers such as the chital of India (Axis axis) and the wapiti of Canada (Cervus canadensis). Samotherium major is an extreme case, as it resembles the wildebeest (Connochaetes taurinus). Samotherium major probably grazed on C3 variegation.

Tooth microwear analysis The isotopic data cannot resolve the C3 signal of grazing in a C3 grassland versus the signal of browsing in a C3 forest. Tooth microwear analysis can resolve this issue. Tooth microwear analyses of Miocene ungulates shows the presence of browsers, mixed feeders, and a few grazers (Table 22.1). Since the isotopic signal is C3, and the paleobotanical information that of sclerophyllous woodlands, then it can be assumed that browsing, mixed feeding, and the observed scarce grazing occurred within forests. It is a major Wnding that the archaic grazers grazed within forests. Several species of hipparions are mixed feeders and thus dietarily diVerent from modern horses and zebras (Table 22.1). SpeciWcally hipparions can be interpreted as a perissodactyl version of modern hypsodont browsers or mixed feeders; camels and takins (Camelus–Budorcas) or wapiti and sambar (Cervus canadensis–Cervus unicolor) respectively. The wapiti and sambar are ideal ecomorphs for Pikermian hipparions and bovids, as they are mixed feeders which inhabit woodlands. In addition the gaur, the mountain anoa, and the wood bison are grazer bovids which also inhabit woodlands (Bos gaurus, Bubalus quarlesi, Bison bison athabascae). Preliminary research in the paleodiet of hipparions suggests that some species approximated these extant taxa and that equid grazing originated in woodlands. Samotherium major was a grazing giraYd (Solounias et al., in press). Our data show that several Pikermian Bovidae were also mixed feeders and grazers but morphologically diVerent from modern mixed feeders and grazers (Solounias & Moelleken, 1992; Solounias et al., 1995). Dietary shifts from browsing to mixed feeding and grazing evolved in several independent lineages and within forests; hypsodonty evolved in parallel several times (Solounias & Dawson-Saunders, 1988). C3 grasses grow in glades and riverine margins of the forested regions and thus browsing ungulates experimented there for grazing. We suggest that the Wrst mixed feeders and grazers began feeding on C3 grasses within the Pikermian Biome woodlands. This

445

Palaeoenvironments: mammalian evidence

446

type of feeding was an exaptation for a dietary change to the modern C4 dominated African savannas. A critical issue of the paleodiet research is that the Pikermian taxa were exapted for the environments of the modern savannas of Africa which did not begin to form until the Pleistocene. This time gap between the Late Miocene Pikermian fauna and the Pleistocene aridity in Africa allows time for the evolution of modern savanna species and explains the dramatic morphological diVerences observed between the Pikermian taxa and those of modern African savannas (Solounias et al., in press).

A new model for the origin of African savannas It is accepted that a number of extant, regional mammalian faunas arose in part from the migration of faunas that evolved in other regions. For example, numerous extant South American carnivores and ungulates arose in North America or Asia, and migrated to South America in a large scale faunal interchange. Other examples of large scale immigration and faunal turnover have been documented for extant faunas in Asia and Europe, as in the immigration of horses, camels and dogs into Eurasia from North America. Similar events of faunal turnover are well documented for extinct faunas throughout the world at various times in the past; for example extinct camels, hipparions, and canids arose in North America and migrated to the Old World during the Late Miocene. Thus, large scale migrations have played an important role in the development of regional faunas through time, throughout the world. Migrations can also be expressed as extensions of range when climatic conditions change in both regions. In contrast, it is believed by some that extant African savanna faunas, particularly ungulates and carnivores, evolved from endemic African ancestors (for example, Andrews & Van Couvering 1975; Van Couvering & Van Couvering, 1976; Churcher, 1978; Coppens, 1994). Thus, African mammals have widely been seen as having their evolutionary roots within Africa. This hypothesis holds that during the Late Miocene and Pliocene, the climate of northern and eastern Africa underwent a progressive ‘drying out’, because of the formation of the rift valleys with a reduction in annual rainfall and consequent change from dense forest to open woodland and savanna environments. In conjunction with this, endemic African faunas evolved adaptations for inhabiting open country, and in particular many ungulates evolved adaptations for grazing in open grasslands. However, current paleontological, paleobotanical, and paleoclimatological evidence can be interpreted as pointing to a more complex situation, in which much of the current African savanna fauna did not evolve in situ, but

Paleoecology of the Pikermian Biome

migrated from more northerly latitudes, replacing endemic African species. The implication is that many modern African savanna dwelling mammals originated not from forest dwelling African Miocene relatives, but rather from the Pikermian Biome which was located north of Africa. Maglio (1978) and Savage (1978) indicate such movements of taxa into Africa. We suggest that with the drying out of Africa, large mammals from the sclerophyllous woodland of more northern climates migrated into Africa (simply extending their range). Animals of the Pikermian Biome would have possessed a number of exaptations that would have allowed them to easily adapt to the modern savanna environment of Africa. The hypothesis explicitly states that later Miocene faunas such as Pikermi or Samos should be more closely related to modern African taxa than Middle and Late Miocene African faunas. As a result, a major component of the modern large mammal African savanna faunas (e.g. Masai Mara, Serengeti, and Tsavo in Kenya; and Kruger in the Republic of South Africa; and Virunga in Zaire), evolved from mammal species of this Pikermian Biome (Fig. 22.1). The temporal climatic changes of such a model are as follows. (A) During the early Miocene many of the temperate Eurasian regions were laurophyllous woodlands and forests which were replaced by the sclerophyllous evergreen woodlands. The laurophyllous woodlands survive today in regions of the Canary Islands. This botanical change from laurophyllous to scelophyllous was the result of increasing seasonally and aridity which was taking place during the Miocene. (B) African stalks of mammalian megafauna species expand their range into Eurasia during the early and Middle Miocene. This expansion from the African tropical forests and woodlands to laurophyllous Eurasian ones was possible. (C) In Eurasia, the African elements mix with endemic taxa and together they evolve into new clusters of species which became fully adapted for the sclerophyllous evergreen woodlands. Their adaptations were exaptations for the modern African savanna. During the Pliocene, the Pikermian taxa contributed a foundation for the newly formed African savanna environment and expanded their range into Africa. In a diVerent perspective, this expansion of Pikermian taxa into the savannas of Africa can be envisioned as lineage sorting bias, such that lineages with Pikermian roots have become relatively (though not necessarily absolutely) more frequent and lineages with deep African roots relatively less frequent (Werdelin 1998, personal communication). This expansion of Pikermian taxa into the savannas of Africa can gain strength using biogeographic and phylogenetic information. We predict that where data are comparable in age from Africa and the Pikermian Biome (i.e. 10–7 Ma), Pleistocene and extant African taxa should be more closely related to Pikermian forms than to endemic African species.

447

Palaeoenvironments: mammalian evidence

448

[Figure 22.1] Proposed model of some of the climatic changes through time: (A) Archaic African rain forest 20 Ma. (B) During the early Miocene many of the temperate Eurasian regions were laurophyllous woodlands and forests which were replaced by the sclerophyllous evergreen woodlands (C) by 8 Ma. African stalks of mammalian megafauna species expand their range into Eurasia during the early and Middle Miocene (e.g. curved small arrow at 16 Ma). (C) In Eurasia, the African elements mix with endemic taxa and together they evolve into new clusters of species which became adapted for the sclerophyllous evergreen woodlands; these are the Pikermian Biome. Their adaptations were exaptations for the modern African savanna. During the Pliocene, the Pikermian taxa contributed a foundation for the newly formed African savanna environment and expanded their range into Africa (large curved arrow). While this proposed emigration and immigration of African mammals was crucial in shaping modern savannas, a continuous set of lineages must have stayed endemic within Africa (straight vertical arrows).

Presently the strongest cases derive from Hominidae and GiraYdae because these families have been speciWcally researched from the point of view of determining derivations from the Pikermian species. In the Hominidae, the Wndings are published. In the GiraYdae, the results are not published. Hominids are more closely related to Pikermian forms than to endemic African species (Andrews, 1992; Dean & Delson, 1992; Andrews et

Paleoecology of the Pikermian Biome

al., 1997). These authors as well as Bonis (1998, personal communication) have suggested that Proconsul and other Middle Miocene African hominoids are less closely related to Pan, Gorilla and to Australopithecus than Ouranopithecus–Graecopithecus, which is from the Pikermian Biome. The dentition of Graecopithecus is similar to that of Hominidae. Several dental specimens of this species from Vathylakkos, Greece resemble in morphology those of Gorilla. In the type of wear, the thickness of enamel, and the slow maturation into adulthood Graecopithecus resembles Australopithecus (Bonis 1998, personal communication). Similarly, the Middle Miocene African giraYd ‘Palaeotragus’ primaevus has four ossicones which have rugose surface and sivathere-like protrusions. It also has specialized cervical vertebrae and plesiomorphic premolars. This species from Fort Ternan of Kenya is more similar to GiraVokeryx from the Chinji Formation of Pakistan than to the type of Palaeotragus from Pikermi. ‘Palaeotragus’ primaevus is less closely related to the modern African GiraVa than Palaeotragus rouenii and especially Bohlinia attica which are from the Pikermian Biome. Palaeotragus rouenii from Pikermi has two ossicones with smooth surface, displays some elongation of cervical vertebrae, and premolars which foreshadow those of the modern GiraVa. Bohlinia attica from Pikermi can be envisioned as a specialized Palaeotragus. Bohlinia displays numerous dental, cranial, and postcranial similarities to the modern GiraVa. Additional promising Pikermian taxa for similar connections to African savanna species can be derived from the following list although we presently do not elaborate. The oldest colobine monkey is Mesopithecus from Pikermi which is the best sister taxon to the modern colobines. Metailurus from Pikermi could be a sister taxon to the African Pleistocene Dinofelis. Hyaenids evolved in Eurasia not in Africa and expanded their range in Africa. Belbus beaumonti and Adcrocuta eximia of the Pikermian Biome are plausible sister taxa to the African Miocene Hyaena abronia and ultimately to the modern Hyaena hyaena and to Crocuta crocuta. In the hyaenids, the link to Eurasia is of Late Miocene age. Felids and mustelids from Eurasia also appear in Africa in the Late Miocene (Werdelin 1998, personal communication). Both African rhinoceroses, Diceros and Ceratotherium, also have their oldest representatives in the Pikermian Biome. Finally, the bovid Palaeoryx pallasi can be considered as the sister taxon for Oryx.

Acknowledgments The present study was supported by NSF IBN 9420184. We thank the mammal departments of the National Museums of Kenya, The Natural

449

Palaeoenvironments: mammalian evidence

450

History Museum, the Museum of Comparative Zoology, the American Museum of Natural History, and the National Museum of Natural History for the extant mammal comparative material. We also thank Richard Tedfrod at the American Museum of Natural History for providing tooth samples for isotopic work. We thank Mikael Fortelius and Lars Werdelin for comments.

References Abel, O. 1927. Lebensbilder aus der Tierwelt der Vorzeit, 2nd edn. Stuttgart, Jena Verlag Gustav Fisher. Andrews, P. 1992. Evolution and environment in the Hominoidea. Nature, 360, 641–6. Andrews, P. & Van Couvering, J. A. H. 1975. Paleoenvironments in East African Miocene. In Approaches to Primate Paleobiology, Szalay, F. (ed.), pp. 62–103. Basel, S. Karger Press. Andrews, P., Begun, D. R. & Zylstra, M. 1997. Interrelationships between functional morphology and paleoenvironments in Miocene hominoids. In Function Phylogeny and Fossils, Begun, D. R. (ed.), pp. 29–57. New York, Plenum Press. Axelrod, D. 1975. Evolution and biogeography of the Madrean Tethyan sclerophyll vegetation. Annals Missouri Botanical Gardens, 62, 280–334. Benda, L. 1971. Principles of the palynologic subdivision of the Turkish Neogene. Newsletters of Stratigraphy, 1 (3), 23–6. Benda, L. & Bruijn, H. de 1982. Biostratigraphic correlations in the Eastern Mediterranean Neogene. Newsletters of Stratigraphy, 11 (3), 128–35. Berggen, W. A. & Van Couvering, J. A. 1974. The late Neogene. Palaeogeography, Palaeoclimatology, Palaeoecology, 16, 1–216. Bernor, L. R. 1978. The mammalian systematics, biostratigraphy and biochronology of Maraghegh and its importance for understanding Late Miocene hominoid evolution. University of California, Los Angeles, unpublished Ph.D. Dissertation. Bernor, L. R. 1983. Geochronology and zoogeographic relationships of Miocene Hominoidea. In New Interpretations of Ape and Human Ancestry, Ciochon, R. L. & Corruccini, R. S. (eds.), pp. 21–64. Plenum Publishing Company. Bernor, L. R., Andrews, P. J., Solounias, N. & Van Couvering, J. A. H. 1979. The evolution of ‘Pontian’ mammal faunas: some zoogeographic, paleoecologic and chronostratigraphic considerations. Annales Ge´olologiques Pays Helleniques, VII International Congress on Mediterranean Neogene, Athe´ nes, 1, 81–9. Bernor, R. L. & Franzen J. L. 1997. The equids (Mammalia, Perissodactyla) of Dorn-Du¨rkheim 1 (Germany, Rheinhessen). Courier Forschungs-Institut Seckenberg, 197, 117–86. Bernor, R. L., Kovar-Eder, J., Lipscomb, D., Ro¨gl, F., Sen, S. & Tobien, H. 1988. Systematic, stratigraphic, and paleoenvironment contexts of Wrst-appearing Hipparion in the Vienna Basin, Austria. Journal of Vertebrate Paleontology, 8, 427–52. Bernor, R. L., Kovar-Eder, J., Suc, J.-P. & Tobien, H. 1990. A contribution to the evolutionary history of European Late Miocene age hipparionines (Mammalia:

Paleoecology of the Pikermian Biome

Equidae). Pale´obiologie Continetale, Montpellier, 17, 291–309. Bernor, R. L., Solounias, N., Swisher III, C. C. & Van Couvering, J. A. 1996. The correlation of the classical ‘Pikermian’ mammal faunas – Maragheh, Samos, and Pikermi, with the European MN unit system. In The Evolution of the Western Euraasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 137–54. New York, Columbia University Press. Bonis, L. de, Bouvrain, G., Geraads, D. & Koufos, G. 1992. Diversity and paleoecology of Greek late Miocene mammalian faunas. Palaeogeography, Palaeoclimatology, Palaeoecology, 91, 99–121. Cerling, T. E., Harris, J. M., MacFadden, B. J., Leakey, M. G., Quade, J., Eisenmann, V. & Ehleringer, J. R. 1997. Global vegetation change through the Miocene/Pliocene boundary. Nature, 389, 153–8. Churcher, C. S. 1978. GiraYdae. In Evolution of African Mammals, Maglio, V. J. & Cooke, H. B. S. (eds.), pp. 509–35. Cambridge, Massachusetts, Harvard University Press. Coppens, Y. 1994. The east side story: the origin of humankind. ScientiWc American, 270, 88–95. Crusafont-Pairo, M. 1950. La cuestion del llamado Meotico espaniol. Arrahona (Sabadell), 1–5. Dean, D. & Delson, E. 1992. Second gorilla or third chimp? Nature, 359, 676–7. Dorofeyev, P. I. 1966. Flora of the Hipparion epoch. International Geological Reviews, 8, 1109–17. Estes, R. 1971. Social organization of African Bovidae. In The Behaviour of Ungulates and its Relation to Management, Geist, V. & Walter, F. (eds.), pp. 166–205. Moreges, Switzerland, International Union for Conservation of Nature and Natural Resources. Gaudry, A. 1862–7. Animeaux Fossiles et Ge´ologie de l’Attique. Martinet, Paris, pp. 476. Givulescu, R. & Florei, N. 1960. Die fossile Flora von Sinersig, Ruma¨nien. Geologia, 9, 799–813. Gregor, H.-J. 1982. Die jungetertiaeren Floren Suddeutschlands. Stuttgart, Enke Verlag. Guernet, C., Keraudren, B. & Sauvage, J. 1976. La se´rie ‘Levantine’ du cap Phocas (ıˆle de Kos, Dode´cane`se, Gre`ce): Stratigraphie, palynologie et pale´oe´cologie. Revue de Micropale´ontologie, 19 (2), 61–73. Hayek, L.-A., Bernor, R. L., Solounias, N. & Steigerwald, P. 1992. Preliminary studies of hipparionine horse diet as measured by tooth microwear. In Bjo¨rn Kurte´n – A Memorial Volume, Forste´n, A. Fortelius, M. & Werdelin, L. (eds.), pp. 187–200. Annales Zoologici Fennici volume 28. Helsinki, Finland, Finnish Zoological Publishing Board. Ioakim, Ch. & Solounias, N. 1985. A radiometrically dated pollen Xora from the Upper Miocene of Samos Island, Greece. Revue de Micropale´ontologie, 28 (3), 197–204. Janis, C. M. 1982. Evolution of horns in ungulates: ecology and paleoecology. Biological Reviews, 57, 261–318. Janis, Ch. 1989. A climatic explanation for patterns of evolutionary diversity in ungulate mammals. Paleontology, 32, 463–81. Kovar-Eder, J. 1987. Pannonial (Upper Miocene) vegetational character and climatic inferences in the Central Paratethys area. Annalen des Naturhistorischen Museums Wien, Serie A, 88, 117–29.

451

Palaeoenvironments: mammalian evidence

452

Kurte´n, B. 1952. The Chinese hipparion fauna. Societas Scientiarium Fennica. Commentationes Biologicae, 13, 1–82. Kurte´n, B. 1971. The Age of Mammals. London, Weindelfeld and Nicolson. Leopold, E. B. 1969. Late Cenozoic. In Aspects of Palynology, Tschudy, R. H. & Scott, R. A. (eds.), pp. 377–438. New York, Wiley-Interscience. MacFadden, B. J. 1992. Fossil Horses. New York, Cambridge University Press. Maglio, V. J. 1978. Patterns of faunal evolution. In Evolution of African Mammals, Maglio, V. J. & Cooke, H. B. S. (eds.), pp. 603–19. Cambridge, Massachusetts, Harvard University Press. Matthew, W. D. 1926. The evolution of the horse. Quarterly Review of Biology, 1, 139–85. Orgetta, M. 1979. Erste Ergebnisse einer palynologischen Untersuchung der Lignite von Pikermi/Attica. Annales Ge´olologiques Pays Helleniques, VII International Congress on Mediterranean Neogene, Athe´nes, 2, 909–21. Osborn, H. F. (1910). The Age of Mammals. New York, MacMillan. Quade, J., Solounias, N. & Cerling, T. E. 1994. Stable isotopic evidence from paleosol carbonates and fossil teeth in Greece for forest or woodlands over the past 11 Ma. Palaeogeography, Palaeoclimatology, Palaeoecology, 108, 41–53. Raven, P. H. 1971. The relationship between ‘Mediterranean’ Xoras. In Plant Life in South-West Asia, Davis, P. H., Harper, P. C. & Hedge, I. C. (eds.), pp. 119–34. Botanical Society Edinburgh. Salitova, I. & Ramishvili, Sh. 1984. Climate and Xora of the Neogene of Western Georgia (USSR). Pale´ontologiques Contributions Montpellier, 14, 423–32. Sauvage, J. 1977. Les e´tudes palynologiques du Ne´oge´ne et du Quaternaire en Gre´ce et leurs applications a` la ne´otectonique des Helle´nides (Corinthie, Eube´e, Be´otie, Phocide et Attique). Bulletin Societe´ ge´ologique de France, 19, (3), 695–700. Savage, R. J. G. 1978. Carnivora. In Evolution of African Mammals, Maglio, V. J. & Cooke, H. B. S. (eds.), pp. 249–67. Cambridge, Massachusetts, Harvard University Press. Schaller, G. 1967. The Deer and the Tiger. Chicago, The University of Chicago Press. Simpson, G. G. 1951. The Major Features of Evolution. New York, Oxford University Press. Simpson, G. G. 1953. Horses. New York, Columbia University Press. Solounias, N. 1981a. Mammalian fossils of Samos and Pikermi. Part 2. Resurrection of a classic Turolian fauna. Annals of Carnegie Museum, 50 (8), 231–70. Solounias, N. 1981b. The Turolian fauna from the island of Samos, Greece. Contributions to Vertebrate Evolution, 6, 1–232, 42 tables, 67 Wgures, 1 map. Solounias, N. & Dawson-Saunders, B. 1988. Dietary adaptations and paleoecology of the late Miocene ruminants from Pikermi and Samos in Greece. Palaeogeography, Palaeoclimatology, Palaeoecology, 65, 149–72. Solounias, N. & Hayek, L.-A. C. 1993. New methods of tooth microwear analysis and application to dietary determination of two extinct antelopes. Journal of Zoology, London, 229, 421–45. Solounias, N. & Moelleken, S. M. C. 1992. Dietary adaptation of two goat ancestors and evolutionary considerations. Geobios, 6 (25), 797–809. Solounias, N. & Moelleken, S. M. C. 1993a. Determination of dietary adaptations of extinct ruminants through premaxillary analysis. Journal of Mammalogy, 74, 1059–74.

