Quick viewing(Text Mode)

Young-JR-1987-Phd-Thesis.Pdf

Young-JR-1987-Phd-Thesis.Pdf

NEOGENE CALCAREOUS NANNOFOSSILS FROM THE MAKRAN REGION OF PAKISTAN AND THE INDIAN OCEAN

To Christopher Tarquin Teale

A thesis submitted for the degree of Doctor of Philosophy of the University of London and for the Diploma of membership of Imperial College.

Jeremy Young BA MSc DUC Geology Department September 1987 Royal School of Mines Imperial College London SW7 ABSTRACT A biostratigraphical study has been made of Neogene calcareous nannofossils from the Makran accretionary terrane of South West Pakistan. They are used to help reconstruct the stratigraphy and geological evolution of the region. Comparative material from Deep Sea Drilling Project Sites in the Indian Ocean (Legs 23-26), has also been examined. In particular Reticulofenestra size variation patterns in the Miocene are documented; a Late Miocene (NN10) "small Reticulofenestra interval” is identified; assemblages from above and below it are differentiated; and the biostratigraphic, biological and taxonomic implications of the trends are discussed; the biostratigraphic results are applied in the Makran study. In addition the biological literature on nannoplankton has been comprehensively reviewed, and is used as the basis for analysis of: nannoplankton functional morphology; structural development; and intraspecific variation. The analysis of structural development has produced a predictive model of heterococcolith morphology. This model is applied and tested via: detailed investigation of reticulofenestrid structure, and biometric variation; discussion of the structure of other nannofossils; and development of a computer program for drawing elliptical coccoliths. Analysis of intraspecific variation in living nannoplankton, and evolution in calcareous nannofossils, has lead to significant reassessment of species concepts. It is recommended that more use is made of infraspecific taxa, and in particular varieties, in nannofossil taxonomy. These principles are applied in the systematics section (where most Neogene nannofossil taxa are discussed and illustrated). Forty recombinations of species as varieties are recommended. The taxonomic revisions do not involve any radical name changes, but result in a system that describes nannofossil variation more accurately than the current system. It is also more flexible, and should be easier to learn. Furthermore since it is more closely related to biological species concepts it should form a sounder base for studies of nannofossil evolution, diversity variation, and palaeoecology.

1 C.QN.TSNXS.

Abstract 1 Contents and List o£ Figures 2

1. INTRODUCTION 8 1.1 Background to and development o£the project. 8 1.2 Structure and content of thethesis. 9 1.3 Acknowledgements. 11

PART A - PALABOBIOLOGY 2. INTRODUCTION TO NANNOPLANKTQN BIOLOGY 2.1 INTRODUCTION 13 2.2 HIGHER CLASSIFICATION, THE NATURE OF 14 2.3 LIFE-CYCLES, AND 17 2.4 CELL-WALL COMPONENTS AND ORGANISATION 19 2.4.1 Plasmalemma and columnar layer. 19 2.4.2 Organic scales. 22 2.4.3 Heterococcoliths. 22 2.4.4 Holococcoliths. 23 2.4.5 Partially calcified coccoliths. 23 2.5 COCCOLITHOGENESIS 24 2.5.1 Intracellular coccolithogenesis. 24 2.5.2 Extracellular calcification. 25 2.6 LITERATURE SOURCES 26 2.7 GLOSSARY OF TERMS USED IN BIOLOGICAL DESCRIPTION OF 28 3. FUNCTIONAL MORPHOLOGY 3.1 RELATIONSHIP OF COCCOLITHS AND ORGANIC SCALES 31 3.2 PROTECTION 32 3.3 FLOTATION 34 3.4 LIGHT REGULATION 38 3.5 ENVIRONMENTAL BUFFERING 39 3.6 DISCUSSION 40 4. CQCCQLITH STRUCTURE AND DEVELOPMENT 4.1 SOME GEOMETRICAL ASPECTS OF COCCOLITH MORPHOLOGY 42 4.1.1 Rim width, and ellipticity variation. 43 4.1.2 Element orientation. 43 4.1.3 Element spacing. 45 4.1.4 Discussion. 45 4.2 COCCOLITH DEVELOPMENT DURING COCCOLITHOGBNESIS 46 4.3 A BASIC MODEL FOR HETEROCOCCOLITH DEVELOPMENT 48 4.4 RETICULOFENESTRID STRUCTURE 49 