Paleoecology of the Pikermian Biome

Solounias, N. & Moelleken, S. M. C. 1993b. Determination of dietary adaptation in an archaic antelope through tooth microwear and premaxillary analysis. Lethaia, 26, 261–8. Solounias, N., Moelleken, S. M. C. & Plavcan, J. M. 1995. Predicting the diet of extinct bovids using masseteric morphology. Journal of Vertebrate Paleontology, 15, 795–805. Solounias, N., McGraw, W. S., Hayek, L.-A. C. & Werdelin, L. (in press). The paleodiet of GiraYdae. In The Ruminants: Past, Present and Future, Vrba, E. (ed.). New Haven, Yale University Press. Solounias, N., Teaford, M. & Walker, A. 1988. Interpreting the diet of extinct ruminants: the case of a non-browsing giraYd. Paleobiology, 14 (3), 287–300. Sondaar, P. 1971. The Samos Hipparion. Koninklijke Nederlandse Akademie van Wetenschappen, Series B, 74 (4), 417–41. Takhajan, A. 1957. On the origin of the temperate Xora of Eurasia. Botanic Zurnal (Moscow and Leningrad), 42, 935–53 [in Russian with English summary]. Thenius, E. 1959. Handbuch der Stratigraphichen Geologie, Tetria¨r. Stuttgart, Ferdinand Enke Verlag. Thenius, E. 1966. Die Vorgeschichte der Einhufer. Sonderdruk aus Saugetierkunde, 31, 150–71. Thenius, E. 1969. Phylogenie der Mammalia. Berlin, Walter de Gruyter & Co. Udvardy, M. D. F. 1978. World Biogeographical Provinces (a map). UNESCO project 8. The Coevolution Quarterly, Box 428, Sausalito, California, 94966. Van Couvering, J. A. H. & Van Couvering, J. A. 1976. Early Miocene mammal fossils from East Africa. In Human Origins, Isaac, G. Ll. & McCown, E. R. (eds.), pp. 155–207. Menlo Park, California, A Stapes Press Book. Velitzelos, E. & Gregor, H.-J. 1985. Neue pala¨eoXoristische Befunde im Neogen Griechelands. Documenta Naturae Mu¨nchen, 25, 1–4. Velitzelos, E. & Gregor, H.-J. 1986. Gelologische daten zu den fossilfu¨ hrenden Fundstellen Lava, Prosilion, und Likudi (Griechenland) nebst Bemerkungen zu deren Frucht und SamenXoren. Documenta Naturae Mu ¨ nchen, 29, 34–40. Velitzelos, E. & Knoboloch, E. 1986. Die Plioza¨ne Flora von Skoura bei Sparta aus dem Peloponnes (Griechenland). Documenta Naturae Mu ¨ nchen, 29, 21–8. Webb, S. D. 1983. The rise and fall of the late Miocene ungulate fauna in North America. In Coevolution, Nitecki, M. H. (ed.), pp. 267–306. Chicago, University of Chicago Press. Weerd, A. van de 1983. Palynology of some upper Miocene and Pliocene Formations in Greece. Geologsches Jahrbuch, Stuttgart, B. H., 48, 1–62. Weidmann, M., Solounias, N., Drake, R. E. & Curtis, G. H. 1984. Neogene stratigraphy of the Mytilini Basin, Samos Island, Greece. Geobios, 17 (4), 477–90. Willman, R. 1980. Die Alterstellung kontinentintaler Neogen ablagerungen in der ¨ga¨is (Rhodos und Kos/Dodekanes, Dacta/Su su ¨ dostlichem A ¨ dwestanatolien). Newsletters in Stratigraphy, 9/1, 1–18. Willman, R. 1983. Neogen und jungtertia¨re Entwicklung der Insel Kos (A¨ga¨is, Griechenland). Geologisches Rundshau, 72/3, 815–60. Woodburne, M. O. 1989. Hipparion horses: a pattern of endemic evolution and intercontinental dispersal. In The Evolution of Perissodactyls, Prothero, D. R. & Schoch, R. M. (eds.), pp. 197–233. New York, Oxford.

453

23 Vicariance biogeography and paleoecology of Eurasian Miocene hominoid primates Peter Andrews and Raymond L. Bernor

Introduction A characteristic of Eurasian Miocene hominid primates is their patchy distribution, both in time and space. It is exceptional to have a concentration of more than 140 specimens from a single locality, such as occurs at Rudaba´nya (Begun & Kordos, 1997). However, Rudaba´nya, like most Eurasian localities, represents a restricted stratigraphic interval. We never have well developed stratigraphic sequences of abundantly represented evolving hominid lineages comparable to rodent assemblages, or large samples of a taxon occurring among dozens of localities as for instance with late Miocene hipparionine horses. The only Eurasian hominid stratigraphic sequences of any length are those from Spain (c. 12–9.5 Ma; Garce´s et al., 1996), Italy (Rook et al., this volume) and the extraordinary record from the Siwaliks (12.5–7 Ma; Kappelman et al., 1996). Despite this patchy concentration, there is a relatively great diversity of generic-level lineages of higher primates which exhibit an interesting, albeit confusing, array of primitive characters (McCrossin & BeneWt, 1997) and clear homoplasies in both the cranial and postcranial complexes (Begun & Kordos, 1997). Moreover, the fossil record is vastly improved compared to 30 years ago (Pilbeam, 1996), and this record allows meaningful comparison of a variety of hominid phylogenetic hypotheses. In recent years there have been a number of new competing hypotheses published which implicitly propose phylogenetic relationships between fossil and extant species. These phylogenetic reconstructions have been based on morphological evidence without regard to geographical or ecological factors that regulate species dispersion and evolution, and it is our intention here to investigate these latter aspects of Eurasian Miocene hominoid evolution. We do not address the varying methodologies used in erecting these various phylogenies, and we wish to be quite clear that we are not attempting any taxonomic revision in this paper. We provide an updated biochronologic and paleogeographic context as a backdrop to our evaluation of the various phylogenetic hypotheses. We will apply present knowledge of the past biogeographic distributions and ecological associations to the phylogenies as a test of their possible signiWcance and validity.

Vicariance biogeography

Stratigraphic and chronologic background We present here an updated summary of Western Eurasian Miocene faunas (Lindsay et al., 1989; Bernor et al., 1996a). This includes a chronology of the principal characterizing mammal faunas (Steininger et al., 1996; Krijgsman et al., 1996; Daams et al., in prep. (pers. commun.); Lunkka et al., this volume) and all the major catarrhine faunas from the three regions under consideration. The MN unit system (Mein, 1975, 1979, 1989) has itself undergone a major philosophical reconsideration (Fahlbusch, 1991; Bernor et al., 1996a) and will undoubtedly continue to do so in the future, in particular in relation to long-distance biochronologic correlations, which are confounded by regional provinciality due to climatic and environmental diVerences (Bernor et al., 1996a). Provincial mammalian time scales need to be further developed in Spain (Garce´s et al., 1996; Krijgsman et al., 1996), the Central and Eastern Paratethys (Ro¨gl & Daxner-Ho ¨ ck, 1996) and Southeast Europe and Southwest Asia (Bernor et al., 1996c), and periodically evaluated and integrated into a Western Eurasian time scale (Steininger et al., 1989, 1996). Figures 23.1–23.3 present this provincial correlation and indicate those taxa known from each catarrhine-bearing locality, and we summarize these results below by successive MN units.

Late Early Miocene MN 5 The base of MN 5 is variously calibrated as between 17.0 and 16.5 Ma (Steininger et al., 1996), 17.26 Ma (Krijgsman et al., 1996), or most recently 16.0 Ma (Daams et al., this volume). Bernor (in Steininger et al., 1996) has argued that the base of MN 5 corresponds with the terminal Burdigalian regression, c. 17–16.5 Ma, an interval of global sea-lowering (Haq et al., 1987) and expansion of continental margins which accommodated mammalian migration between Africa and Eurasia (Mein, 1975, 1979, 1989; Bernor et al., 1989). Catarrhines are amongst the immigrants into Western Eurasia during MN 5. Two species of Pliopithecus occur in Europe: P. antiquus in Western Europe (Fig. 23.1) and P. platyodon in Central Europe (Fig. 23.2). There is a single specimen of a large bodied hominid from Engelswies, southern Germany (Heizmann, in prep.), referred by Andrews et al. (1996) to ?Griphopithecus (Fig. 23.2).

455

Palaeoenvironments: mammalian evidence

456

[Figure 23.1] Miocene mammal faunas of Western Europe (c. 17–5.3 Ma). Localities with radiometric or magnetostratigraphic dates are shown with ** together with letter notations for presence of primate taxa as indicated. Ages shown in square brackets signify alternative dates determined by Krijgsman et al. (1996) and Daams et al. (this volume).

Vicariance biogeography

457

[Figure 23.2] Miocene mammal faunas of Central Europe (c. 17–5.3 Ma). Localities with radiometric or magnetostratigraphic dates are shown with ** together with letter notations for presence of primate taxa as indicated. Ages shown in square brackets signify alternative dates determined by Krijgsman et al. (1996) and Daams et al. (this volume).

Palaeoenvironments: mammalian evidence

458

[Figure 23.3] Miocene mammal faunas of Italy, Greece and Southwest Asia (c. 17–5.3 Ma). Localities with radiometric or magnetostratigraphic dates are shown with ** together with letter notations for presence of primate taxa as indicated. Ages shown in square brackets signify alternative dates determined by Krijgsman et al. (1996) and Daams et al. (this volume).

Vicariance biogeography

Middle Miocene MN 6 Bernor (in Steininger et al., 1996) has correlated the base of MN 6 with the terminal Langhian regression c.  / = 14.8 Ma. This age determination draws support from Sen’s (1996) magnetostratigraphic correlation of 15.2 Ma for Sansan (France), the type MN 6 locality, and Bernor & Tobien’s (1990) correlation of the Pas¸alar (Turkey) fauna c. 15 Ma. Alternatively, Krijgsman et al. (1996) have calibrated the Wrst occurrence of MN 6 small mammal faunas in Spain as being 13.75 Ma (Caltayud-Daroca Basin), which has been maintained in subsequent revision by Daams et al. (pers. commun.). Catarrhine primates achieve greater diversity in MN 6. The Pliopithecidae are represented by two genera, Plesiopliopithecus (P. auscitanensis [WE]; P. lockeri [CE]) and Pliopithecus (P. antiquus [WE and CE]; P. vindobonensis [CE]). Harrison (in Andrews et al., 1996) believes that the diversity of European pliopithecids likely reXects multiple lineage geographic extension into Europe from, perhaps, North Africa. Recently, Harrison & Gu (in review) have presented new evidence that the Pliopithecidae are represented in the early Miocene (MN 4 = c. 18 Ma) Chinese locality of Sihong. Harrison & Gu further argue that these primates are members of a primitive subfamily of pliopithecids, the Dionysopithecinae. It would appear that the pliopithecids and hominoids diverged as early as the late Oligocene, and it was in the earlier Miocene that members of the Pliopithecidae extended their range from Africa to South and East Asia. The diversity of pliopithecids in the European middle Miocene may reXect either the entry of multiple lineages from the east, or the rapid in situ evolution as species Wlled a variety of frugivorous and folivorous niches previously unoccupied in Eurasia (Kay & Ungar, 1997). Speciation may have been facilitated by discontinuities of continental biogeographic connections typical of Western Eurasia’s tectonically active and paleogeographically varied landscape. Large bodied hominids are represented solely by the genus Griphopithecus and are found only in Turkey (Pas¸alar) and Austria (Klein Hadersdorf) at this time.

MN 7 + 8 De Bruijn et al. (1992) have argued that MN 7 and MN 8 should be combined as a single unit because they lack any clear biostratigraphic basis for distinguishing them as individual units. Steininger et al. (1996) adopt Ro¨gl &

459

Palaeoenvironments: mammalian evidence

460

Daxner-Ho ¨ ck’s (1996) estimate for the lower boundary of this unit as being 12.5 Ma. This corresponds closely with Krijgsman et al.’s correlation of the base of MN 7 + 8 in Spain (Calatayud-Daroca Basin) as being 12.75 + / − 0.25 Ma. Catarrhine diversity remains high during MN 7 + 8 with the Pliopithecidae being represented by Plesiopliopithecus (P. rhodanica) and Crouzeliinae indet. in Western Europe and Pliopithecus (P. antiquus) in Central Europe. The large bodied hominid genus Dryopithecus Wrst appears in MN 7 + 8 and it remains unclear whether this represents an evolutionary derivation from indigenous Griphopithecus populations or is a separate European immigration. What is clear is that this is a rather advanced hominid which is variously related as the sister-taxon of the African and Asian great apes (Harrison & Rook, 1997), the Ponginae (Moya`-Sola` & Ko¨hler, 1993; Schwartz, 1990) or the African great ape–human clade (Begun & Kordos, 1997; Begun et al., 1997). Dryopithecus is represented by D. carinthiacus in Central Europe and 2–3 species in Western Europe, D. fontani and D. laietanus [arguably, D. crusafonti (Begun, 1992), both Spain only].

Late Miocene MN 9 The base of the marine late Miocene (= base of the Tortonian stage; 11.2 Ma) closely corresponds with the base of MN 9 marked by the Wrst appearance of the tridactyl equid ‘Hipparion’: the so-called ‘Hipparion Datum’. This correlation can be recognized across Eurasia and Africa (Woodburne et al., 1996a) with a slight discrepancy in age ranging from 11.2 Ma in Middle Europe (Ro¨gl & Daxner-Ho ¨ ck, 1996), to 11.1 Ma in Spain (Krijsman et al., 1996; Garce´s et al., 1996 [Valle´s-Penede´s]), to 10.8 Ma in the Siwaliks (Pilbeam et al., 1996) and Turkey (Lunkka et al., this volume). This discrepancy is slight compared to its status 10 years ago when the European datum was calculated as being 12.5 Ma (Berggren & Van Couvering, 1974) and the Siwalik one was 9.5 Ma (Barry et al., 1985). The revision is due to redating critical deposits at Ho¨wenegg, Germany (c. 10.3 Ma; Swisher, 1996; Woodburne et al., 1996b), recalibration of the magnetostratigraphic time scale (Cande & Kent, 1995; Steininger et al., 1996), and a critical reevaluation of the entire ‘Hipparion Datum’ concept (Bernor et al., 1988b, 1989; Sen, 1989, 1996; Garce´s et al., 1996). The Pliopithecidae decline substantially in the Vallesian, occurring in Western Europe (P. antiquus) and Central Europe (Pliopithecidae gen. and sp. indet.). A new pliopithecid genus, Anapithecus Wrst occurs in MN 9. A.

Vicariance biogeography

hernyaki is represented from the Central Paratethys locality of Rudaba´nya (Hungary; Kordos, 1991) and, to a lesser extent from Go¨tzendorf (Ro¨gl et al., 1993; Andrews et al., 1996). The French locality of Priay II Upper has a form originally referred to Pliopithecus priensis by Welcomme et al. (1991). Dryopithecus reaches its greatest geographic and taxonomic extent during this interval, occurring at a number of Spanish localities (D. laietanus and D. crusafonti (Begun, 1992); re: Fig. 23.1; Agustı´ et al., 1996) as well as a number of Central European localities (D. carinthiacus and, arguably, D. brancoi; Begun & Kordos, 1993; Fig. 23.2). Garce´s et al. (1996) have argued that Central and Western European MN 9 faunas have a high faunal similarity. However, current study and comparison of the Rudaba´nya fauna with the Spanish MN 9 faunas suggests some biogeographic vicariance in several of the small and large mammal lineages (several authors in Bernor & Kordos (eds.), in prep.). Recent work at Sinap, Turkey (Kappelman et al., 1996), has yielded a second partial cranium with associated postcranial material of Ankarapithecus meteai from Middle Sinap (Alpagut et al., 1996). Both this specimen, and the specimen earlier reported by Ozansoy (1965; loc. 8b) are from MN 9 correlative horizons.

MN 10 This is a unit that is well characterized and deWned in Spain, elusive in Central Europe, and virtually unrecognizable in a biochronologic sense in Southeast Europe/Southwest Asia. The type area is the Valle`s-Penede`s region in Spain and there is good magnetostratigraphic control suggesting an age of base MN 10 as being 9.7 Ma (Krijgsman et al., 1996; Garce´s et al., 1996). The alternative correlation of 9.5 Ma based on North African, Eastern Mediterranean (Sen, 1996; Steininger et al., 1996) and Central Paratethys estimates compares closely with the Spanish calibration. MN 10 witnesses the extinction of European Pliopithecidae and Dryopithecus. This shift was greatest in Southeast Europe and Southwest Asia and least in Central Europe (Bernor et al., 1996c; Fortelius et al., 1996). In Spain, the Pliopithecidae occur up to nearly the end of MN 10 where it is recorded at Terrassa as Crouzeliine gen. and sp. nov. by Harrison (in Andrews et al., 1996; also Moya`-Sola`, in prep.). Dryopithecus likewise occurs abundantly in Spain (D. laietanus) and potentially Germany (indeterminate ages; Andrews et al., 1996: Table 12.7, p. 198). The large hominid Graecopithecus (= Ouranopithecus) is known from Greece (de Bonis & Koufos, 1996).

461

Palaeoenvironments: mammalian evidence

462

MN 11 The base of MN 11 has been calibrated in the type area of Teruel (La Gloria section) as corresponding to an age of 8.7 Ma (+ / − 0.1 Ma; Krijgsman et al., 1996). Garce´s et al. (1996) have indicated that actually the uppermost Vallesian (MN 10) of the Valle`s-Penede`s is 9.0 Ma and does not overlap chronologically (and presumably faunally) with the early Turolian of the Teruel Basin where the 8.7 Ma base of MN 11 is established. Bernor et al. (1996b) and Steininger et al. (1996) estimated the base of MN 11 to be c. 9.0 Ma based on the chronology of the ‘typical’ Turolian megafauna occurring in Lower Maragheh, Iran. However, there is no systematic basis whatsoever for correlating fauna of Lower Maragheh with the type Turolian of Spain. If we accept the Spanish type section correlation of MN 11 as being 8.7 Ma, that would give a chronologic correlation of the lower portion of the Lower Maragheh fauna with MN 10 and further indicate a precocious development of Southeast European–Southwest Asian Pikermian faunas compared to those of Western Europe (Bernor, 1983, 1984; Bernor et al., 1979, 1996b; de Bonis & Koufos, 1996; Fortelius et al., 1996) MN 11 lacks the earlier pliopithecid–dryopithecine fauna common to Central and Western Europe with the exception of some localities of questionable age from Germany. However, the record of Oreopithecus begins and is documented at several localities in Italy (Fig. 23.3 [Rook, pers. commun.]; Harrison & Rook, 1997). Koufos (1993) and de Bonis and Koufos (1996) report a new specimen of Ouranopithecus (= Graecopithecus) from Nikiti, which is believed to be early MN 11 in age. The colobine primate Mesopithecus pentelici makes its Wrst appearance in Greece and Iran at this time. Western European localities are devoid of catarrhine primates during MN 11.

MN 12 The base of MN 12 is deWned in the Je´car-Cabriel Basin, Spain, as being 7.5 + / − 0.1 Ma (Krijgsman et al., 1996). Bernor et al. (1980, 1996c) and Steininger et al. (1996) estimated MN 12 as being c. 8.2 Ma based on the correlation of the Wrst occurrence of Hipparion prostylum in Middle Maragheh, and its species-level correlation with the French MN 12 fauna, Mt. Luberon (= Cucuron). If we accept the Spanish correlation, this results in Middle Maragheh and Pikermi (at the very least) being shifted downward in their chronologic correlation to MN 11. Uppermost Maragheh and the bulk of the Samos quarries (Q1–5) would still remain in MN 12. Other similar changes in chronologic correlation can be expected for Greek, Turkish and Afghanistan faunas.

Vicariance biogeography

MN 12 records the maximum extension of Pikermian megafaunas across Eurasia and Africa (Bernor et al., 1996b; Fortelius et al., 1996). Concommitant with this is a catarrhine ‘vacuum’ in Western and Central Europe, with the exception of the one Hungarian fauna of Hatvan which records an inXux of Pikermian faunal elements into the Central Paratethys (Kordos, pers. commun.). Oreopithecus bambolii continues its record in Italy as does Mesopithecus pentelici in Greece and Southwest Asia.

MN 13 The base of MN 13 has been correlated with the base of the Messinian by Bernor et al. (1996a), Ro¨gl & Daxner-Ho ¨ ck (1996) and Steininger et al. (1996). Agustı´ (in Steininger et al., 1996) has argued for a younger base of MN 13 which is 6.7 Ma (Figs. 23.1–23.3 here). All hominid primates disappeared from Eurasia by the base of MN 13 (= 7.1 Ma), apparently with the exception of the Hylobates and Pongo lineages in subtropical Southeast Asia. This extinction would appear to also extend to South (Sivapithecus) and East Asian (Lufengpithecus) hominid faunas. Fortelius et al. (1996) and Bernor et al. (1996c) have noted a major extinction of several ‘Pikermian’ large mammal lineages at the base of MN 13 (especially depletion of herbivores) which is believed to reXect increased seasonality. This interval also correlates closely with the carbon-isotope shift recorded by Quade et al. (1989).

Hominoid phylogeny and biogeography The procedure that we will follow here will be to present Wve morphologybased cladograms and evaluate their various strengths and incongruencies with biogeographic and paleoenvironmental data. The patterns for the branching points will be inferred where possible by commonality in the end points. If two sister-taxa share the same geographic distribution, their common ancestor will be inferred to have shared the same distribution. If they share similar ecologies, their common ancestor will likewise be assumed to have shared the same ecology. This procedure was used in an earlier phyloecologic analysis (Andrews, 1982) in which it was shown that, based on the cladistic analysis of Delson & Andrews (1975), the common ancestor of the hominoid primates occupied tropical forests whereas the common ancestor of the cercopithecid monkeys occupied non-forest environments. The Hominidae, by which is meant the great ape and human clade, including the fossil genera Sivapithecus and Gigantopithecus

463

Palaeoenvironments: mammalian evidence

464

(Andrews, 1982), was found to be associated with open forest/woodland environments. In addition to the Wve morphology-based cladograms, we present a sixth cladogram in which we attempt to combine data from the Wrst Wve. This we will term a concordance cladogram, and it represents another alternative which incorporates what we consider to be the best of the Wve morphology cladograms. For example, we accept Schwartz’s (1990) concept of a greater orang utan clade; we accept Harrison & Rook’s (1997) concept of relationship between Oreopithecus and Dryopithecus; these authors and Andrews (1992b) also propose relationship between Griphopithecus and Kenyapithecus, and this also is accepted here. Dryopithecus is considered here as a stem-hominid. The cladograms that we will use in the present analysis are based on the following branching sequence in living hominoids: gibbons–orang utan– gorilla–chimpanzee–human. This gives rise to the higher order classiWcation proposed by Goodman (1963) of Hominidae for the great ape and human clade, within which Ponginae (orang utan clade) and Homininae (African ape and human clade) are distinguished as subfamilies of Hominidae. Further divisions within Homininae are unnecessary for the scope of the present paper, but it will be assumed that this clade, at least as represented by extant species, had an African origin. This is based on the African distributions of chimpanzees and gorillas and the fact that early human ancestry in the period 5–2 Ma was also exclusively African, so that it may be inferred that the last common ancestor of the living forms lived in Africa.

Cladogram number 1 The Wrst phylogeny considered here (see Fig. 23.4) was proposed by Schwartz (1990), who identiWed an extended pongine clade including Sivapithecus, Lufengpithecus, Graecopithecus and Dryopithecus (= Rudapithecus). The entire spectrum of Eurasian hominids with the exception of Oreopitheus is thus grouped with the orang utan. Griphopithecus is not mentioned by Schwartz (1990), but its absence from the analysis, which was restricted to the pongine clade, implies its omission was deliberate and hence it is implicitly excluded from Ponginae. This scheme was followed almost exactly by Moya`-Sola` & Ko¨hler (1995) as regards the pongine clade, and in addition they positioned Afropithecus and Kenyapithecus as stemhominids, diverging before the branching point of the pongine and hominine clades (Fig. 23.4). Again no mention is made of Griphopithecus, mainly because Moya`-Sola` & Ko¨hler based their analysis on cranial characters and no cranium of Griphopithecus has yet been found.

Vicariance biogeography

[Figure 23.4] Cladogram 1, from Schwartz (1990) of Hominoidea.

Schwartz’s (1990) phylogeny linking all European and Asian middle to late Miocene large-bodied hominids with the pongine clade makes a lot of sense biogeographically (Schwartz, 1997). It assumes an African origin for the Hominoidea, with emigration of the gibbons to Asia at an unknown stage of evolution. Hominids also arise in Africa, probably from within the Kenyapithecinae (sensu Andrews, 1992b) at the end of the early Miocene or beginning of the middle Miocene (c. 17–15 Ma). Hominids extended their range into Eurasia and diVerentiated within the new Eurasian clade into both arboreal forms (Dryopithecus) and others that were probably at least partly ground dwelling (Graecopithecus, Lufengpithecus, Ankarapithecus, Sivapithecus). Speciation within the clade would have occurred vicariantly as a result of the emergence of physical and environmental barriers to genetic continuity between the various populations. This model can certainly accommodate the MN 5 immigrant form Griphopithecus as sister group to this clade and the African ape and human clade, forming a branch between the kenyapithecines and the greater Pongo-clade. Seemingly nearly simultaneous Wrst occurrences of Dryopithecus in Europe and Sivapithecus

465

Palaeoenvironments: mammalian evidence

466

in IndoPakistan at 12.5 Ma further supports this model by documenting potential vicariance of European and Asian radicles of the greater Eurasian ape-clade.

Cladogram number 2 The second phylogeny (see Fig. 23.5) is that of Andrews (1992b) which places most fossil hominoids (including Dryopithecus) as stem-hominids, i.e. branching from the hominid lineage before the divergence of the orang utan. Three tribes are recognized for the stem-hominids: Afropithecinae (Afropithecus and Heliopithecus); Kenyapithecinae (Kenyapithecus and Griphopithecus); Dryopithecinae (Dryopithecus [which includes in this context Lufengpithecus and other Chinese hominids]). These three tribes were seen as more or less distinct radiations of middle to late Miocene hominids not related directly to any extant form. Graecopithecus is grouped with Homininae and Sivapithecus (together with the now separated genus Ankarapithecus from Turkey) with Ponginae as in the Wrst phylogeny as also proposed by Begun (1992, 1993, 1994) and Pilbeam (1996). Biogeographically, this cladogram suggests that there was a stemhominid lineage with a common African–Eurasian biogeographic distribution in the late early Miocene to early middle Miocene leading to the Afropithecus–Kenyapithecus–Dryopithecus trichotomy. It further suggests that the Pongo clade (which here includes Sivapithecus–Pongo, but implicitly includes Lufengpithecus and Ankarapithecus) is a distinctly Asian clade ranging from Turkey to Southeast Asia. The Wrst recorded occurrence of this clade would be the Siwaliks, c. 12.5 Ma (Kappelman et al., 1991). This leaves Graecopithecus as a European sister-taxon of the African ape–human clade with several diVerent possible biogeographic interpretations: 1) that Graecopithecus evolved as a vicariant eastern Mediterranean species of a greater Eurasian ‘clade’; 2) that Graecopithecus had a Southeast European origin and immigrated into Africa in the late Miocene to found the African ape–human clade; 3) that Graecopithecus is a member of the African ape–human clade that arose in Africa and extended its range into the eastern Mediterranean in the early late Miocene. A fourth possiblity exists, which we will explore later, and that is Graecopithecus exhibits an extreme case of character convergence (= homoplasy) with the African hominines as a product of its adaptation to ground dwelling and hard-object feeding (Kay & Ungar, 1997). If this case were to Wnd eventual anatomical support, Graecopithecus could be viewed as being a vicariant lineage of the Eurasian greater pongine clade as suggested by Schwartz (1990).

Vicariance biogeography

[Figure 23.5] Cladogram 2, from Andrews (1992a) of Hominoidea.

Cladogram number 3 Harrison & Rook (1997) have presented quite a diVerent, and far ranging cladogram and accompanying classiWcation of hominoids (their Table VI) (see Fig. 23.6). Harrison & Rook consider Proconsulidae to be a stem group not belonging to the Hominoidea, and they also recognize Afropithecidae (Afropithecus and Heliopithecus) as a possible non-hominid family of Hominoidea. Four subfamilies of Hominidae are recognized: Kenyapithecinae (= Kenyapithecus and Griphopithecus; [also, ?Maboko hominid]); Dryopithecinae (Tribe Dryopithecini [Dryopithecus]); Tribe Oreopithecini [Oreopithecus]); Subfamily Ponginae (Pongo, Sivapithecus, Gigantopithecus, ?Lufengpithecus); Subfamily Homininae (Tribe Gorillini [Gorilla]; Tribe Panini [Pan]; Tribe Hominini [Ardipithecus, Australopithecus, Homo]; Tribe incertae sedis [Graecopithecus]). Otavipithecus is placed in Hominidae incertae sedis. Hominidae is presented as a natural group with the Kenyapithecinae being the most primitive member (for a diVerent argument see McCrossin & BeneWt, 1997). Harrison & Rook group Dryopithecus

467

Palaeoenvironments: mammalian evidence

468

[Figure 23.6] Cladogram 3, from Harrison & Rook (1997) of Hominoidea.

and Oreopithecus together as a clade which is a stem member of the Hominidae, encompassing the South and East Asian Miocene hominid radiation and the African ape and human clade. Interestingly, Graecopithecus is not seen as being either a member of the Dryopithecini or Pongini, but is placed distinctly within the African great ape–human clade on the tree. Harrison & Rook’s (1997) cladogram is biogeographically the most complicated. The cladogram could be interpreted as supporting multiple independent African biogeographic extensions into Eurasia: (1) Pliopithecids are represented by Dionysopithecus in the early Miocene of Asia (Bernor et al., 1988a); (2) an extension of Griphopithecus into Europe in MN 5; (3) an extension of an unknown African sister-taxon of Dryopithecus–Oreopithecus into Europe in the late middle Miocene, c. 12.5–11. 1 Ma; (4) an extension of a sister-taxon of the Pongo–Sivapithecus–Gigantopithecus (and implicitly

Vicariance biogeography

Ankarapithecus and Lufengpithecus) into West, South and East Asia also c. 12.5 Ma; and (5) an extension of Graecopithecus into Greece in the Vallesian (c. 11–9 Ma). Harrison & Rook’s cladogram can alternatively be interpreted so that the Kenyapithecus–Griphopithecus clade includes a primitive hominid stem species that extended its range throughout Eurasia with resulting vicariant evolution of a European Dryopithecus–Oreopithecus clade and a Pongo– Sivapithecus–Gigantopithecus [and implicitly Ankarapithecus–Lufengpithecus] clade. The divergence time of these two greater Eurasian clades would be set by the Wrst occurrence of Sivapithecus at 12.5 Ma. This interpretation Wnds support in the divergent patterns of postcranial evolution with the Dryopithecus–Oreopithecus clade evolving suspensory repertoires considered to be autapomorphic for the extant great ape–human clade by some (Begun & Kordos, 1997; McCrossin & BeneWt, 1997), and more conservative ground dwelling repertoires for the Pongo–Sivapithecus–Gigantopithecus [and Ankarapithecus–Lufengpithecus] clade. This second biogeographic hypothesis conforms in its essential content with that of Schwartz (1990) and Andrews (1992b).

Cladogram number 4 An extreme variant of this third scheme has been proposed by Cameron (1997); see Fig. 23.7. He distinguishes the Spanish and Hungarian specimens of Dryopithecus, placing the former in the pongine clade and the latter as a stem-hominid. In doing this, he resurrected the genus Hispanopithecus for the Spanish fossil ape to distinguish it from Dryopithecus, which is retained for the Hungarian and other fossils. In addition, Cameron (1997) linked Graecopithecus with the gorilla, but he did not comment further on other fossil apes. Cameron’s (1997) cladogram is in some ways the most controversial. It recognizes the similarities between Gorilla and the European fossil ape Graecopithecus as synapomorphies and links them in a single clade (Cameron, 1997). Consequently, there must be a geographical event leading to the origin of the gorilla; the Graecopithecus–African ape–human clade arose either in Greece or Africa with species extension to one or the other geographic areas. Also in this cladogram, Dryopithecus is presented as a stem-hominid with the Spanish species of Hispanopithecus (previous junior synonym for Dryopithecus) being found to have synapomorphies common with the Sivapithecus–Pongo clade from Asia. Cameron’s cladogram is similar to Schwartz’s (1990) model in that it is biogeographically consistent with a Eurasian origin for the pongine clade. The remaining European

469

Palaeoenvironments: mammalian evidence

470

[Figure 23.7] Cladogram 4, from Cameron (1997) of Hominoidea.

Dryopithecus species are recognized as being stem-hominids as in Andrews (1992b) and Proconsul is positioned as the outgroup with no other fossil apes considered.

Cladogram number 5 The Begun et al. (1997) cladogram (see Fig. 23.8) is based on 240 characters roughly subdivided between cranial and postcranial characters derived from participants in the volume Function, Phylogeny, and Fossils – Miocene Hominoid Evolution and Adaptations. Also, Begun (pers. commun.) would place Ankarapithecus as the sister-taxon of the Sivapithecus– Pongo clade. This cladogram excludes Proconsul, Afropithecus and Kenyapithecus from the Hominoidea. Begun et al. (1997) argue that the evidence suggests that: ‘Proconsul is the outgroup to an Afropithecus– Kenyapithecus clade, and all three of these taxa form a monophyletic clade that is the sister group to the hylobatids and the clade that includes fossil and living great apes and humans’. This is not the solution on their

Vicariance biogeography

[Figure 23.8] Cladogram 5, from Begun et al. (1997) of Hominoidea.

471

`

most parsimonious cladogram (Begun et al., 1997, Wgure 1), but it is one of three alternatives with the addition of one extra branching point. Oreopithecus and Dryopithecus are broadly separated cladistically, with Oreopithecus being the sister-taxon of all the Asian and African large bodied apes included in their analysis, while Dryopithecus is the sistertaxon to the extant African great ape–human clade. Begun (pers. commun.) acknowledges some uncertainty in the position of Graecopithecus and Dryopithecus on the tree. Species included in the Pongo clade decrease substantially here from most phylogenies that have been proposed to date. According to this analysis, Proconsul is considered to be a generalized catarrhine which is a sister-taxon to the Hominoidea. The stem-hominoid Afropithecus is limited to East Africa, but Kenyapithecus (and implicitly its sister-taxon Griphopithecus) have a distribution from Africa to Europe and West Asia during the MN 5–6 interval. This clade then is poised as the sister-taxon of the Eurasian–African hominoid radiation that places

Palaeoenvironments: mammalian evidence

472

Hylobates at its base. The biogeographic relationships that can be interpreted from the remainder of the cladogram are mixed. 1. Oreopithecus is an early endemic branch of the hominid clade which would have diverged from the remainder of the group early in the Middle Miocene. 2. Lufengpithecus likewise is a primitive clade that would have diverged from its sister-taxa prior to the branching of the Sivapithecus–Pongo clade, c.  12.5 Ma. 3. The Ouranopithecus clade would have diverged prior to the Wrst occurrence of Dryopithecus c. 12.5–11.1 Ma, and evolved in isolation from all other taxa from that point onward. Implicitly the same basic relationship to the Sivapithecus–Pongo clade would hold for Ankarapithecus. 4. The Dryopithecus–African ape–human clade then would have been established as early as the 12.5–11.1 Ma interval. Dryopithecus is considered by most authors, however, to be a stem-hominid for the entire great ape–human clade and their extinct sister-taxa and its close position to the African ape–human clade is inconsistent with several cladistic hypotheses. In addition, the early divergence of Oreopithecus separates it from Dryopithecus and suggests extensive homoplasy in postcranial anatomy if we accept Harrison & Rook’s (1997) interpretation of its postcranial anatomy.

Cladogram number 6 In the concordance cladogram shown in Fig. 23.9 we position Proconsul as a stem-hominoid before the branching of Hylobates. This cladogram next positions Afropithecus, Griphopithecus and Kenyapithecus as stemhominids. The next branch recognizes the Dryopithecus–Oreopithecus clade based largely on its postcranial anatomy which is adapted for suspensory behavior and unique for all known Miocene hominoids. The next branch recognizes the Lufengpithecus–Graecopithecus–Ankarapithecus– Sivapithecus/Gigantopithecus–Pongo clade as a natural group of pongine apes, which in turn is the sister-taxon of the gorilla–chimp–Ardipithecus clade. The Miocene members of this clade were partly terrestrial, not having acquired the suspensory anatomical modiWcations present in living hominids, Dryopithecus and Oreopithecus. Our concordance cladogram is strongly inXuenced by Schwartz’s (1990), Andrews’s (1992b) and Harrison & Rook’s (1997) analyses, and the congruent biogeographic and paleoecologic conclusions that we derive from them.

Vicariance biogeography

[Figure 23.9] Consensus cladogram of Hominoidea.

It suggests that stem-hominoids have an African origin. During the later early Miocene a continental connection between Africa and Eurasia allowed an extension of Dionysopithecus into South Asia, no later than 18 and 16 Ma (Bernor et al., 1988a; Harrison, in press). The Hominidae also appear to have an African origin (Afropithecus), but established themselves in Turkey and Central Europe sometime between 17 and 15 Ma. Griphopithecus is a kenyapithecine and was adapted to some degree of ground dwelling (Ersoy et al., in prep.). We pose the hypothesis that this generalized quadruped, or its sister-taxon, founded Miocene hominids in Central and Western Europe which evolved suspensory behaviors and soft-object frugivory (Dryopithecus) and folivory (Oreopithecus) appropriate for the subtropical evergreen to the slightly more temperate semi-deciduous forests that they

473

Palaeoenvironments: mammalian evidence

474

inhabited, while the pongine clade remained postcranially conservative and successfully adapted to ground dwelling. The immediate consequence of this set of relationships is to pose the hypothesis that anatomical modiWcations for suspensory behavior evolved at least three times: (1) in recent Hylobatids; (2) in Dryopithecus– Oreopithecus; (3) in Pongo. A second possibility is that it evolved twice only, in gibbons and at the time of the Dryopithecus–Oreopithecus divergence, so that it was primitively retained in orang utans and the African apes and secondarily lost in the Lufengpithecus to Sivapithecus group. A consequence of the second possibility is that Sivapithecus and the other parts of the ‘greater orang utan clade’ are not in fact related to the orang utan. If we accept that suspensory adaptations evolved three times, we have to recognize at least two immigration events founding Eurasian hominids: one for the hylobatid clade, a second for the Dryopithecus–Oreopithecus and pongine clades. A third event could separate the latter two clades if the Dryopithecus–Oreopithecus and pongine clades entered Eurasia independently, but if only two, a necessary consequence would be that Griphopithecus, with possible ground dwelling adaptations and thick enamelled teeth, could be a paraphyletic sister-taxon for both the western Dryopithecus–Oreopithecusclade and the pongine clade, so that it is possible that these adaptations could have been primitive to these two clades, as it could also have been for the African ape–human clade. Finally, this cladistic hypothesis denies the sister-taxon relationship between Graecopithecus and the African ape–human clade, their similarity being due to character convergence (Kay & Ungar, 1997). If we accept the second possibility, that suspensory locomotion evolved just twice, then again there must have been at least two immigration events into Eurasia. The Wrst would have led to the gibbons again, and the second produced the suspensory Dryopithecus–Oreopithecus clade and the orang utan. The rest of the middle to late Miocene fossil apes of Eurasia could have been derived from this, with secondary return to quadrupedal partial terrestrialism, or it could have been derived from a third and independentimmigration represented perhaps by Griphopithecus. This option is probably more likely given the earlier appearance of Griphopithecus in Europe and Turkey.

Paleoecological context There is little information on the paleoeocological context of most European hominid sites, but the climatic and environmental conditions in Europe as a whole has been documented in some detail in the present volume. We

Vicariance biogeography

summarize the main results here as they relate to the phylogeny and biogeography of Eurasian Miocene catarrhines, with particular regard to hominid vicariant evolution and biogeography. Brief mention can also be made of the signiWcance of Dionysopithecus in Asia, which represents a prehominoid immigration event possibly related to the ‘Proboscidean Datum’ (= Wrst occurrence) which is correlated with MN 4 (18–17 Ma) in Europe and potentially as early as the basal Miocene in south Asia (Bernor et al., 1988a). Continued plate tectonic encroachment along the Alpine–Himalayan orogenic system restricted the Tethys Sea connection with the Indian Ocean (Ro¨gl, this volume). As theTethys diminished, two distinct large water bodies remained, the Paleo-Mediterranean Sea in the south and west, and a greatly reduced Paratethys Sea to the north and east. Two major consequences ensue from these oceanic reductions. The large expanse of midcontinental seas connected to the Indian Ocean must have had an ameliorating eVect on climate in the surrounding areas, particularly aVecting Southern Europe and Southwest Asia. It is diYcult to predict what this eVect may have been precisely, but the great extent of subtropical forests in Southern Europe lasting even into the Pliocene in places (Thompson & Fleming, 1996; Suc et al., this volume) must surely be related to this. The reductions in subtropical forests during the middle and late Miocene followed the oceanic reduction. The other major consequence is that with the contact between the Arabian plate and the Anatolian plate, land bridges formed between Africa and Eurasia which opened and closed throughout the middle Miocene. The Tethys and Paratethys were connected at times, forming intermittent barriers to migrations, and they were also connected to the Arctic Ocean through the Turgai Strait in Western Siberia, preventing faunal interchange between Europe and Asia. Of particular signiWcance to hominoid migrations was the end-Burdigalian regression of the restricted Tethys Sea at 17–15.5 Ma. This only shortly predates the Wrst appearance of hominids in Europe, which is documented during this interval: Griphopithecus at Engelswies in southern Germany (MN 5). Griphopithecus appears to be related to East African species of Kenyapithecus, suggesting derivation from that continent (Andrews, 1992b). Thereafter, Griphopithecus is known from Pas¸alar in Turkey. During the early middle Miocene, climates were warm with wet subtropical forests indicated for much of southern Europe through to China (KovarEder et al., 1996; Suc et al., this volume). The habitats were not uniform, however, and there is some indication of slightly diVerent forest associations between South central/Western Europe and the eastern Mediterranean region of Turkey. For example, the environments reconstructed for the eastern region show a more seasonal form of forest, which was probably

475

Palaeoenvironments: mammalian evidence

476

more open in structure and more deciduous (for example at Pas¸alar), than the wet subtropical forests of Central and Western Europe at this time (Andrews, 1990, 1992a, 1006; de Bonis et al., 1992). Griphopithecus at Pas¸alar exhibits the morphology of a generalized quadruped with some similarities to terrestrial monkeys in phalangeal morphology (Ersoy et al., in press), and it is similar in this respect both to Kenyapithecus (BeneWt & McCrossin, 1995) and to Sivapithecus (Pilbeam et al., 1990). This suggests that the association between partly terrestrial hominids and seasonal forests was widespread in the middle Miocene, in contrast to the association of pliopithecids with arboreal suspensory adaptations and wetter, less seasonal and probably evergreen subtropical forests during the same period (Andrews et al., 1997). The eastern environments became drier and more open through the middle Miocene and into the late Miocene (Fortelius et al., 1996), with persistence of hominids having terrestrial adaptations and large thick-enamelled teeth for eating hard fruit diets, until just after 10 Ma (Alpagut et al., 1996). Little is known of the environment of Ankarapithecus in Turkey, but some unpublished information on the postcranial bones provides evidence that this fossil ape was adapted for ground living. For example the radius is extremely robust, and a femur provisionally attributed to this genus is more robust even than gorilla femora. The masticatory apparatus of Ankarapithecus was adapted for heavy chewing, with massively built and robust jaws, large molars and premolars with Xattened crowns and thick enamel, and robust zygomatic bones and processes for insertion of muscles of mastication. In addition, the faunas associated with Ankarapithecus are precocious Pikermian open country fauna (Kappelman et al., 1996; Lunkka et al., this volume; Solounias et al., this volume). Ankarapithecus’s anatomy, faunal association and environmental context all suggest that it was a habitually terrestrial form that was dependant on hard-object frugivory, and it occupied a distinct niche in seasonal environments where food sources would have been patchier and both quantity and quality of food sources would have been strongly seasonal. In Western and Central Europe, wet subtropical conditions were maintained at least until early in the late Miocene (MN 10). Dryopithecus appeared late in the middle Miocene (MN 7–8), and where evidence exists for associated environment, for example at Rudaba`nya (Bernor & Kordos, in prep.), wet evergreen subtropical forests are indicated. Dryopithecus has been interpreted as having been a generalized suspensory hominid with a powerful grasping pollex (Begun, 1993) well adapted for its reconstructed subtropical forest habitat. The ancestry of this genus is in doubt, however, and it is not known at present if it represents a separate immigration event into Europe or if it is derived from one of the non-suspensory and partly

Vicariance biogeography

terrestrial forms like Kenyapithecus or Griphopithecus. This is an important question in determining the environmental and functional development during hominid evolution. Rudaba`nya is one of the few sites where fossil hominids are associated with pliopithecids, in this case Anapithecus, and this genus also is interpreted as suspensory and similar to howler monkeys in positional behaviour (Begun, 1993). Later in the Miocene, one of the signiWcant events in the vegetational history of Europe was the disappearance of the mangrove Avicennia from the northern Mediterranean and Paratethys in the middle to late Miocene. At the same time, the Taxodium and Sequoia humid subtropical forests were replaced by deciduous mesophytic forests by the Pliocene, and after 3.5 Ma much of the northern Mediterranean coastal area changed to steppic associations dominated by Artemisia (Suc et al., this volume). This is matched by marine invertebrates, so that for instance the larger benthic foraminifera which have temperature tolerances of 18–27 °C, corresponding to latitudes today of 0–35° north and south (Jones, this volume), have distributions in the Tethyan region indicating warm conditions in the early to middle Miocene, with persistence of warm conditions in the eastern Paratethys longer than in the west. Similarly, z-coral communities that today thrive in the Caribbean and IndopaciWc were present in the Tethys, disappearing by the late Miocene (Rosen, this volume). The northern limit for z-corals in the early Miocene was at the latitude of Southern France and Anatolia, shifting northwards in the middle Miocene to the area of the Vienna Basin, and southwards in the late Miocene to the present northern coast line of the Mediterranean. This indicates warmer conditions in Europe during the early to middle Miocene, with declining temperatures in the late Miocene and thereafter. The long sequence of deposits in Central Spain has a decline in species diversity at 16.5–15.5 Ma, with a second drop at 12.3 Ma and gradual increase thereafter until about l0 Ma (Daams et al., this volume). These changes are congruent with the shrews (see Reumer, this volume) and are related more strongly to changes in humidity than to changes in temperature as indicated by oxygen isotope curves. This is in line with O’Brien’s model for Southern Africa, which shows that a large proportion (79%) of woody plant species richness is accounted for by two aspects of climate (O’Brien, 1993), annual rainfall and an optimized function of energy (minimum monthly potential evapotranspiration). These two measures can predict variations in woody plant species richness, whereas temperature is only weakly related to diversity changes and its addition to the model fails to increase its resolving power (O’Brien & Peters, this volume). Hominid fossils almost disappear from Western Europe after about 9.5 Ma. Oreopithecus appeared in MN 12–13 deposits in Italy, and Dryopithecus

477

Palaeoenvironments: mammalian evidence

478

survives into MN 10 in Spain (at La Tarumba I). There is also the much disputed tooth from Salmendingen, the MN 11 type site for ‘Neopithecus’ brancoi, but its identiWcation is still uncertain. The sudden disappearance of hominids and pliopithecids follows the changing conditions indicated by plants and both land and marine animals, but in contrast to this, the persistence of hominids in Southeast Europe well into the late Miocene is consistent with the evidence of the Xora and fauna for the continuation of warmer and wetter conditions in this region long after the rest of Europe had become cooler and drier. For example, the foraminifera and corals both show declining temperatures throughout the tethyan region during the middle to late Miocene, but the northeastern parts of the sea were the last to be aVected by these changes (i.e. the region adjacent to southeastern Europe and Turkey). The Laurel forests that were present in the early Miocene (and earlier; Axelrod, 1975) disappear during the middle Miocene in Central Europe, although they persist for longer in Southeast Europe (for example in Greece; Kovar-Eder et al., 1996), where they include many paleotropical species. Deciduous taxa invaded Europe gradually during the middle Miocene, with examples of replacement even within the same genus, for example with Platanus leucophylla replacing the tropical P. neptuni. At the same time, ecologically tolerant taxa were present in Europe, such as Liquidambar europaea and other thermophillous taxa. These changes indicate replacement of tropical to subtropical plant associations by more temperate taxa by the end of the Miocene (Kovar-Eder et al., 1996). At the present time, Sequoia forests with laurophyllous understorey persist along the west coast of California, and this probably provides a good model for the transition phase of the Miocene European environments. The forests have essentially two canopies, a high emergent canopy that is evergreen but discontinuous and composed almost entirely of redwoods, with a more continuous partly deciduous canopy dominated in places with laurel. Two photographs of this forest in Northern California are shown here in Fig. 23.10, and while it is evident that the redwoods would have contributed little to the diet of essentially frugivorous Miocene hominids, the lower canopy and the rich ground vegetation would have oVered much more. At about the same time as the early radiation of dryopithecines, the earliest hominids in Southern Asia appear. Sivapithecus is Wrst known around 12.5 Ma in Indo-Pakistan (Kappelman et al., 1991), but the relationships of this orang-like genus are not entirely clear at present. It is not known, for instance, if Sivapithecus is related to or descended from any of the other Eurasian hominids or if it was derived from a separate immigration event from Africa. There is still little information about the probable environments present in the Indian subcontinent during the late Miocene,

Vicariance biogeography

479

[Figure 23.10] Sequoia/laurel forest in Northern California.

but predominance of browsing ungulates from late Miocene faunas at Samos suggests environmental conditions diVerent from the traditional view of a savanna-like environment (Solounias et al., this volume). The best analogy is with present day mixed monsoon forests with grassland glades of north Central India, as represented for instance at Kanha Park. The mammalian fauna from this subtropical summer-rainfall region closely matches those of the Pikermian faunas of Samos, Pikermi, Maragheh and the Thessaloniki faunas reported here (de Bonis & Koufos, this volume).

Summary Hominoid primates arguably arise with the Wrst appearance of Proconsul in the African terminal Oligocene. They evolve isolated in Africa in a tropical context (Andrews & Van Couvering, 1975) until the end of the early Miocene (MN 5 of European stratigraphic terminology) when they Wrst immigrate into Europe with a diverse eucatarrine community of species belonging to the Pliopithecidae (Bernor, 1983, 1984; Andrews et al., 1996) and one species of hominid. The Asian eucatarrhine Dionysopithecus represents an even

Palaeoenvironments: mammalian evidence

480

earlier Miocene (c. 18–16 Ma; Bernor et al., 1988a) biogeographic incursion into East and South Asia. Hominids entered Europe in the middle Miocene because land crossings of the Tethys were possible at this time and the subtropical forest environments were suitable. By virtue of the taxonomic diversity and restricted biogeographic ranges alone, Eurasian Miocene hominids show strongly vicariant evolutionary patterns. This suggests a model whereby a founding species extends its range under favorable and speciWc environmental circumstances and then becomes geographically restricted to refugia by geographic (= tectonic and paleogeographic) and/or environmental events. A frequent byproduct of vicariance, exercised over millions of years time, is homoplasy, and it is evident that there was a great deal of homoplasy in Miocene hominids. The nature of the environments occupied by apes in Europe had many structural similarities with the environments in Africa with which they are associated at this time (Andrews & Humphrey, 1998). Paleoecological evidence suggests that African middle Miocene apes lived in seasonal woodlands and forests, for example at Fort Ternan (Andrews, 1996) and Maboko Island (BeneWt & McCrossin, 1995). The hominids at these sites were partly terrestrial and with their large thick-enamelled teeth were adapted for similar diets to some of the European apes. The earliest European apes were similar in being both partly terrestrial and with almost identical dietary adaptations. The community structure of the mammalian faunas in the African and European sites is extremely similar (Andrews, 1992a), and by inference the ecosystem they occupied was also similar. The similarities in locomotor and dietary adaptations of the African and European apes at this early stage indicates further that their position in their respective ecosystems was also very similar. In one sense, therefore, these middle Miocene taxa in Europe were not as distinct from their African relatives as taxonomic divisions and their geographic separation may appear to indicate. Towards the end of the middle Miocene, at 13–12 Ma, the trends of partial terrestriality and thick-enamelled frugivory continued in a group of fossil apes assigned either to the pongine clade or to a paraphyletic group unrelated to any living ape (Andrews, 1992b; Pilbeam, 1996), as in our concordance cladogram (Fig. 23.9). They are associated with a range of open forest to woodland environments ranging from Southeast Europe to China, and one genus at least may be related to the orang utan. At the same time, the more arboreal and suspensory Dryopithecus emerged in association with closed subtropical forest environments where they are sometimes found associated with Pliopithecus and Anapithecus. Similar environments (Harrison & Harrison, 1989) and a similar, possibly heritage, adaptation for

Vicariance biogeography

suspensory locomotion (Harrison, 1991) persisted in Oreopithecus, although it has recently been argued (Ko¨hler & Moya`-Sola`, 1997) that this fossil ape may also have had adaptations for terrestrial bipedal mode of locomotion and the evolutionary relationships of Oreopithecus are still unclear. Hominid primates enjoyed favorable environmental circumstances in the late early and middle Miocene intervals over much of Eurasia. However, the Alpine–Himalayan orogeny caused major changes in land–sea relations, global climatic circulation patterns and seasonality particularly in Central Asia. Regression of the Paratethys likewise caused a shift of habitats to greater seasonality and replacement of evergreen subtropical forests by deciduous woodlands and, progressively in the late Miocene, more seasonal warm temperate woodlands with progressively more open habitats. Hominid primate distribution tracked these changes closely during the 12–9 Ma interval, contracting their ranges from both west and east and Wnding temporary refuge in Southeastern Europe, where favorable subtropical conditions persisted for a time after being lost elsewhere. Hominids disappeared from this region Wnally during MN 11, although they persisted until MN 12 in local insular habitats in Italy and the latest Miocene of China.

Acknowledgements We are most grateful to Terry Harrison and David Begun for constructive comments on the text of this paper. We also gratefully acknowledge Annette Moth Wiklund and the European Science Foundation for funding the Network on Hominoid Evolution and Environmental Change in the Neogene of Europe, Jorde Agustı´ for initiating the Network and Lorenzo Rook for organizing the second workshop in Siena. Some of this work has been funded by the L. S. B. Leakey Foundation and National Geographic (grants for Rudaba´nya and Pas¸alar excavations) and the Alexander von Humboldt Foundation. We would also like to acknowledge discussions with the following over a number of years when some of these issues have been raised: David Begun, David Cameron, Eric Delson, Jens Franzen, Judith Harris, Terry Harrison, Meike Ko¨hler, Laszlo Kordos, Lawrence Martin, David Pilbeam, Lorenzo Rook, JeV Schwartz, and Salvador Moya`-Sola`.

References Agustı´, J., Ko¨hler, M., Moya`-Sola`, S., Cabrera, L., Garce´s, M. & Pares, J. M. 1996. Can Lobateres: the pattern and timing of the Vallesian hominoid radiation reconsidered. Journal of Human Evolution, 31, 143–55.

481

Palaeoenvironments: mammalian evidence

482

Alpagut, B., Andrews, P., Fortelius, M., Kappelman, J., Temizsoy, I., Celebi, H. & Lindsay, W. 1996. A new specimen of Ankarapithecus meteai from the Sinap Formation of central Anatolia. Nature, 382, 349–51. Andrews, P. 1982. Ecological polarity in primate evolution. Zoological Journal of the Linnean Society, 74, 233–44. Andrews, P. 1990. Palaeoecology of the Miocene fauna from Pas¸alar, Turkey. Journal of Human Evolution. Hum. Evol., 19, 569–82. Andrews, P. 1992a. Community evolution in forest habitats. Journal of Human Evolution, 22, 423–38. Andrews, P. 1992b. Evolution and environment in the Hominoidea. Nature, 360, 641–6. Andrews, P. 1996. Palaeoecology and hominoid palaeoenvironments. Biological Reviews, 71, 257–300. Andrews, P., Begun, D. & Zylstra, M. 1997. Palaeoecology of Miocene hominoids. In Miocene Hominoid Fossils: Functional and Phylogenetic Implications, Begun, D. (ed.), pp. 29–58. Plenum Press, New York. Andrews, P., Harrison, T., Martin, L., Delson, E. & Bernor, R. 1996. Systematics and biochronology of European Neogene catarrhines. In Evolution of Neogene Continental Biotopes in Europe and the Eastern Mediterranean, Bernor, R. & Fahlbusch, V. (eds.), pp. 168–207. Columbia University Press. Andrews, P. & Humphrey, L. 1998. African Miocene environments and the transition to early hominines. In African Biogeography, Climate Change, and Early Hominid Evolution, Bromage, T. & Schrenk, F. (eds.). Oxford, Oxford University Press. Andrews, P. & Van Couvering, Judith, A. H. 1975. Environments in the East African Miocene. In Contributions to Primatology, Szalay, F. S. (ed.), Vol. 5, pp. 62–103. Axelrod, D. 1975. Evolution and biogeography of Madrean-Tethyan sclerophyll vegetation. Annals of the Missouri Botanical Garden, 62, 280–334. Barry, J. C., Johnson, N. M., Raza, S. M. & Jacobs, L. L. 1985. Neogene mammalian faunal change in southern Asia: Correlations with climatic, tectonic, and eustatic events. Geology, 13, 637–40. Begun, D. R. 1992. Dryopithecus crusafonti sp.nov. a new Miocene hominoid species from Can Ponsic (Northeastern Spain). American Journal of Physical Anthropology, 87, 291–309. Begun, D. R. 1993. New catarrhine phalanges from Rudabanya (northeastern Hungary) and the problems of parallelism and convergence in hominoid postcranial morphology. Journal of Human Evolution, 24, 373–402. Begun, D. R. 1994. Relations among the great apes and humans: new interpretations based on the fossil great ape Dryopithecus. Yearbook of Physical Anthropology, 37, 11–64. Begun, D. R. & Kordos, L. 1993. Revision of Dryopithecus brancoi Schlosser, 1901, based on the fossil hominoid material from Rudaba´nya. Journal of Human Evolution, 25, 271–85. Begun, D. R. & Kordos, L. 1997. Phyletic aYnities and functional convergence in Dryopithecus and other Miocene and living hominoids. In Function, Phylogeny and Fossils, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 291–316. New York, Plenum Press. Begun, D. R., Ward, C. V. & Rose, M. D. 1997. Function, Phylogeny and Fossils. New York, Plenum Press.

Vicariance biogeography

BeneWt, B. & McCrossin, M. 1995. Miocene hominoids and hominid origins. Annual Review of Anthropology, 24, 237–56. Berggren, W. A. & Van Couvering, J. A. 1974. The Late Neogene: Biostratigraphy, Geochronology, and Palaeoclimatology of the last 15 million years. Elsevier. Bernor, R. L. 1983. Geochronology and zoogeographic relationships of Miocene Hominoidea. In New Interpretations of Ape and Human Ancestry, Ciochon, R. L. & Corruccini, R. S. (eds.), pp. 21–64. Plenum Press, New York. Bernor, R. L. 1984. A zoogeographic theater and biochronologic play: The time/biofacies phenomena of Eurasian and African Miocene mammal provinces. Pale´obiologie Continentale, 14, 121–42. Bernor, R. L., Andrews, P. J., Solounias, N. & Van Couvering, J. A. 1979. The evolution of Pontian mammal faunas: Some zoogeographic, palaeoecologic, and chronostratigraphic considerations. Annales Ge´ologiques des Pays Helleniques, hors. se´rie, 1, 81–9. Bernor, R. L. & Tobien, H. 1990. The mammalian geochronology and biogeography of Pas¸alar (middle Miocene, Turkey). Journal of Human Evolution, 19, 551–68. Bernor, R. L., Woodburne, M. O. & Van Couvering, J. A. 1980. A contribution to the chronology of some Old World Miocene faunas based on hipparionine horses. Ge´obios, 13, 25–59. Bernor, R. L., Flynn, L., Harrison, T., Hussain, T. & Kelley, J. 1988a. Dionysopithecus from southern Pakistan and the biochronology and biogeography of early Eurasian catarrhines. Journal of Human Evolution, 17, 339–58. Bernor, R. L., Kovar-Eder, J., Lipscomb, D., Ro¨gl, F., Sen, S. & Tobien, H. 1988b. Systematics, stratigraphic, and paleoenvironmental contexts of Wrst-appearing Hipparion in the Vienna Basin, Austria. Journal of Vertebrate Paleontology 8, 427–52. Bernor, R. L., Tobien, H. & Woodburne, M. W. 1989. Patterns of Old World hipparionine evolutionary diversiWcations and biogeographic extension. In Topics on European Mammalian Chronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 263–319. New York, Plenum. Bernor, R. L., Fahlbusch, V., Andrews, P., de Bruijn, H., Fortelius, M., Ro¨gl, F., Steininger, F. F. & Werdelin, L. 1996a. The evolution of western Eurasian Neogene mammal faunas: a chronologic, systematic, biogeographic and palaeoenvironmental synthesis. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 449–70. New York, Columbia University Press. Bernor, R. L., Solounias, N., Swisher, C. C. & Van Couvering, J. A. 1996b. The correlation of three clasical Pikermian faunas – Maragheh, Samos and Pikermi – with the European MN unit system. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 137–56. New York, Columbia University Press. Bernor, R. L., Fahlbusch, V., Mittmann, H.-W. & Rietschel, S. 1996c. The Evolution of Western Eurasian Neogene Mammal Faunas In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 1–6. New York, Columbia University Press. de Bonis, L. & Koufos, G. 1996. The phylogenetic and functional implications of Ouranopithecus macedoniensis. In Function, Phylogeny and Fossils, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 317–26. New York, Plenum Press.

483

Palaeoenvironments: mammalian evidence

484

de Bonis, L., Bouvrain, G., Geraads, D. & Koufos, G. 1992. Multivariate study of late Cenozoic mammalian faunal compositions and paleoecology. Paleontologie y Evolution, 24–25, 93–101. de Bruijn, H., Daams, R., Daxner-Ho¨ck, G., Fahlbusch, V., Ginsburg, L., Mein, P. & Morales, J. 1992. Report of the RCMNS working group on fossil mammals, Reisensburg 1990. Newletters on Stratigraphy, 26, 65–118. Cameron, D. W. 1997. A revised systematic scheme for the Eurasian Miocene fossil Hominidae. Journal of Human Evolution, 33, 449–77. Cande, S. C. & Kent, D. V. 1995. Revised calibration of the geomagnetic time scale for the Cretaceous and Cenozoic. Journal of Geophysical Research, B100, 6093–5. Delson, E. & Andrews, P. 1975. Evolution and interrelationships of the catarrhine primates. In Phylogeny of the Primates: A Multidisciplinary Approach, Luckett, W. P. & Szalay, F. S. (eds.), pp. 405–46. New York, Plenum Press. Ersoy, A., Alpagut, B. & Andrews, P. (in press). Comparative anatomy of Griphopithecus phalanges from Pas¸ alar, Turkey. Journal of Human Evolution. Fahlbusch, V. 1991. The meaning of MN zonation: considerations for a subdivision of the European continental Tertiary using mammals. Newsletters on Stratigraphy, 24, 159–73. Fortelius, M., Werdelin, L., Andrews, P., Bernor, R., Gentry, A., Humphrey, L., Mittmann, W. & Viranta, S. 1996. Provinciality, diversity, turnover and palaeoecology in land mammal faunas of the later Miocene of western Eurasia. In Evolution of Neogene Continental Biotopes in Europe and the Eastern Mediterranean, Bernor, R. & Fahlbusch. V. (eds.), pp. 414–48. Columbia University Press. Garce´s, M., Agustı´, J., Cabrera, L. & Pares, J. M. 1996. Magnetostratigraphy of the Vallesian (late Miocene) in the Valle´s-Penede´s Basin (northeast Spain). Earth and Planetary Science Letters, 142 (1996), 381–96. Goodman, M. 1963. Man’s place in the phylogeny of the primates as reXected in serum proteins. In ClassiWcation and Human Evolution, Washburn, S. L. (ed.), pp. 204–34. Chicago, Aldine Press. Haq, B. U., Hardenbohl, H. & Vail, P. R. 1987. Chronology of Xuctuating sea levels since the Triassic. Science, 235, 1156–7. Harrison, T. 1991. The implications of Oreopithecus bambolii for the origins of bipedalism. Cahiers de Pale´oanthropologie, CNRS 1991, 235–44. Harrison, T. & Harrison, T. 1989. Palynology of the late Miocene Oreopithecus-bearing lignite from Baccinello, Italy. Palaeogeography Palaeoclimatology Palaeoecology, 76, 45–65. Harrison, T. & Rook, L. 1997. Enigmatic anthropoid or misunderstood ape? The phylogenetic status of Oreopithecus bambolii reconsidered. In Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 327–62. New York, Plenum Press. Kappelman, J., Kelley, J., Pilbeam, D., Sheikh, K., Ward, S., Anwar, M., Barry, J., Brown, B., Hake, P., Johnson, N., Mahmood Raza, S. & Ibrahim Shah, S. 1991. The earliest occurrence of Sivapithecus from the middle Miocene Chinji Formation of Pakistan. Journal of Human Evolution, 21, 61–73. Kappelman, J., Sen. S., Fortelius, M., Duncan, A., Alpagut, B., Crabaugh, J., Gentry, A., Lunkka, J. P., McDowell, F., Solounias, N., Viranta, S. & Werdelin, L. 1996. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L.,

Vicariance biogeography

Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 78–95. New York, Columbia University Press. Kay, R. F. & Ungar, P. S. 1997. Dental evidence for diet in some Miocene catarrhines with comments on the eVects of phylogeny on the interpretation of adaptation. In Function, Phylogeny and Fossils, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 131–52. New York, Plenum Press. Ko¨hler, M. & Moya`-Sola`, S. 1997. Ape-like or hominid-like? The positional behaviour of Oreopithecus bambolii reconsidered. Proceedings of the National Academy of Sciences, 94, 11747–50. Kordos, L. 1991. Le Rudapithecus hungaricus de Rudaba´nya (Hongrie). L’Anthropologie, 95, 343–62. Koufos, G. D. 1993. Mandible of Ouranopithecus macedoniensis (Hominidae, Primates) from a new late Miocene locality of Macedonia (Greece). American Journal of Physical Anthropology, 91, 225–34. Kovar-Eder, J., Kvacek., Zastawniak, E., Givulescu, R., Hably, L., Mihajlovic, D., Teslenko, J. & Walther, H. 1996. Floristic trends in the vegetation of the Paratethys surrounding areas during Neogene times. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 395–413. New York, Columbia University Press. Krijgsman, W., Landereis, C. G., Daams, R. & van der Meulen, A. J. 1996. Magnetostratigraphic dating of the middle Miocene climate change in the continental deposits of the Aragonian type area in the Calatayud-Teruel Basin (Central Spain). Earth and Planetary Science Letters, 142, 367–80. Lindsay, E. H., Fahlbusch, V. & Mein, P. 1989. Topics on European Mammalian Geochronology, New York, Plenum Press. McCrossin, M. L. & BeneWt, B. R. 1997. On the relationships and adaptations of Kenyapithecus, a large bodied hominoid from the middle Miocene of eastern Kenya. In Function, Phylogeny and Fossils, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 241–68. New York, Plenum Press. Mein, P. 1975. Resultats du groupe de Travail des Vertebres. In Report on Activity of the RCMNS Working Groups (1971–1975), Senes, J. (ed.), pp. 78–81. Bratislava, SAV. Mein, P. 1979. Rapport d’activite du Groupe de Travail Vertebres: Mise a jour de la biostratigraphie du Neogene basee sur les mammiferes. Annales Geologie Pays Hellenica, 1973, 1367–72. Mein, P. 1989. European mammal correlations. In Topics on European Mammalian Geochronology, Lindsay, E. H., Fahlbusch, V. & Mein, P. (eds.), pp. 73–90. New York, Plenum Press. Moya`-Sola`, S. & Ko¨hler, M. 1993. Recent discoveries of Dryopithecus shed new light on evolution of great apes. Nature, 365 543–5. Moya`-Sola`, S. & Ko¨hler, M. 1995. New partial cranium of Dryopithecus lartet 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain. Journal of Human Evolution, 29, 101–39. O’Brien, E. M. 1993. Climatic gradients in woody plant species richness: towards an explanation based on an analysis of southern Africa’s woody Xora. Journal of Biogeography, 20, 181–98. Ozansoy, F. 1965. Etude des gisements continentaux et de mammiferes du Cenozoique de Turquie. Memoires Societe Geologique de France, 44, 1–92.

485

Palaeoenvironments: mammalian evidence

486

Pilbeam, D. R. 1996. Genetic and morphological records of the Hominoidea and hominid origins: a synthesis. Molecular Phylogenetics and Evolution, 5, 155–68. Pilbeam, D. R., Morgan, M., Barry, J. C. & Flynn, L. 1996. European MN units and the Siwalik faunal sequence of Pakistan. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 96–105. New York, Columbia University Press. Pilbeam, D. R., Rose, M. D., Barry, J. C. & Shah, S. M. I. 1990. New Sivapithecus humeri from Pakistan and the relationship of Sivapithecus and Pongo. Nature, 348, 237–9. Quade, J., Cerling, T. E. & Bowman, J. R. 1989. Development of Asian monsoon revealed by marked ecological shift during the latest Miocene in Northern Pakistan. Nature, 342, 163–6. Ro¨gl, F. & Daxner-Ho¨ck, G. 1996. Late Miocene Paratethys correlations. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 47–55. New York, Columbia University Press. Ro¨gl , F., Zapfe, H., Bernor, R., Brzobohaty, R., Daxner-Ho¨ck, G., Draxler, I., Fejfar, O., Gaudant, J., Herrmann, P., Rabeder, G., Schultz, O. & Zetter. R. 1993. Die primatenfundstelle Go¨tzendorf an der Leitha, Nierdero¨sterreich. Jahrbuch der Geologischen Bundesanstalt, 136 (2), 503–26. Schwartz, J. H. 1990. Lufengpthecus and its potential relationship to an orang-utan clade. Journal of Human Evolution, 19, 591–605. Schwartz, J. H. 1997. Lufengpithecus and hominoid phylogeny: problems in delineating and evaluating phylogenetically relevant characters. In Function, Phylogeny and Fossils, Begun, D. R., Ward, C. V. & Rose, M. D. (eds.), pp. 363–88. New York, Plenum Press. Sen, S. 1989. Hipparion datum and its chronologic evidence in the Mediterranean area. In European Neogene Mammal Chronology, Lindsay, E. H. (ed.), pp. 15–46. New York, Plenum. Sen, S. 1996. Present state of magnetostratigraphic studies in the continental Neogene of Europe and Anatolia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 56–63. New York, Columbia University Press. Steininger, F. F., Berggren, W. A., Kent, D. V., Bernor, R. L., Sen, S. & Agustı´, J. 1996. Circum-Mediterranean Neogene (Niocene and Pliocene) Marine-continental chronologic correlations of European mammal units. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 7–46. New York, Columbia University Press. Steininger, F. F., Bernor, R. L. & Fahlbusch, V. 1989. European Neogene marine/continental chronologic correlations. In European Neogene Mammal Chronology, Lindsay, E. H. (ed.), pp. 15–46. New York, Plenum. Swisher, C. 1996. New 40Ar/39Ar dates and their contribution toward a revised chronology of the late Miocene nonmarine of Europe and west Asia. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 64–77. New York, Columbia University Press. Thompson, R. S. & Fleming, R. F. 1996. Middle Pliocene vegetation: reconstructions, paleoclimatic inferences, and boundary conditions for climatic modeling.

Vicariance biogeography

Marine Micropaleontology, 27, 13–26. Welcomme, J.-L., Anguilar, J.-P. & Ginsburg, L. 1991. De´couverte d’un nouveau Pliopithe`que (Primates, Mammalia) associe´ a` des rongeurs dans les sables du Mioce`ne supe´rieur de Priay (Ain, France) et remarques sur la pale´oge´ographie de la Bresse au Valle´sien. Comptes Rendues Hebdomadaires de l’Academie des Sciences, 313, Se´r 11, 723–9. Woodburne, M. O., Bernor, R. L. & Swisher, C. C. 1996a. An appraisal of the stratigraphic and phylogenetic bases for the Hipparion datum in the Old World. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 124–36. New York, Columbia University Press. Woodburne, M. O., Theobald, G., Bernor, R. L. Swisher, C. C., Ko¨nig, H. & Tobien, H. 1996b. Advances in the geology and stratigraphy at Howenegg, southwestern Germany. In The Evolution of Western Eurasian Neogene Mammal Faunas, Bernor, R. L., Fahlbusch, V. & Mittmann, H.-W. (eds.), pp. 106–23. New York, Columbia University Press.

487

Index

Abies, 382, 383 Abruzzi-Apulia palaeobioprovince, 191–3, 197 Aceraceae, 184 Aceratherium, 217, 221, 223 Aceratherium incivisum, 167, 170 Aceratherium kiliasi, 212 Aceritherium incisivum, 119 Aceritherium tetradactylum, 119 Acerorhinus, 253, 261 Acteocemas, 96, 98 Adcrocuta, 104, 114, 209, 212 Adcrocuta eximia, 118, 167, 173, 209, 211, 214, 215, 218, 220, 222, 230, 449 Aden, Gulf of, Pliocene tephra correlation, 24, 31, 31–51 Aegean, rodent faunas, 17 aeolian deposits, Upper Valdarno basin, 363–5 Aepycerotini, 444 Afghanistan, Miocene primate taxa, 458 Africa climate, 62–4 climate/vegetation relationship, 64–8, 290, 292 correspondence analysis, recent faunas, 419–29 palaeoenvironment reconstruction, 290, 292 Pliocene environmental change, 66–77, 446 savanna fauna evolution, 437–8, 446–9 Africa-Eurasia palaeogeography, 14–17 African entrants into Europe, 96, 98, 99, 101, 106, 115, 197, 271, 291, 297, 406 African monsoon, 62 Afromontane vegetation, 293 Afropithecus, 463–73 Agenian, 90 Agnotherium antiquus, 116 Albanensia, 104, 146, 181

Albanensia grimmi, 157, 158 Albanohyus, 403 Albanohyus pygmaeus, 119 Alcelaphini, 444 algal symbionts, 282–3, 311, 315 Alicornops alfambrense, 119, 173 Alicornops simorrense, 114, 119 Alilepus, 145, 198–9 Allohyaena kadici, 173 Allosoricinae, 392, 394 Allospalax, 152 alluvial sediments NE Spain, 398–9 Sinap, 243–7 Tuscany, 365–8 Alpine foredeep, 13, 15 altitudinal position, 89–90 altitudinal trees, 379, 383 Altomiramys, 96 Alveolinella, 291 Amblycoptus, 267, 268, 270, 393 American mammal immigration to Europe 9, 13, 17, 446;, see also Bering landbridge American shrews, 392, 393, 394–5 Ammonia, 291 Amphechinus, 265 amphibians, Italy, 191 Amphicyon, 14, 103 Amphicyon castellanus, 116 Amphicyon major, 116, 172 amphicyonids, 103, 104, 172, 174, 403 Amphilagus, 267 Amphilagus sarmaticus, 265, 266 Amphimoschus-like cervid, Italy, 192 Amphiope bioculata, 278 Amphiprax, 104 Amphiprox anocerus, 120, 170 Amphitragulus, 94, 100 Anacus arvernensis, 116, 173 Anapithecus, 116, 295, 460–1, 477

Index

Anapithecus cf. hernyaki, 169, 174, 461 Anapithecus hernyaki, 169 Anatolia, Turkey see Sinap Formation Anatolian plate, 14–17 Anchitherium, 13, 90, 96, 113, 119, 133, 252 Anchitherium aurelianensis, 172 Anchitherium sampelayoi, 119 Ancyclotherium pentelici, 119 Ancylotherium, 212, 415 Ancylotherium pentelicum, 218, 220, 221 Andegameryx, 94, 96, 98 Angustidens, 178 Ankarapithecus, 252, 261, 295, 465–76 Ankarapithecus meteai, 461 anoa, 445 anomalomyids, 105, 149, 150–2, 182 Anomalomys, 105, 150, 182, 265, 267 Anomalomys cf. gaudryi, 157 Anomalomys gaillardi, 158 Anourosorex, 178, 184 Anourosorex kormosi, 162 Anourosorex squamipes, 393 Anourosoricini, 393, 395 Antarctic ice, 17, 68, 403 antelopes, 436, 444 Antemus, 402 Anthracoglis, 195 Anthracoglis cf. marinoi, 195 Anthracoglis marinoi, 194, 195 Anthracomys lorenzi, 198–9 Anthracomys majori, 195, 196 anthracotherids, 95, 105–6, 195 Anthracotherium, 11 Apodemus, 106, 151, 271 Apodemus aV. primaevus, 89 Apodemus etruscus, 198–9 Apodemus gudrunae, 151, 164, 200 Apula amberti, 341, 342 Apula escoYerae, 338, 340 Apula koehnei, 345 Aquitanian vegetation, 380–1 Arabian plate, 14–17 Arabian Sea, Pliocene tephra correlations, 23–51 Aragonian MN units, Spain, 86, 86–7, 96–102 rodent assemblages, Spain, 127–38 shrews, 392

Aragonian-Vallesian transition, 113–24 Aragoral mudejar, 121 Archaeodesmana, 145, 176, 178, 184 Archaeodesmana cf. pontica, 163 Archaeodesmana vinea, 157, 158, 161, 162, 163 Arctaphicyon, 173 Ardipithecus, 467, 473 Ardroverictis schmidkit., 117 aridity, 24, 28, 357–65, 407–8, 437 Africa, 64, 68, 69, 70, 71–3, 77, 446, 447 mock aridity, 69, 70, 73 Armantomys, 94, 96, 136 Artemisia steppe, 379, 383, 477 artiodactyls, 94, 96, 98, 100–1, 106, 119 Central Europe, 170 Greece, 206, 208 Italy, 192, 197, 198 Spain, 403 Arvicolidae, 152, 162 Asellia, 145 Asellia cf. mariatheresae, 158, 200 Asia, correspondence analysis, recent faunas, 419–29 Asian entrants into Europe, 95, 98, 99, 101, 106, 178–81, 197, 271, 406 Asoriculus, 176, 177, 393 Asoriculus gibberodon, 393 Asteraceae, 382 Atlanteroxerus, 136, 252 Atlanteroxerus cf. rhodius, 200 atmospheric circulation, 59–62, 372, 397, 407 Aureliachoerus, 96 Australopithecus, 449 Austria Miocene primate taxa, 457 non-marine molluscs, 334–5, 336–7, 339 see also Central Europe Austroportax, 101, 102, 106, 121 Avicennia mangrove, 380, 381, 383 Axis axis, 445 Baranogale adroveri, 117 Baranogale cf. adroveri, 173 Barberahyus, 114 Barberahyus castellensis, 1194 basin correlations, Eurasia, 85–90, 258–9

489

Index

490

basin facies analysis, Tuscany, 355–72 beavers 103, 104, 105, 166, 169, 181, 184; see also Castoridae Begertherium, 253 Belbus, 252 Belbus beaumonti, 218, 449 Bensonomys gileyi, 402 benthonic foraminifera, 282–93, 477 Beremendia, 394 Beremendia Wssidens, 394 Beremendia minor, 394 Beremendiini, 394, 395 Bering landbridge, 11, 17, 287, 290, 293, 296, 298, 393 Beryslavsky mammalian complex, 268–70 Betulaceae, 184 Bilkynsky mammalian complex, 270–1 biochronology hominoid primates, 455–63 non-marine molluscs, 329 biodiversity, and soricids, 391 biogeography and coral data, 312–14 hominoid primates, 454–81 biome method, 380 biostratigraphy Late Miocene, Central Europe, 167–74, 178–82 Miocene, Spain, 84–107 MN mammal units, 84–107 Oligocene-Miocene, W Eurasia, 274–98 Upper Miocene, France, 153–4 Birgerbohlinia, 105, 121 Bison bison athabascae, 445 Bithinia, 366 Bitlis Zone, 16 bivalves, Paratethys, 12, 278 Blackia, 146, 181, 265 Blackia miocaenica, 157, 158, 162 Blarinella, 176, 393 Blarinellini, 392–3, 394 Blarinini, 393 Blarinoides, 145, 164, 393 Bohlinia, 253, 261 Bohlinia attica, 211, 213, 215, 221, 223, 230, 441, 449 Bohlinia nikitiae, 213 Bos gaurus (gaur), 445

boselaphine bovids, 113, 212, 224 bovids, 98, 99, 101, 102, 104, 106–7, 113, 114, 425, 441 Greece, 206, 208–9, 212, 224, 227, 228, 229, 230, 231, 415, 418, 422, 429 Italy, 195, 198, 200 Siwaliks, 405 Turkey, 254 Brachyodus, 95, 96 Brachyodus onoideus, 205, 210 Brachypotherium, 119, 253, 404 Brachypotherium goldfussi, 167, 173 Bransatoglis, 96, 104, 134, 403 Bubalus quarlesi (mountain anoa), 445 Budorcas (takins), 445 Bunolistriodon, 13, 98, 100, 206, 253 Bunolistriodon lockarti, 210 Byzantinia, 252, 270 Cainotherium, 96, 100 Calatayud-Daroca Basin, 127–38 Calatayud-Teruel Basin, 113–24 Calligonum, 382 Calomyscus, 86 Camelopardis attica, 230 camels, 445 Can Vilella, Spain, 89 Canalicia, 332, 333 Canalicia attracta, 332 Canis cipio, 116 ‘Caprotragoides’, 99 Cardium lipoldi, 12 Carnivora Late Miocene, Central Europe, 174 Vallesian, Spain, 116–18, 403 Carychiopsis schwageri, 332 Carychium (Carychiella) puisseguri, 342 Carychium (Saraphia) nouleti, 335 Carychium (Saraphia) pachychilum, 342, 344 Carychium (Saraphia) pseudotetrodon, 344 Caspicyclotus, 333 Castillomys, 106, 198–9 Castor, 181 Castor cf. praeWber, 198–9 Castoridae 145–6, 402, 403, 424;, see also beavers Castromys, 151, 164

Index

Cathaya, 379, 382, 383 Cedrus, 379, 382, 383 Celadensia, 153 Celadensia grossetana, 157 cenogram analysis, faunal diversity, 122–3 Central Europe Wller Late Miocene faunal province, 417–18 Late Miocene mammals, 165–85 Miocene primate taxa, 457 palaeoclimate change and non-marine molluscs, 328–49 Centralomys benericettii, 200 ‘Cepaea’, 333 ‘Cepaea’ concudensis, 338 ‘Cepaea’ moroguesi, 331 ‘Cepaea’ silvana silvana, 331 ‘Cepaea’ subsulcosa, 331 Ceratonia, 382 Ceratotherium, 252, 261, 449 Ceratotherium neumayri, 212, 214, 216, 217, 221, 223 cercopithecids 169, 214, 216, 223, 224, 225, 296, 405, 413, 414;, see also Mesopithecus Cervavitulus mimus, 173 cervids, 96, 99, 101, 104, 105, 107, 441, 444 Central Europe, 172, 173, 174 Greece, 415, 417 Italy, 192, 198, 200 Turkey, 252, 261 Cervus canadensis (wapiti), 444, 445 Cervus unicolor (sambar), 444, 445 Chalicomys, 103, 104, 146, 403 Chalicomys jaegeri, 157, 158, 161, 163, 176 Chalicoteriidae as palaeoenvironmental indicators, 415–19 Chalicotheriinae, 214, 415 Chalicotherium, 98, 415 Chalicotherium goldfussi, 170, 217, 223 Chalicotherium grande, 119 Chasmaporthetes, 225 Chasmaporthetes bonisi, 214, 223 chelonians, Italy, 192 Cherevychansky mammalian complex, 271 Chilotherium, 253, 261 Chilotherium persiae, 218 China, 182, 426, 443, 459

Chiroptera, 145, 157–64, 195 chitals, 445 Chlamys albina, 278 Chlamys submalvinae, 278 Choerolophodon, 254 Choerolophodon chioticus, 206, 210 Choerolophodon pentelici, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 223, 224, 225, 230 Chondrinidae, 329, 333 chronostratigraphy, 48–51, 84–5 hominoid primates, 455–63 lignites, Tuscany, 365–8 macromammals, Spain, 123 non-marine molluscs, 329 Oligocene-Miocene, W Eurasia, 274–98 Sinap Formation, 247–9 Valle`s-Penede`s Basin, 397–408 Circamustela dechaseauxi, 117 Cistus, 379 cladograms, 463–74 Clausilia, 335 Clausilia baudoni, 342 Clausilia baudoni baudoni, 345 Clausilia baudoni tillensis, 345 Clausilia hollabrunnensis, 340 Clausilia portisi, 345 Clausilia produbia, 345 Clausilia rolfbrandti, 345 Clausilia strauchiana, 345 Clausilioidea, 329–47 climate, Late Miocene, France, 153 climate change, and mammal turnover, Late Miocene, Spain, 397–408 climate change and faunal succession, 88–9 climate parameters, 55–77 climate-modifying factors, 60–6 climate/vegetation relationship, 64–6, 71–7 climatostratigraphy, Oligocene-Miocene, W Eurasia, 274–98 cluster analysis, Vallesian large mammals, 114–15 Clypeaster, 278 Clypeaster intermedius, 277 Clypeaster martini, 278 coccolithophorids, 48 Cochlodina, 333, 335

491

Index

492

Cochlodina berthaudi, 342 Cochlodina laminata, 345 Cochlodina oppoliensis, 339–40 Cochlodina perforata, 332 Cochlodina prolaminata, 345 Cochlostoma, 333, 334, 343 commonality, foraminifera assemblages, 285–94, 298 Completeness Index, 251–7 coniferous forests, 271, 380 Connochaetes taurinus (wildebeest), 445 Conohyus, 99, 104 Conohyus simorrensis, 172 Constricta, 332, 333 Constricta tenuisculpa, 332 Convolvulus, 382 corals, Miocene, 293, 309–24, 398, 477, 478 Correspondence Factor Analysis, 419–29 Craspedopoma, 335, 341 Craspedopoma conoidale, 342 Cremohipparium matthewi, 441 Cremohipparium mediterraneum, 441 Cremohipparium proboscideum, 440, 441 Creneatachea, 336 cricetids, 11, 86–8, 88, 91, 97–100, 103, 104, 105, 131, 133, 134, 402–8 Central Europe, 181 France, 149–50, 153–4 Ukraine, 265–71 Cricetodon, 97, 102, 134 Cricetodon albanensis, 102 Cricetodon lavocati, 102 Cricetodon versteege, 98 Cricetodontinae, 87 Cricetulodon, 87, 103, 104, 150, 181, 402–8 Cricetulodon complicidens, 266 Cricetulodon hartenbergeri, 87 ‘Cricetum vacuum’, 95, 97 Cricetus, 17, 106 Cricetus cf. barrierei, 200 Criotherium, 254 Criotherum argalioides, 217, 441 Crocidosorex, 195 Crocidosorex antiquus, 191 Crocidosoricinae, 391–2, 395 Crocidura, 394 Crocidurinae, 394 crocodiles, Italy, 192

Crocuta crocuta, 449 crouzelines, 169, 295, 460, 461 Crusafontina, 178, 184, 265, 266 Crusafontina endemica, 157, 393 Crusafontina kormosi, 144, 157, 158, 161, 163 ctenodactylids, 192 Cupressaceae, 379 Cyclophoridae, 333 Cyperacea marshes, 382 Cyrtochilus, 332, 333, 336 Czech Republic, non-marine molluscs, 332 Dacian Basin, 17 dambos, 72, 290, 293, 295 Danish–Polish Strait, 11 De Geer Route, 9 Deamaninae, 268 Decennatherium, 103, 208, 209, 253 Decennatherium macedoniae, 209, 212 Decennatherium pachecoi, 121, 209 deciduous forests, 381, 382 Deep-Sea Drilling Project (DSDP), 23 deer, modern 173, 444; see also cervids Deinotherium, 14, 98, 103, 404 Central Europe, 165, 167, 170–2 Greece, 210, 211, 219, 230, 417, 418 Turkey, 254 Deinotherium bavaricum, 165, 170 Deinotherium giganteum, 116, 165, 167, 170–2, 172, 218, 220, 221 Deinotherium gigantissimum, 170–2 Deinotherium levius, 170 DeinsdorWa, 392 Democricetodon, 87, 91, 95–7, 133, 140, 150, 181, 252 Democricetodon cf. nemoralis, 157 Democricetodon freisingensis, 102 Democricetodon galliardi, 134 Democricetodon hispanicus, 133 Deperetomys hagni, 102 desertiWcation, 407 deserts, Africa 69; see also aridity Desmana, 176, 267, 268, 270 Desmanella, 145, 158, 159–60, 160, 178, 185 Desmanella cf. crusafonti, 161, 162, 163 Desmanella crusafonti, 161, 162, 163, 164 Desmanella stehlini, 157

Index

Dibolia, 176 ‘Dicerorhinus’, 198 Dicerorhinus cf. megarhinus, 198, 200 Dicerorhinus orientalis, 207, 217, 230 Dicerorhinus pikkermiensis, 217, 220, 221 ‘Dicerorhinus’ steinheimensis, 104, 119, 403 Diceros, 449 Diceros cf. pachygnathus, 198 Diceros pachygnathus, 207, 438 Dicroceros elegans, 172 dietary reconstructions, Pikermian ungulates, 439, 444–6 Dimylids, 144, 266 ‘Dinarica’ dalpiazi, 339 Dinocrocuta, 173, 226, 252 Dinocrocuta gigantea, 209, 212 Dinocrocuta salonicae, 213 Dinofelis, 449 Dinosorex, 144, 178, 185, 265, 266 Dinosorex cf. pachygnathus, 157, 158 Dinosorex pachygnathus, 157 Dionysopithecinae, 459, 468, 475 Diplommatinidae, 333 Dipodidae, 270 Dipoides, 146, 181 Dipoides problematicus, 161, 163, 198–9 Discostrobilops, 333 Discostrobilops uniplicatus, 331 Discus (Discus) euglyphusi, 332 Discus (Discus) vireti, 331 Discus laetumbilicus, 344 Discus pantanelli, 344 diversity Aragonian rodent faunas, 127, 127–38 foraminiferal assemblages, 284–98 Spanish Vallesian, 115–24 Vallesian macromammals, 115–24 Dolichopithecus, 297 Domninoides, 265 Dorcatherium, 96, 103, 106, 114, 206, 208, 210, 221 Dorcatherium crassum, 120 Dorcatherium naui, 120, 170 Dorcatherium penecki, 210 Dorcatherium puyhauberti, 215, 223, 224 dryopithecines, 294 Dryopithecus, 16, 101, 103, 104, 169, 172, 174, 294, 403, 417, 456–61, 463–80

Dryopithecus brancoi, 169, 461 Dryopithecus carinthiacus, 169, 460, 461 Dryopithecus crusafonti, 116, 460, 461 Dryopithecus fontani, 460 Dryopithecus laietanus, 116, 460, 461 Dryopithecus rhenanus, 169 East Africa, Pliocene tephra correlations, 23–51 Eastern Europe, Vallesian 240, 261; see also Eurasia Ebromys autolensis, 93 echinoids, 278 Echinolampas, 278 Echinolampas hemisphaericus, 278 Echinosoricinae, 144, 162 EKR and LKR, 135 Eliomys, 146–7, 158, 182, 403 Eliomys truci, 164 Ellobiidae, 331 endemic lineages, 85–90 endemicity, corals, 313 endemism, Italy, 192–7, 197 energy hypothesis, 309–24, 477 energy and water regimes, 55–6, 64–6, 75–7, 447 Engelhardia, 381, 383 Enhydriodon laticeps, 222 eolian dust variations, Pliocene-Pleistocene, 27–8, 34, 68 Eomellivora wimanni, 170 Eomuscardinus, 104, 146, 182, 403 Eomuscardinus cf. sansaniensis, 157 eomyids, 11, 88–9, 91, 94, 95, 134, 147, 149, 182–3 Eomyops, 89, 134, 182, 265, 266 Eomyops catalaunicus, 147, 157, 158, 159, 160, 162, 164 Eostrobilops, 333, 341, 343, 347 Eostrobilops aloisii, 344 Eostrobilops duvali, 342 Eostrobilops patuliformis, 344 Eotragus, 98, 101, 121, 206, 210 Eozapus, 178, 182, 268, 270 Eozapus intermedius, 150, 159, 160, 161, 164 Epimeriones, 150, 167, 181 Epimeriones aV. austriacus, 89, 162, 164

493

Index

494

Episoriculus, 176, 265, 266 Episuriculus aV. gibberodon, 200 Eptesicus cf. campanensis, 145, 158, 162 Eptesicus cf. noctuloi¨des, 145, 158, 162 equids 96, 102, 103, 113, 133, 417, 418, 422, 430, 441; see also Hipparion equitability, Aragonian rodent faunas, 127, 128–9, 136–7 Equus burchelli, 444 Equus grevyi, 444 Ericaceae moors, 381 Erinaceidae, 142–4, 198–9, 266 Erinaceus, 177 Ethiopia, tephra correlations, 23, 31, 48, 49 Etruria viallii, 195 Eualopia, 332, 333 Eualopia bulimoides, 332 Euboictis aliverensis, 206, 210 Eucricetodon, 90–2, 94, 95 Eucricetodon aquitanicus, 90, 91, 92, 94 Eucricetodon collatus, 91 Eucricetodon gerandianus, 90, 91, 92, 94 Eucricetodon hesperius, 91, 92 Eucricetodon infralactorensis, 90, 91, 92, 94, 95 Eucyon, 198–9 Eucyon monticinensis, 200 Eumaiochoerus etruscus, 195, 196 Eumyarion, 97, 98, 104, 150, 181 Eumyarion cf. latior, 157 Euprox, 106 Euprox dicranocerus, 120, 172 Euprox elsanus, 198 Euprox furcatus, 120, 172 Euprox minimus, 120 Eurasia Eocene plate tectonics, 9 Late Miocene, Ukraine, 265–71 late Miocene palaeoecology, 436–49 mammal turnover and climate change, 403, 405 Miocene hominoid primates, 454–81 Oligocene, faunal exchange, 12–17 Oligocene-Miocene palaeogeography, 274–308 Vallesian, 261, 403 Eurolagus fontannesi, 145, 159 Eurolistriodon, 98

Europe see Central Europe; Eastern Europe; Western Europe Euroxenomys, 104, 403 Eusmilus, 11 Eustrobilops Wscheri, 332 Euxinian Basin, 17 evaporites, 293, 357–63, 371–2 evapotranspiration, 61, 62, 64–5, 66, 76, 477 evergreen woodland, 174, 295, 442, 476 facies analysis, 355–72 FAD and LAD, 85–107, 134, 171, 174 Fahlbuschia, 97, 103, 136 Fahlbuschia crusafonti, 102, 103 felids, 95, 102, 103, 114, 222, 449 Felis, 438 Felis antediluviana, 118 Felis attica, 173, 218, 222 Felis ex. gr. attica-christoli, 200 Wr, 379 Flabellipecten larteti, 278 Xood-dominated deposits, Tuscany, 368–71, 371 Xora 166, 167, 170–71 174, 184–5, 440–43; see also palaeobotany; vegetation Xuvio-deltaic deposits, Tuscany, 369 Xuvio-lacustrine deposits, Tuscany, 363–5 Xying squirrels, 103, 104, 105, 146, 153, 184, 402, 403 FO and LO, 135 foraminifera, 9, 12–13, 16, 276–308, 320, 477, 478 forest habitats, 413, 422–4, 439–46, 475–9 Central Europe, 169–70, 174, 184–5 Western and circum-Mediterranean Europe, 379–84, 399–408 forest-steppes, Ukraine, 271 Forsythia, 265 Fortuna, 343 Fortuna seringi, 342 France Miocene primate taxa, 456 non-marine molluscs, 331–2, 334, 337, 338, 340–1, 343, 346, 347 Upper Miocene micromammals, 140–54 Frechenia, 343 Frechenia ducrosti, 341, 345

Index

Frechenia nayliesi, 341, 342, 345 Frechenia quadrifasciata, 342 Frechenia reichenbachi, 345 fruit production, and climate change, 383 Galactochilus, 332, 333, 339, 341 Galactochilus leobersdorfensis, 337 Galactochilus locardi, 338, 340 Galerix, 142, 144, 157, 159, 160, 161, 162, 163, 164, 177, 267, 270 Galerix aV. depereti, 200 Galerix cf. socialis, 157, 158, 198–9 Galerix iberica, 158 Gastrocopta, 337 Gastrocopta (Albinula), 332, 336, 337, 338, 339, 340, 341, 343, 347 Gastrocopta (Albinula) acuminata, 336 Gastrocopta (Albinula) acuminata acuminata, 335 Gastrocopta (Albinula) acuminata fossanensis, 344 Gastrocopta (Albinula) dupuy, 342 Gastrocopta (Albinula) edlaueri, 336 Gastrocopta (Albinula) turgida, 331–2 Gastrocopta (Sinalbinula), 332, 333, 334, 335, 336, 338, 339, 340, 346 Gastrocopta (Sinalbinula) baudoni, 342 Gastrocopta (Sinalbinula) larteti, 335 Gastrocopta (Sinalbinula) nouletiana, 335, 337 Gastrocopta (Sinalbinula) serotina, 336 Gastrocopta (Vercopsis) dehmi, 344 Gastrocoptinae, 329, 331, 333 gastropods, terrestrial, 191 gaur, 445 Gauss Gilbert palaeomagnetic reversal, 50 Gazella, 106, 212, 213, 214, 217, 218, 219, 222, 224, 230, 231, 254, 444 Gazella capricornis, 221, 228, 438 Gazella capricornis-gaudryi, 441 Gazella cf. gaudri, 216, 230, 231 Gazella deperdita, 121, 216, 218, 223, 228, 230 Gazella granti, 445 Gazella pilgrimi, 214, 215, 216, 217, 227, 228 General Circulation Models (GCMs), 28, 379

Gentrytragus, 99, 101 genus richness, Sinap Formation, Turkey, 257–8, 260–1 Georgiomeryx georgalasi, 207, 210 Geotrypus oschiriensis, 191 Geraniaceae, 382 Gerbillidae, 106, 198–9 Germany Miocene primate taxa, 457 non-marine molluscs, 331, 332, 334, 335, 343 see also Central Europe Gigantopithecus, 463, 467, 468 GiraVa attica, 438 giraVes, modern, 437, 449 giraYds, 95, 99, 101, 102, 103, 105, 113, 195 Greece, 415, 417, 418, 422, 424 Italy, 195, 207, 208, 228 Pikermian Biome, 436, 441, 445, 448–9 Turkey, 253 GiraVokeryx, 449 glacial cycles, 28, 73, 76, 397 Gliaudinus, 147 glirids, 88–9, 92, 94, 95–6, 103–40, 134, 136, 402–3 Central Europe, 182–3 France, 146–7 Greece, 211 Italy, 191, 195 Siwaliks, 404 Ukraine, 266–7 Glirudinus, 146 Glirudinus glirulus, 96 Glirudinus gracilis, 96 Glirudinus undosus, 157 Glirulus, 104, 142, 147, 182 Glirulus cf. diremptus, 147, 157, 159, 161 Glirulus cf. lissiensis, 158, 159, 160, 162, 163 Glirulus lissiensis, 147 Glis, 147, 176–8, 182 Glis major, 191 Glis minor, 164 Glis vallesiensis, 265 global climate change, and mammal turnover, 397–408 Globigerina marl facies, Paratethys, 9

495

Index

496

Globigerinoides extremus, 277 Globigerinoides ruber, 277 Globigerinoides subquadratus, 277 Globorotalia conomiozea, 358, 359 Globorotalia (Fohsella), 278–81 Globorotalia (Fohsella) fohsi, 278–81 Globorotalia (Fohsella) fohsi lobata, 278 Globorotalia (Fohsella) fohsi robusta, 278 Globorotalia (Fohsella) peripheroacuta, 278–81 Globorotalia (Fohsella) peripheroronda, 277 Globorotalia (Fohsella) praefohsi, 278–81 Globorotalia mayeri, 277 Globorotalia menardii, 277 Gomphotheridae, 14, 96, 200 Gomphotherium, 14–15, 192, 210, 254 Gomphotherium angustidens, 116, 172 Gomphotherium landbridge, 14–17, 191, 291 Gorilla, 449, 465–73 Graecopithecus, 295, 449, 461, 462, 463–74 Graecopithecus freybergi, 217, 230 Grand Coupure, 9, 392 Graphiurops, 147, 182 Graphiurops austriacus, 159, 160, 161–2 grass pollen as aridity indicator, 71–3, 77 grassland habitats, 71–3, 290, 404–5, 422–4, 436, 438, 439–43 Greco-Iranian faunal province, 417, 424, 426–31 Greece, Miocene macromammals, 205–31 micromammals, 206, 207, 208 primate taxa, 458 Griphopithecus, 16, 294, 455, 459, 460, 463–77 Grytsivsky mammalian complex, 265–7 gypsum units, Tuscany, 357–63 halite, 357, 363 Helicella (Xerotricha), 339 Helicigona, 340 Helicigona chaignoni, 341, 345 Helicigona schwarzbachi, 345 Helicigona truci, 342 Helicodonta hecklei, 332 Helicoidea, 329–47

Heliopithecus, 466, 467 ‘Helix’ bolivari, 338 ‘Helix’ vilanovai, 338 Helladorcus, 429 Helladorcus geraadsi, 212 Helladotherium, 211, 253, 2219 Helladotherium cf. duvernoyi, 231 Helladotherium duvernoyi, 213, 214, 216, 217, 219, 220, 221, 230, 441 Helladotherium mediterraneum, 229 Hemicyon, 96, 252 Hemisorex, 392 Herpestes dissimilis, 117 Heterocricetodon, 13 Heteroprox, 99, 101 Heteroprox larteti, 120, 172 Heterosoricids, 144, 266 Heteroxerus, 136, 145–6, 162 Heteroxerus grivensis, 157, 159 Heteroxerus huerzlerei, 158, 161 Hexaprotodon crusafonti, 339 Hexaprotodon pantanelli, 198 Hexaprotodon succulus, 198 Hipparion, 17, 101, 102, 113, 114 France, 140 Greece, 209, 212, 213, 219, 225, 226, 227, 230, 231, 429–30, 430 Italy, 198 Turkey, 240, 252 Ukraine, 198 Hipparion brachypus, 21, 221, 229, 230 Hipparion catalaunicum, 119 Hipparion concudense, 119 ‘Hipparion Datum’, 460 Hipparion dietrichi, 214, 215, 216, 217, 219, 227, 228, 440, 441 ‘Hipparion event’, 296, 401 ‘hipparion faunas’, 436–7, 443 Hipparion gromovae, 119 Hipparion koenigswaldi, 229–30 Hipparion macedonicum, 211, 212, 213, 214, 215, 216, 226, 227, 228 Hipparion matthewi, 217, 223, 224, 230 Hipparion mediterraneum, 119, 217, 219, 220, 221, 222, 223, 224, 228, 229, 230 Hipparion periafricanum, 223, 229 Hipparion primigenium, 119, 208, 211, 212, 266

Index

Hipparion proboscidium, 214, 217, 227, 228 hippopotamids, 17, 339 Hippopotamodon, 253, 261 Hippopotamus, 106 Hipposideros, 145 Hipposideros cf. collongensis, 158 Hippotherium, 102–3, 401 Hippotherium giganteum, 440, 441 Hippotherium primigenium, 166–7, 170, 430 Hippotragini, 444 Hispanodorcas, 106, 121 Hispanomeryx, 100–1, 104, 120 Hispanomys, 86, 102, 105, 106, 149, 150, 163, 267 Hispanomys aragoniensis, 86 Hispanomys bijugatus, 86, 157 Hispanomys cf. mediterraneus, 159, 160 Hispanomys cf. peralensis, 162 Hispanomys decedens, 86 Hispanomys dispectus, 86 Hispanomys mediterraneus, 86, 158, 161, 1529 Hispanomys nombrevilliae, 86 Hispanomys thaleri, 86 Hispanopithecus, 469 Hispanotherium, 98, 99, 100, 133 hominid localities, tephra correlations, 23–51 hominoids, 101, 103 biogeography and palaeoecology, 454–81 Central Europe, 169, 172, 174 and climate change, 397, 403, 404–5 disappearance, 296, 383 Greece, 208, 209, 212, 213, 226 Italy, 193–5, 197, 365 Oligocene-Miocene land migration, 290, 292, 293–4, 296 palaeoenvironment, Macedonia, 413–31 phylogenetic hypotheses, 463–74 Pikermian-African correlations, 448–9 Siwaliks, 404–5 Turkey, 253, 261 Homo, 465–71 homoplasy, 472, 480 Honanotherium schlosseri, 441 Hoplitomeryx, 192 Hoploaceratherium, 253

Huerzelerimys, 105, 106, 153 Huerzelerimys minor, 151, 152, 160, 161 Huerzelerimys oreopitheci, 195, 196 Huerzelerimys vireti, 105, 142, 151, 154, 162, 163, 194, 195 humidity Late Miocene, France, 153 and rodent habitats, Aragonian, 132–8 and shrews, 391, 395–6 and species diversity, 477 Vallesian, Spain, 122–3 western Europe, 424 Hungary, Miocene primate taxa, 457 Hyaena abronia, 449 Hyaena chaeretis, 438 Hyaena eximia, 438 Hyaena hyaena, 449 Hyaena salonicae, 226 Hyaenictis, 104 Hyaenictis almerae, 118 Hyaenictis graeca, 222 Hyaenictitherium hyaenoides, 218, 222 hyaenids, 101, 102, 104, 114, 167, 173, 174, 198–9, 209, 403, 436, 449 Hyaenotherium, 252 Hyaenotherium wongi, 215, 218, 222 Hydrocena, 343, 346 Hydrocena dubrueilliana, 342, 344 hydrological cycle, 58, 61, 70 Hylobates, 463, 465–74 Hylopetes, 146, 157, 159, 160, 181, 200 Hyotherium, 13 Hyotherium soemmeringi, 172 Hypolagus, 198–9 Hypposioderos (Syndemostis) cf. vetus, 200 Hypsodontus, 253 hypsodonty, 194, 436, 438, 444 Hyracoidea, 119, 424, 436 Hystrix, 182 Hystrix primigenia, 198–9, 200, 219, 221, 222, 224, 229, 438 Iberian peninsula landbridge, 297 Iberus, 340 Iberus dupuydelomei, 337, 338 ice age cycles, 28, 73, 76, 397 Ictitherium, 167, 252 Ictitherium pannonicum, 118

497

Index

498

Ictitherium viverrinum, 214, 215, 216, 218, 222 Indarctos, 102, 103, 114, 252 Indarctos anthracitis, 195 Indarctos arctoides, 170 Indarctos atticus, 116, 218, 221 Indarctos laurillardi, 196 Indarctos vireti, 116 Indian monsoon, 24, 26–8, 34 Indo-PaciWc gateway, 15–16 insectivores, 88, 104 Central Europe, 175, 178–81 France, 142–5, 157–64 Italy, 198–9 Ukraine, 265–71 insolation, and climate change, 56–7, 66–8 insular endemism, Italy, 192–7, 197 Intertropical Convergence Zone (ITCZ), 60, 62, 64, 70, 371 Iran, Miocene primate taxa, 458 Iranian plate, 13 Ischymomys, 270, 404 Ischymomys quadriradicatus, 267 Ischyrictus petteri, 117 isotopic research, 48–51, 249, 320–4, 406, 438–40 Italy facies analysis, 355–72 Miocene land mammals, 191–7 Miocene primate taxa, 458 non-marine molluscs, 339, 341, 346 Janschinella, 12 Janulus, 333, 334, 336, 340, 341, 342, 343 Janulus angustiumbilicatus, 344 Janulus densestriatus, 332 Janulus olisipponensis, 338 Janulus supracostus, 335 Juglandaceae, 381 Kalfynsky mammalian complex, 267 Kenya, tephra correlations, 23, 31, 37–51 kenyapithecines, 294, 473 Kenyapithecus, 463–77 Keramidomys, 89, 147, 182, 265 Keramidomys cf. pertesunatoi, 164 Keramidomys pertesunatoi, 159, 160 Klikia, 332, 333

Korynochoerus palaeochoerus, 225 Korynochoerus provincialis, 198, 200 Kowalskia, 105, 150, 167, 181, 195, 270 Kowalskia aV. lavocati, 89 Kowalskia cf. fahlbuschia, 270 Kowalskia nestori, 198–9 Kowalskia progressa, 270 Kubanochoerus, 99, 253 lacustrine sediments Sinap, 243–7 Tuscany, 365–8 Lago Mare biofacies, 357–8 Lagomeryx, 96, 99, 210 Lagomorphs, 11, 13, 15–57, 88, 104, 145, 157–64, 198–9, 265–71, 424 Laminifera (Laminiplica), 343 Laminifera (Laminiplica) cesseyensis, 345 Laminifera (Laminiplica) meini, 342 Laminifera (Laminiplica) villafranchiana, 345 Laminifera mira, 332 Langhian vegetation, 381 Lanthanotherium, 178, 185, 265, 266 Lanthanotherium sanmigueli, 144, 157, 158, 159, 160, 161, 162 Lartetomys, 97 Lartetotherium sansaniense, 104, 119, 403 Lartetotherium schleiermacheri, 170 latitudinal position, 88–9 laurel forests, 478 leafy Xora, Vallesian Central Europe, 166, 167, 170 Leiostyla, 334, 335, 347 Leiostyla capellinii, 344 Leiostyla gottschicki, 344 Leiostyla (Leiostyla), 332, 333, 343 Leiostyla (Leiostyla) austriaca, 336 Leiostyla priscilla, 342 leporids, 145, 270 Leptopelsictis aurelianensis, 117 Leucochroopsis, 332, 333, 335 Leucochroopsis apicalis, 332 Leucochroopsis dufrenoyi, 337 Leucochroopsis kleini, 335, 337 Leucochroopsis leptoloma leptoloma, 331 Leucochroopsis pisum, 331, 336 Leucochroopsis v. semenazensis, 338

Index

Leucochroopsis valentinensis, 338, 340 Ligerimys, 90, 92, 94, 95 lignites, Tuscany, 365–8 Limnocardiiae, 340 Limnoecinae, 394 Liquidambar europaea, 478 Listriodon, 98, 99, 100, 104, 114, 210, 253, 403 Listriodon splendens, 119, 172 lithostratigraphy, Oligocene-Miocene, W Eurasia, 274–98 Lokochot TuV, 42, 45, 49–51 Lomogol TuV, 41, 44–5, 49–51 Lophiomyidae, 271 Lophocricetinae, 270 Lophocricetus, 265, 266, 271 Lophocricetus complicidens, 268 Lophocricetus maeoticus, 270 Lophocricetus sarmaticus, 270 Lophocyon paraskevaidisi, 207, 210 Lucentia, 105, 106, 120 Lufengpithecus, 463, 463–74 Lutra aYnis, 225 lutrine carnivore, Italy, 192 Lycyaena, 114 Lycyaena chaeretis, 118, 218, 222 Lygeum, 382 Macaca, 17, 106, 297 Macedonia Andrews’s collection, London, 213, 226, 413 Arambourg collection, Paris, 413 macromammal succession, 208–26, 226 primate localities, 413–31 Machairodus, 102, 103, 114, 214, 215 Machairodus alberdiae, 118 Machairodus aphanistus, 118, 170 Machairodus giganteus, 118, 198, 218, 220, 222 Macrogastra, 335 Macrogastra densestriata, 344 Macrogastra loryi, 342, 344 Macrogastra multistriata, 344 Macrogastra schlickumi, 344 Macrogastra sessenheimensis, 345 macromammals Central Europe, Vallesian, 165–74

Greece, 205–31, 414 Spain, Vallesian, 113–24 Turkey, 257–8 Western Europe, 106–7 Macrotherium, 415 Macrotherium macedonicum, 224 Macrozonites casteti, 342 Macrozonites collongeoni, 342 MaWa, 393 magnetobiostratigraphy, 93, 127, 207 magnetostratigraphy lignites, Tuscany, 365–8 Oligocene-Miocene, W Eurasia, 274–98 mangroves, 291, 320, 380, 399, 477 Maragheh, Iran, 443 Marcetia santigae, 117 Maremmia haupti, 195, 196 Maremmia lorenzi, 195, 196 marine connections of Tethys, Middle Miocene, 277–81 marine excursion, Oligocene-Miocene, 12–14 marine faunal similarities, Mediterranean-Paratethys, 8–17 marine invertebrate palaeogeography, 274–308 marine sediments NE Spain, 398, 399 Tuscany, 357–63, 368–71, 371 Martes, 117 Martes andersoni, 117 Martes basilii, 117 Martes burdigaliensis, 117 Martes lefkonensis, 225 Martes mellibula, 117 Martes munki, 117 Martes paleosinensis, 117 Martes woodwardi, 222 masticatory morphology, ungulates, 437, 444–5 Mastodon pentelici, 230 mastodons, 191, 436 Mediterranean areas, Neogene vegetation changes, 378–84 Mediterranean palaeogeography, 8–17 Mediterranean xerophytic vegetation, 379, 382, 383 Mediterranean-Indian Ocean gateway, 13

499

Index

500

Megacricetodon, 86, 95–100, 103, 104, 140, 150, 181, 402, 405 Megacricetodon cf. freudenthali, 157 Megacricetodon collongensis, 97–9, 133, 137 Megacricetodon crusafonti, 86, 97, 100, 103 Megacricetodon germanicus, 86 Megacricetodon gersii, 97, 100, 103, 137 Megacricetodon gregarius, 86 Megacricetodon ibericus, 86, 87, 102, 103 Megacricetodon minor, 134 Megacricetodon minor-debruijni, 137 Megacricetodon primitivus, 97, 98 Megacricetodon rafaeli, 134 Megaderma, 142, 145 Megaderma cf. mediterraneum, 200 Megaderma cf. vireti, 158 Megaderma vireti, 164 Megalotachea, 333, 334, 337, 338 Megalotachea christoli, 337 Megalotachea delphinensis, 340 Megalotachea delphinensis tersannensis, 338 Megalotachea gualinoi, 338, 340 Megalotachea turonensis, 331, 336 Megalotachea turonensis larteti, 335 Melanoides, 339, 340, 346 Melanopsidae, 340 Melanopsis, 328, 332, 337, 339, 340, 346 Melanopsis kleini, 338, 340 Melanopsis laevigata, 337, 338 Melanopsis narzolina, 337 Melanopsis narzolina gigantea, 338 Melanopsis requenensis, 338 Melissiodon, 12, 90, 91, 95 Melissiodon cf. dominans, 94 Melissiodon schro¨deri, 90 Mellivora benWeldi, 200 Meomyini, 393 Meotian, Ukraine, 270–1 Mesembriacerus, 208, 209, 429 Mesembriacerus melentisi, 210, 211 Mesodontopsis, 343 Mesodontopsis chaixi, 341, 342, 345 Mesodontopsis doderleini, 339 Mesodontopsis heriacensis, 338, 340 Mesodontopsis ludovici, 335 Mesodontopsis nehringi, 345 mesoloph(id) length, rodents, 132–3

Mesomephitis medius, 117 Mesopithecus, 142, 163, 198–9, 296, 414, 418, 449, 456–9, 462, 463 Mesopithecus aV. pentelicus, 223, 224 Mesopithecus cf. montspessulanus, 224 Mesopithecus cf. pentelicus, 225 Mesopithecus delsoni, 214, 227 Mesopithecus monspessulanus, 229 Mesopithecus pentelicus, 198–9, 200, 221, 229, 230, 462, 463 Mesopithecus pentelicus microdon, 223 Mesopithecus pentelicus pentelicus, 223 Messinian Eurasia, 297–8 evaporites, Tuscany, 357–63, 371–2 Italy, 193, 196–7 non-marine molluscs, 339, 340, 348 salinity crisis, 86–7, 297, 322, 357–63, 381, 406 Metailurus, 438, 449 Metailurus major, 118, 198, 218, 220, 222 Metailurus parvulus, 118, 198, 218, 220, 222 Microdryomys aV. koenigswaldi, 191 Microdyromis, 146, 182 Microdyromis complicatus, 134, 137, 158 Microlophiomys vorontsovi, 271 micromammals Late Miocene Central Europe, 167, 174–85 France, 140–54 Ukraine, 265–71 Late Vallesian, Spain, 123 Miocene, Greece, 206, 207, 208 Micromeryx, 99, 106, 120, 173 Microstonyx, 104, 107, 219, 230, 253, 261 Microstonyx antiquus, 172 Microstonyx erymanthius, 120, 173 Microstonyx erymanthius brevidens, 173 Microstonyx major, 120, 173, 198, 213, 215, 217, 220, 222, 226 Microstonyx major erymanthius, 214, 215, 216, 219, 221, 227 Microstonyx major major, 223, 229 Microtia fauna, Italy, 193 Microtocricetus, 150, 181, 266, 267, 404 Microtocricetus cf. molassicus, 157 Microtoscoptes, 270, 271

Index

migration Africa-Eurasia-Africa, 8–17, 274, 281–2, 285–94, 437–8, 446–9 America-Europe 9, 13, 17, 446;, see also Bering landbridge Milankovitch cycles, 58, 66–7, 76 Miliolida, 283 Mimomys pliocaenatus zone, 346 Miniopterus, 158 Miniopterus cf. fossilis, 164 Miniopterus fossilis, 145 Miocene chronology and mammal faunas, Sinap Formation, 238–61 corals, 309–24 land mammals, Italy, 140–54 macromammals, Spain, 113–24 mammals, Central Europe, 165–85 micromammals, France, 140–54 micromammals, Ukraine, 265–71 MN units, Western Europe, 84–107 non-marine molluscs, 331–40 palaeogeography, 8–22 plate tectonics, 8–17 rodent assemblages, Spain, 127–38 shrews and palaeoclimate, 392–6 Miocene-Pliocene facies analysis, Tuscany, 355–72 Miodyromys, 266, 403 Miodyromys aegerci, 93 Miodyromys griciviensis, 265 Miodyromys hugueneyae, 93 Miogypsina intermedia, 291 Miogypsina tani, 288 Miomachairodus, 252 Miopetaurista, 104, 146, 181, 265, 266 Miopetaurista cf. crusafonti, 162 Miosorex, 142, 144, 270 Miosorex cf. grivensis, 157, 158 Miotragocerus, 99, 104, 404, 441 Miotragocerus monacensis, 121 Miotragocerus pannoniae, 121, 167, 170 MN mammalian units, re-evaluation, 84–107, 455 MN zones Aragonian rodent assemblages, 127–38 hominoid primates, 455–63 macromammals, Greece, 205–31

macromammals, Vallesian, Spain, 115 non-marine molluscs, 328–49 soricids, 391–4 Moiti TuV, 38–9, 47, 48 molluscs, 9, 11, 12, 166, 167, 278, 328–49, 366 Monoptychia monoptyx, 344 Monosaulax, 265, 270 monsoonal circulation, 24, 26–8, 34, 60, 61, 62, 405, 408 Mormopterus, 145 Mormopterus helvetica, 157 morphology cladograms, 463–74 moschids, 99, 100–1, 104, 198 mountain anoa, 445 multivariate analysis, late Miocene primate localities, 419–29 Muntiacus, 441 murids, 104, 105, 106, 296, 402–8, 405, 407 Central Europe, 182–3 France, 151–5, 162 Turkey, 261 Ukraine, 267, 270 Muscardinus, 104, 134, 146, 182, 198–9, 200, 270, 403 Muscardinus aV. vireti, 89, 198–9 Muscardinus austriacus, 157, 159, 160, 161, 162 Muscardinus davidi, 164 Muscardinus hispanicus, 157, 158 Muscardinus thaleri-hispanicus, 137 Muscardinus topachevskii, 265 Mustela majori, 195 mustelids, 209, 212, 449 Mygalinia, 267, 268 Mykhailivsky mammalian complex, 267–8 Myocricetodon, 86, 252 Myoglis, 89, 104, 134, 146, 181, 266, 403 Myoglis meini, 137, 157, 159 Myoglis ucrainicus, 265 Myomiminae, 136 Myomimus, 136, 147, 182, 252, 403 Myomimus dehmi, 134, 158, 162 Myotis, 145, 157, 159, 162, 163, 164 Myotis antiquus, 157, 158–9 Myotis boyeri, 161, 164 Myotis cf. boyeri, 158, 200 Myotis cf. murinoi¨des, 159

501

Index

502

Myotis murinoi¨des, 158 Myoxus, 178 nannoplankton, calcareous, 276 Nannospalax, 270, 271 Nannospalax compositodontus, 271 Negulus, 332, 333, 334, 335, 339, 341, 343, 347 Negulus bleicheri, 342 Negulus raricostatus, 332 Negulus suturalis gracilis, 336 Negulus truci, 344 Negulus villafranchianus, 344 Neocricetodon, 105, 106, 142, 157, 158, 159, 163, 164 Neocricetodon grangeri, 150 Neocricetodon lavocati, 150, 164 Neocricetodon lucentensis, 164 Neocricetodon skoXeki, 150, 159, 161, 162, 163 Neomyini, 393, 395 ‘Neopithecus’ brancoi, 478 Neritaea, 346, 347 Neritidae, 340 Nesiotites, 393 Neumayria, 346 Neurada, 382 nimravids, 103, 104, 403 Nisidorcas, 229 Nisidorcas planicornis, 213, 214, 215, 217, 227 Nitraria, 382 Norbertia hellenica, 224 Nordsieckia, 340 Nordsieckia Wscheri, 342 Nuragha schreuderae, 191 obsidian, 29, 31 Occitanomys, 105, 106, 151, 152, 271 Occitanomys adroveri, 164 Occitanomys cf. adroveri, 151, 164 Occitanomys clauzoni, 151, 152, 159, 160, 161, 162 Occitanomys faillati, 142 Occitanomys hispanicus, 105, 150, 151, 152, 157, 159, 161, 404 Occitanomys sondaari, 105, 150, 151, 154, 161, 162

Ocean Drilling Program, 23–4 oceanic circulation, Miocene, 397, 399–401, 402 oceanic currents, Africa, 61, 64, 68 Ochotonidae, 267, 270 Oioceros, 213, 429, 444 Oioceros aV. atropatanes, 213 Oioceros rothi, 221, 228 Oioceros wegneri, 217, 228, 441 okapi, 437, 443 Olea, 379, 382, 383 Oleacinidae, 329, 331 Oligocene, plate tectonics, 8–17, 391–2 Oligocene-Miocene palaeogeography, 8–22, 274–308 Omphalosagda subrugulosa, 331 Opeas, 336, 339 Opeas minutum, 335 open-habitat vegetation, 406, 413–31, 418, 426, 481 orbitally tuned ages, 48–51 Orbulina, 277 Orbulina universa, 277, 278 Oreopithecus, 462–81 Oreopithecus bambolii, 193, 194, 195, 197, 365, 463 Orycteropus, 254, 436 Orycteropus cf. gaudryi, 200 Orycteropus gaudryi, 218, 223, 438 Orycteropus pottieri, 209, 213 Oryx, 449 ostracods, 363 Otavipithecus, 467 Otonycteris, 145, 158 Ouranopithecus, 295–6, 414, 461, 462 Ouranopithecus macedoniensis, 208, 209, 211, 212, 213, 226 Ouranopithecus-Graecopithecus, 449 Ouzocerus, 209, 211, 213, 429 Ouzocerus cf. gracilis, 224 Ouzocerus gracilis, 209, 211 Ouzocerus pentalophosi, 212 Owen Ridge, Arabian Sea, 25–6 oxide chemistry, tephra, 25, 37–51 Pachytragus, 254, 445 Pachytragus crassicornis, 441 Pachytragus laticeps, 441

Index

Paenelimnoecus, 144, 178, 392 Paenelimnoecus crouzeli, 157 Paenelimnoecus repenningi, 157, 158, 159–60, 161, 162 Paidopithex rhenanus, 169 Pakistan, Dera Bughti, 13 palaeobotany, 288–90, 478–9 Africa, 69, 71–3, 76 Central Europe, 166, 167, 170, 174, 184–5 Mediterranean Miocene, 440–3 United States, 74–6 Palaeochoerus, 13 palaeoclimate, 25, 27–8 Central Europe, 166–7, 174, 184–5 France, 153 marine invertebrate evidence, 274–98 Spain, 122–3, 397–408 palaeoclimate analysis, Aragonian rodent assemblages, 127–38 palaeoclimate change, 477 and mammal turnover, 397–408 non-marine mollusc evidence, 328–49 soricids as indicators, 390–6 palaeoclimate reconstruction by pollen records, 378–84 by sedimentary facies analysis, 355–72 palaeodietary research, Pikermian ungulates, 439, 444–5 palaeoecology hominoid primates, 474–9 Pikermian Biome, 436–49 rodent assemblages, 127–38 palaeoenvironmental change and faunal diversity, 115–24 Late Miocene, France, 153 and mammal turnover, 404–8 palaeoenvironmental diVerences, correlation, 87–8 palaeoenvironmental reconstruction, climatic perspectives, 55–77 palaeoenvironments Late Miocene, Ukraine, 265–71 Late Miocene primate localities, 413 Palaeogale, 206, 210 palaeogeography, 8–79 marine invertebrate evidence, 274–308 Oligocene-Miocene, 8–22, 274–308 Ukraine, 271

Palaeoglandina, 332, 334, 337, 338, 341, 347 Palaeoglandina aquensis, 331 Palaeoglandina gracilis, 331, 335 Palaeoglandina montenati, 346 Palaeoglandina paladilhei, 342 Palaeolaginae, 267 palaeolakes, Africa, 69, 70 palaeomagnetic analysis, Eastern Ebro Basin, 93 palaeomagnetic data, Aragonian rodent assemblages, 127 palaeomagnetic reversal stratigraphy, Sinap Formation, 247–9 Palaeomanis, 438 palaeomerycids, 96, 100 Palaeomeryx, 96, 172, 206 palaeomeryx cf. kaupi, 210 Palaeomeryx magnus, 121 Palaeomys, 176, 181, 265 Palaeomys castoroides, 176 Palaeoplatyceros hispanicus, 120 Palaeoreas, 216, 254, 261, 444 Palaeoreas lindermayeri, 214, 220, 221, 223, 441 Palaeoreas zouvei, 213, 227 Palaeoryx, 106, 211, 212, 229, 438, 445 Palaeoryx ‘aV. stutzeli’, 219 Palaeoryx cf. pallasi, 219 Palaeoryx pallasi, 121, 217, 219, 220, 221, 228, 441, 449 palaeosols, 290, 292, 293, 295, 369, 438–40 palaeotemperatures from Miocene coral record, 309–24 Oligocene-Miocene, W Eurasia, 274–98 sea water, 8, 284, 293, 309–24 Palaeotragus, 101, 102, 121, 253, 449 Palaeotragus cf. coelophrys, 211 Palaeotragus coelophrys, 212, 217, 441, 445 ‘Palaeotragus’ primaevus, 449 Palaeotragus quadricornis, 441 Palaeotragus rouenii, 211, 213, 214, 217, 221, 223, 441, 445, 449 palaeovegetation, 478–9 mapping, 378–84 see also vegetation Palaina, 334, 335, 336, 339 Palerinaceus, 211

503

Index

504

Paludolutra aV. etruria, 195 Paludolutra campanii, 195, 196 Paludolutra maremmana, 195 Paludotona cf. etruria, 195 Paludotona etruria, 194, 195 palustrine deposits, Tuscany, 363–5, 367 Pan, 449, 465–73 Pannonian Basin, 17 Pannonicola, 142 Papyrotheca, 335 Parabos, 107, 198, 200 Paracamelus, 106 Paracervelus, 198 Parachleuastochoerus, 101, 104, 119–20, 404 Paracricetodon, 12 Paraethomys, 106, 198–9 Paraethomys anomalus, 200 Paraethomys cf. anomalus, 198–9 Paraglirulus, 89, 104, 146, 182, 266, 403 Paraglirulus cf. werenfelsi, 265 Paraglirulus werenfelsi, 134, 157, 158, 159 Paralutra, 117 Paramachaerodus ogygius, 170 Paramachairodus, 114 Paramachairodus ogygia, 118 Paramachairodus orientalis, 118, 173, 222 Parapodemus Central Europe, 167, 182 Ukraine, 267, 270, 271 Western Europe, 105, 106, 151, 152, 159, 195, 196 Parapodemus barbarae, 105, 151, 164 Parapodemus cf. barbarae, 153 Parapodemus gaudryi, 270 Parapodemus lugdunensis, 105, 151, 152, 153, 159–63, 267, 270 Parapodemus pasquieri, 151, 153, 160, 162, 163 Parapodemus schaubi, 216, 227 Parasaidomys, 106 Parataxidia maraghana, 218 Paratethys, 9–17, 268–71, 291, 293 Pareptesicus, 145 Pareptesicus priscus, 158–9 Parmacella, 334, 339, 340, 346, 347 Parmacella sayni, 338, 340 Parmacellidae, 340

Parurmatherium rugosifrons, 441 peatland formation, 367–8, 372 Pecora, 253 percrocutids, 167, 173, 174 Peridyromys, 252 Peridyromys murinus, 93, 94 perissodactyls, 98, 119, 170, 198, 403, 417 Persian Gulf, 16 Petauristinae, 266–7, 270 Petenyia, 144, 176, 178 Petenyia dubia, 158, 159–60, 161, 162, 164, 176 Phillyrea, 382 photosynthetic symbionts, 282–3, 311, 315 phylogenetic hypotheses, Miocene hominoid primates, 463–74 Picea, 382, 383 Pikermi, 226, 228–9, 437 Pikermian Biome, 436–9 Pikermian Province Xora, 296 Pilocervus pentelici, 217 Pinaceae, pollen, 378–9 Pipistrellus, 145, 161 Pireddamys rayi, 191 Pistacia, 379, 382 planktonic foraminifera, 277–81 Planorbarius, 338 Planorbarius aV. villatoyensis, 337 Planorbarius villatoyensis, 337 plant remains, Vallesian, Central Europe, 166, 167 plant species prediction, 73–5 Platanus, 478 plate tectonics, 8–17, 70–1, 405, 407, 475, 481 Platycarya, 381, 383 Pleistocene aeolian deposits, 363–5 plate tectonics, 70–1 shrews, 393–4, 395 Plesiodimylus, 178, 185, 265 Plesiodimylus cf. chantrei, 157, 158, 160, 161 Plesiodimylus chantrei, 144, 157 Plesiogulo, 117, 222 Plesiogulo crassa, 198, 215 Plesiomeles cajali, 117 Plesiomeles pachecoi, 117

Index

Plesiopliopithecus, 16, 459 Plesiopliopithecus auscitanensis, 459 Plesiopliopithecus lockeri, 459 Plesiopliopithecus rhodanica, 460 Pleurodiscus, 335 Pleurodiscus falciferus, 332 Pliocene aeolian deposits, 363–5 environmental change, Africa, 66–77, 447 marine transgression, 17 non-marine molluscs, 340–9 shrews, 393, 395 tephra correlations, 23–51 vegetation changes, 406 Pliocervus, 106–7, 220 Pliocervus graecus, 224 Pliocervus pentelici, 221, 222, 223, 441 Pliohyrax, 438 Pliohyrax graecus, 119, 218, 220, 221 Pliopetaurista, 105, 146, 167, 175, 181 Pliopetaurista bressana, 157, 159, 160, 162, 163 Pliopetes, 181 Pliopithecus, 16, 99, 142, 292, 459 Pliopithecus antiquus, 116, 455, 460 Pliopithecus platyodon, 455 Pliopithecus priensis, 461 Pliopithecus vindobonensis, 459 plioplethicids, 169, 172, 174, 292, 417, 456–61, 477 Pliospalax, 17, 252 Plioviverrops, 118 Plioviverrops cf. guerini, 215 Plioviverrops faventinus, 200 Plioviverrops orbignyi, 214, 215, 216, 218, 222 Plithocyon armagnacensis, 116 Poaceae, 382 Poland Gomphotherium, 15 Miocene primate taxa, 457 non-marine molluscs, 333, 335 Polar Sea, 9 pollen Africa, 69, 71–3 as evidence for palaeoclimates, 288, 292, 293, 295, 297, 365

Tuscan sediments, 363, 365, 371, 372 Vallesian, Central Europe, 167 in vegetation mapping, 378–84 Polloneria, 343 Polloneria pliocenica, 345 Pomatias, 333, 334, 335, 339 Pomatias conicus, 336 Pomatiasidae, 331 Pongo, 463, 466 Postpalerinaceus, 200 Postpalerinaceus vireti, 144, 158, 160, 164 Postpotamochoerus, 438 Potwar Plateau see Siwaliks Praearmantomys, 96 Praeorbulinia, 277 precipitation, 55, 58–60, 62–4, 66–73, 477 primates Greece, 211, 212–17, 223, 224, 227, 229, 231, 413–31 Late Miocene, France, 142 Messinian, Italy, 198–9 migration waves, 14, 16 Turkey, 252, 261 Vallesian, Spain, 114, 116, 403, 404–5 see also hominoids proboscideans, 96, 98, 116, 198, 206, 404 ‘Proboscidian Datum’, 475 Procapreolus, 173, 198 Procervulus, 96 Proconsul, 290, 449, 465–73 Prodamaliscus, 438 Progenetta, 104 ProgiraVa, 95 Progonomys, 88, 104, 151, 153, 161, 167, 182, 252, 261, 267, 403 Progonomys castillae, 181 Progonomys cathalai, 152, 157, 209, 211 Progonomys cf. cathalai, 152, 159 Progonomys woelferi, 404 Prolagus, 89, 198–9, 268, 270, 271 Prolagus cf. sorbinii, 200 Prolagus crusafonti, 89, 157, 158, 159, 161, 162, 163, 164 Prolagus michauxi, 89, 198–9 Prolagus ningensis, 157 Promeles, 117, 173 Promeles macedonicus, 225 Promeles palaeattica, 173, 218, 222

505

Index

506

Promephitis, 225 Promephitis larteti, 218, 222 Promephitis pristinidens, 117 Promimomys, 182 Proochonota, 252, 267, 268 Propotamochoerus, 101, 219, 253 Propotamochoerus cf. hysudricus, 214, 217, 227, 228 Propotamochoerus palaeochoerus, 120, 172 Proputorius, 117 Proscapanus, 265 Prosinotragus, 444 Prosinotragus kuhlmanni, 441 Prososthenia, 337, 346, 347 Prospalax, 150, 182, 271 Prospalax petteri, 159, 160, 161, 162, 163 Prostrepsiceros, 220, 226, 229, 254, 261, 429, 438, 444 Prostrepsiceros aV. houtumschindleri, 218, 230 Prostrepsiceros houtumschindleri, 441 Prostrepsiceros houtumschindleri syridesi, 213 Prostrepsiceros rotundicornis, 214, 220, 221 Prostrepsiceros valliensis, 211 Prostrepsiceros woodwardi, 220, 231 Prostrepsiceros zitteli, 214, 215, 216, 217, 228 Protatera, 17, 86, 106 Protictitherium, 100, 104, 252 Protictitherium aV. gaillardi, 211 Protictitherium cf. crassum, 212 Protictitherium crassum, 118, 173, 224 Protodrepanostoma, 332, 333, 335, 343 Protodrepanostoma bernardii, 342, 345 Protodrepanostoma nordsiecki, 332 Protodrepanostoma plioauriculatum, 345 Protoryx, 209, 212, 215, 217, 253, 254, 438 Protoryx carolinae, 209, 221, 228 Protoryx crassicornis, 209, 228 Protoryx laticeps, 209, 217, 228 Protoryx solignaci, 209 Protozapus, 178 Protragelaphus, 217, 438, 444 Protragelaphus skouzesi, 220, 221, 229, 441 Protragelaphus theodori, 223, 224, 229 Protragocerus, 101, 102, 104, 121 Pseudaelurus, 206, 252

Pseudaelurus cf. lorteti, 210 Pseudaelurus lorteti, 103, 118 Pseudaelurus quadridentatus, 103, 118, 210 Pseudaelurus transitorius, 95 Pseudaelurus turnauensis, 95 Pseudarctos, 116 Pseudarctos bavaricus, 172 Pseudidyla, 335, 340 Pseudidyla moersingensis, 335 Pseudochloritis, 332, 333 Pseudochloritis incrassatus, 335 Pseudochloritis mollonensis, 338, 340 Pseudocricetus, 270, 271 Pseudocricetus antiquus, 270 Pseudocricetus kormosi, 271 Pseudocricetus orienteuropaeus, 270 Pseudocyon sansaniensis, 172 Pseudodryomys ibericus, 94, 95 Pseudodryomys simplicidens group, 94, 95 Pseudofahlbuschia, 136 Pseudoleacina, 335 Pseudoleacina (Paraglandina) christoliana, 339 Pseudoleacina (Paraglandina) oligostropha, 332 Pseudomeriones, 86, 106, 182, 271 Pseudomeriones abbreviatus, 271 Pseudomonacha, 336 Pseudomonacha zippei, 332 Pseudooleacina (Paraglandina) christoliana, 331 Pseudorotalia, 291 Pseudotheridomys, 91, 94 Pseudotragus, 254 Pseudotragus capricornis, 217, 441 Puisseguria, 343 Puisseguria idanica, 341 Puisseguria kowalczyki, 345 Puisseguria zilchi, 345 Pupillidae, 329 Quercus ilex, 382 radioisotopic dating, 48–51, 249 ramapithecines, 293–4 Ramapithecus, 293–4 Ramblian, 85, 90–4, 392 Ramys, 147

Index

Ramys multicrustatus, 157, 162 rarefaction method, 131 reciprocal of Simpson’s index, 127 Red Sea graben, 13 Reduncini, 106, 444 ‘reef corals’, 313 reef patterns, Miocene, 322, 324 Regiclausilia, 335 reptiles, Italy, 191 Reticulofenestra pseudoumbilica, 48 Rhagapodemus, 151 Rhagapodemus primaevus, 164 Rhamnus, 379 Rhenopithecus epplesheimensis, 169, 174 Rhinegraben, 13, 15 rhinoceroses, 14, 172, 173, 403, 404, 444 rhinocerotids, 11, 98, 99, 104, 114, 133, 422 Eurasia, 417, 436 Greece, 207, 210, 211, 212, 213, 215, 216, 219 Rhinolophus, 145, 200 Rhinolophus cf. kowalskii, 200 Rhinolophus csakvarensis, 145, 157, 158, 160, 161, 163, 164 Rhinolophus lissiensis, 145, 164 Rhizospalax, 13 Rhodanomys oscensis, 92 Rhodanomys schlosseri, 90, 92 Rhodanomys transiens, 90, 92, 93 Rhodanomys-Ritteneria lineage, 90, 92–3 Rho ˆ ne Valley, Upper Miocene micromammals, 140–54 richness, 64–9, 74–6 fossil corals, 309–24 non-marine molluscs, 328 Sinap Formation, Turkey, 257–8, 260–1 see also species richness Ritteneria manca, 90, 92, 94 Ritteneria molinae, 92, 94 rodent dental features, 132–3 rodent habitat preferences, 132–8 rodents Aegean, 17 Central Europe, 167, 175–8, 181–3 France, 157–64 Greece, 227 Italy, 194, 198–9 Spain, 84–107, 127–38, 402–8

Ukraine, 265–71 Rotaliida, 283 Rotundomys, 104, 105, 150, 163, 404 Rotundomys bressanus, 159, 160, 161 Rotundomys cf. bressanus, 161 Rotundomys cf. montisrotundi, 157 Rotundomys montisrotundi, 158 Rudapithecus, 464 Ruemkalia, 176 Rumina, 340, 346 Rumina cf. decollata, 339 Rumina decollata, 339 Ruminantia, 200 Ruscinian, 150, 172, 194, 196, 340–2, 393 Ruscinomys, 154 Ruscinomys cf. lasallei, 200 Ruscinomys schaubi, 150, 164 Sabdelictus crusafonti, 117 sabkhisation, 362 salinity, 17, 86–7, 293, 297, 322, 357–63, 381 sambar, 444, 445 Samos, 217–18, 227–8, 437, 439–40 Samotherium, 443 Samotherium boissieri, 215, 217, 228, 441, 445 Samotherium cf. boissieri, 224 Samotherium major, 440, 445 Samotherium neumayri, 441 ‘Samotherium’ pamiri, 209 Samotragus, 429 Samotragus crassicornis, 209 Samotragus occidentalis, 200 Samotragus pilgrimi, 121 Samotragus praecursor, 209, 210, 211, 212 Sanitherium, 206 Sanitherium slagintweiti, 210 Sansanosmilus, 103, 252 Sansanosmilus jourdani, 118 Sapotaceae, 383 Sardomys antoniettae, 191 Sardomys dawsoni, 191 Sarmatian, Ukraine, 265–71 Sarmatomys podolicus, 266 savannas, 295, 422–4, 436–49, 479 Savic tectonic phase, 14 Scaptonyx, 177 Schizochoerus, 104, 106, 120, 253

507

Index

508

Schizogalerix, 142, 175, 178, 252, 265 Schizogalerix zapfei, 142–4, 159, 160 Schizotheriinae, 415–19 Schlickumia, 333, 341, 343, 347 Schlickumia alicantensis, 346 Schlickumia aquensis, 331, 336, 337 Schlickumia bottinii, 345 Schlickumia gaspardiana, 342 Schreuderia, 105 Sciuridae 136, 145–6, 153;, see also Xying squirrels Sciurotamias 265, 267, 270;, see also Spermophilinus Scleractinia, 309–24 Scutella lusitanica, 278 Scutella rotundaeformis, 278 sea levels, Miocene, 322, 399–401 sea water palaeotemperatures, 8, 284, 293, 309–24 seasonality, 58, 60, 323, 371, 408, 447, 463, 475–6, 480, 481 sediment distributions, and palaeogeography, 8 sedimentary facies analysis, 355–72 sedimentation rates, 127 sediments, Sinap Formation, 243–7, 258–60 Semigenetta ripol, 117 Semnopithecus eppelsheimensis, 169 Sequoia, 380, 382, 477, 478 Serravallian regression, 16 Serrulastra, 333, 335 Serrulastra amphiodon, 332 Serrulastra brandti, 339 Serrulastra michaudi, 342 Serrulastra polyodon, 332 Serrulella, 333, 335, 343 Serrulella anodon, 344 Serrulella clessini, 335 Serrulella decemplicata, 344 Serrulella michelottii, 342 Serrulella schwageri, 332 Serrulella truci, 344 Shanxsi, China, 443 shrews see soricids ‘sigmodon event’, 404 Similarity CoeYcient (SC), 37–8, 47 similarity measurements, Vallesian large mammals, 114–15

Simocyon primigenius, 117, 220, 222 Simocyon simpsoni, 116 Sinap Formation, 238–61 chronology, 247–9 geology, 241–7 mammal succession, 250–8 richness, 257–8, 260–1 sediments, 243–7 turnover, 257–8 Sinapodorcas, 254 Sinictis pentelici, 222 Sivaonyx lluecai, 117 Sivapithecus, 294, 404–5, 463–78 Sivatherine giraYds, 113 Siwaliks, 14, 397, 401, 402, 404–5, 405, 408, 454 Slovakia Miocene primate taxa, 457 non-marine molluscs, 339 Sminthozapus, 178 Sminthozapus janossy, 89 soil carbonates, 438–40 solar radiation, and climate change, 57–8, 66–8 Soosia godarti, 342 Soosia monikae, 345 Soosia pseudopalnorbis, 341 Sorex, 392 soricids, 144–5, 164, 178, 191, 194, 195, 200, 266, 271 as palaeoclimatic indicators, 390–6 Soricinae, 392–4 Soricini, 392, 394, 395 Soriculus, 176 Spain mammal turnover and global climate change, 397–408 Miocene primate taxa, 456 MN mammal units, 84–107 non-marine molluscs, 331, 337–8, 341, 346, 347 species numbers Aragonian rodent fauna, 131, 135–8 Late Miocene, Central Europe, 183–4 species richness, 64–6, 74–6, 240 Aragonian rodent faunas, 127, 130–8 Central European micromammals, 178–81, 182–3

Index

woody-plants, 64–6, 74–7, 477 Spermophilinus 145, 181, 211, 252; see also Sciurotamias Spermophilinus cf. bredai, 157, 158, 159, 160 Spermophilinus cf. turolensis, 163–4 Spermophilinus turolensis, 162 Sphaeoidinellopsis Zone, 196 Sporadotragus parvidens, 217, 221, 228, 442 spruce, 379 Squared Chord Distance Parameter, 47 Stegolophodon wahlheimensis, 172 Stegotetrobelodon lehmanni, 172 Stehlinocerus, 101, 120 Stenailurus teilhardi, 118 steneoWber, 265 SteneoWber jaegeri, 224, 231 Stephanomys, 106, 151 Stephanomys debruijni, 200 Stephanomys primaevus, 151 Stephanomys ramblensis, 105, 151, 164 Stephanomys stadii, 151, 164 Stephanorhinus, 119, 253 steppes, 271, 298, 379, 383 stratigraphic correlations, 9, 276 Strobilops, 335, 337 Strobilops (Strobilops), 332, 333, 334, 335, 336, 338, 339, 340, 341 Strobilops (Strobilops) costatus, 335 Strobilops (Strobilops) labyrinthiculus, 342 Strobilops (Strobilops) pappi, 337 Strobilops (Strobilops) romani, 342 Strobilops (Strobilops) tiarulus, 337 Strobilopsidae, 329, 331, 333, 340 Struthio caratheodoris, 438 Stylocricetus, 267 Stylocricetus meoticus, 270 Subtropical High Pressure Cells (STHP), 60, 68 subtropical vegetation Neogene NE Spain, 398–9, 403 Neogene Southwestern Europe, 379, 383 Subulinidae, 333 Succineidae, 335 suids, 95, 96, 98, 101, 104, 114, 403, 424 Central Europe, 172, 173, 174 Greece, 226, 227, 228, 229, 415, 417, 418

Italy, 195 Turkey, 261 Sulimskia, 268, 393 Switzerland, non-marine molluscs, 335 symbiosis, 282–3, 311, 315 synchronism, 87–8 Tacheocampylaea, 346 takins, 445 Talpa, 178, 185 Talpa cf. minuta, 157 Talpa gilothi, 157, 158, 159, 160, 161, 163, 164 talpids, 145, 191, 267, 270 Tamias, 145, 159, 160, 163 Tanzania, tephra correlations, 23 Taphozous, 145, 158 tapirids, 103, 114, 173, 402–4, 417, 418, 422 Tapiriscus pannonicus, 167, 173 Tapirus arvernensis, 198 Tapirus cf. arvernensis, 198 Tapirus priscus, 119, 170 Tasodon sansaniensis, 117 Taucanamo, 253 Taxodiaceae, 382–3 Taxodium swamps, 380, 381, 382, 383, 477 taxonomic comparisons, late Miocene primate localities, 414–31 taxonomic richness, fossil corals, 309–24 tayassuids, 14 tectonic subsidence, 368, 372 teeth analysis for dietary intake, 439–40 microwear analysis, 437, 445–6 rodents, 404 ungulates, 436, 437, 444–6 temperature and climate change, 55–6, 76–7 and rodent habitats, 132–4 see also palaeotemperatures tephra correlations, Pliocene, 23–51 terrestrialisation, 367, 372 Teruelia, 95, 96 Teruelia adroveri, 95 Tethys Ocean, 8, 9, 70, 274, 276–8, 313, 407, 475 Tethytragus, 99, 100–1, 254 Tethytragus cf. koheri, 210

509

Index

510

Tethytragus langai, 121 Tetralophodon, 102, 103, 211, 213 Tetralophodon atticus, 221 Tetralophodon longirostris, 116, 170 Thalassictis, 101, 102, 114, 118, 252 Thalassictis gr. chaeretis-macrostoma, 200 Thalassictis hipparionum, 118, 198–9 Thalassictis hyaenoides, 198–9 Thalassictis montadai, 118 Thalassictis robusta, 118, 170 Thalassictis wongi, 211 Thaumastocyon dirus, 116 Theodoxus, 346, 347 Thiaridae, 340 Tibetan Plateau uplift, 405, 407–8 Tiliaceae, 185 Titthodomus, 332, 336 Titthodomus koeneni, 332 Tortonian transgression, 17 Tournouerina, 346 Tragoportax, 104, 114, 213, 226, 227, 253, 261, 444 Tragoportax amalthea, 216, 219, 220, 221, 228, 229, 230, 441 Tragoportax cf. amalthea, 224 Tragoportax gaudryi, 121, 221, 222, 224, 228 Tragoportax rugosifrons, 213, 215, 217, 227, 228, 441 Tragoreas oryxoides, 217, 218 tragulids, 96, 103, 402, 405, 415, 417, 418, 422, 424 Triceromeryx, 100 Triptychia, 343 Triptychia geisserti, 344 Triptychia leobersdorfensis, 337, 339 Triptychia mastodontoohila, 344 Triptychia (Milneedwardsia) lageti, 338, 340 Triptychia (Milneedwardsia) lartei, 335 Triptychia (Milneedwardsia) sinestrorsa, 342 Triptychia (Milneedwardsia) terveri, 342 Triptychia schlickumi, 344 Triptychia (Triptychia) bacillifera, 335 Triptychia (Triptychia) bourguignati, 338, 340 Triptychia (Triptychia) grandis, 335 Trischizolagus cf. maritzae, 200

Trocharion albanense, 117 Trochictis narchisoi, 117 Trogontherium, 146, 181 Trogontherium minutum, 146, 157, 161 Trogontherium rhenanum, 157, 159, 162, 163 Trolliella, 335 tropical coral, 309 tropical marine excursions, Oligocene-Miocene, 12–14 tropical vegetation, 381, 383, 398–9 Tropidomphalus, 332, 333 Truciella ballesioi, 342 Tsuga, 383 Tubulidentata, 413, 424 Tudorella, 333, 338, 340, 341 Tudorella baudoni, 342 Tudorella draparnaudi, 331, 338 Tudorella draparnaudi minor, 337 tuVs, East Africa, 24, 25, 31–51 Z-Tulu Bor TuV, 43, 45–7, 49–51 Turcocerus, 253 Turgai Strait, 9, 11, 287 Turiacemas, 106, 107, 120 Turkana Basin tuVs, 24, 25, 37–51, 69 Turkey, Miocene primate taxa 458; see also Sinap Formation Turkomys, 86 turnover and global climate change, NE Spain, 397–408 turnover statistics, macromammals, Turkey, 257–8 Turolian Central Europe macromammals, 165–74 micromammals, 167, 174–85 Eurasia, 296 France, micromammals, 140–54 Germany, 170 Greece, macromammals, 205–31 Italy, 191–7 Macedonia, primate palaeoenvironments, 413–31 non-marine molluscs, 337–40 Southern Europe, 86 Spain, MN units, 85, 87, 88–9, 105–7 Tusco-Sardinia palaeobioprovince, 191, 193–6, 197

Index

Tyrrenolutra helbingi, 195, 196 Tyrrenotragus, 195 Tyrrenotragus aV. gracillimus, 194 Tyrrenotragus gracillimus, 194, 195 Uganda, tephra correlations, 23, 50 Ukraine, late Miocene small mammals, 265–71 Ulmaceae, 184–5 Ulmus, 381 Umbrotherium azzarolii, 195 United States, species richness prediction, 74–6 upwelling indices, 34 Urotrichini, 265 Urotrichus cf. dolichochir, 145, 157 Ursavus, 114, 218 Ursavus brevirhinus, 116 Ursavus depereti, 116, 167, 173 Ursavus primaevus, 116, 170 ursids, 96, 102, 114

and climate, 64–6, 71–3 and mammal turnover, 404–8 mosaic, Africa, 290, 292, 293 Vallesian, Central Europe, 166, 167 see also palaeovegetation ‘Ventian’ faunas, 86 Vertiginidae, 329, 333, 340 Vertigo (Vertigo) diversidens, 335 Vertigo (Vertigo) nouleti, 342, 344 Vespertillionidae, 145, 160, 163 Veterilepus, 268, 270 Veterilepus hungaricus, 271 Veterilepus lascarevi, 270 vicariance biogeography, hominoid primates, 454–81 Villafranchian faunal remains, Tuscany, 363, 366, 369 Viverra, 198–9 Viverra sansanensis, 118 viverrids, 198–9, 207 volcanic ash correlation, Pliocene, 23–51

Valerymys, 214 Valerymys turoliensis, 195 Valle`s-Penede`s Basin, 113–24, 397–408, 461 Vallesian Central Europe macromammals, 165–74 micromammals, 167, 174–84 Crisis, 2, 88–9, 104, 240, 403, 405, 418–19 France, micromammals, 140–54 Greece, macromammals, 205–31 Italy, 191, 192 Macedonia, primate palaeoenvironments, 413–31 MN unit re-evaluation, 85, 87–9, 102–5 non-marine molluscs, 336–9 shrews, 394–5 Spain, macromammals, 113–24 Turkey, 261 Ukraine, 265–71 Vasseuromys, 92–3, 147, 182 Vasseuromys pannonicus, 161, 164 Vasseuromys rugosus, 93 ‘Vasseuromys’ thenii, 93 vegetation change to open-habitat, 406

wapiti, 444, 445 Wargolo TuV, 38–40, 44, 48 water-energy dynamics, 56, 64–6, 76–7, 477 Wenzia, 333 Wenzia ramondi, 331 Western Europe Late Miocene faunal province, 418, 424–6, 430 Miocene primate taxa, 456 Neogene vegetation changes, 378–84 palaeoclimate change and non-marine molluscs, 328–49 Vallesian crisis, 2, 88–9, 104, 240, 403, 405 wet and dry provinces, Spain, 96 wet and dry rodents, Aragonian, Spain, 132–8 wildebeest, 445 wind inXuence on African climate, 62–4, 68 woodland habitats, 174, 261, 295, 403, 406–8 African, 72, 290, 293, 422–4, 480 Asian, 295, 422–5 Eurasian, 174, 261, 481 Pikermian Biome, 436–46

511

Index

512

woody-plant species richness, 64–6, 74–6, 477 Xenohyus, 95, 96 Xerotricha, 339 Zanclean vegetation, 381–5 Zapodidae, 12, 149, 150, 270

Zarafa, 207 Zelkova, 381 Zizyphus, 382 Zonitidae, 329, 331 Zonitoides wenzi, 338 zooxanthellate corals (z-corals), 282, 310–24, 477, 478 Zygolophodon borsonii, 198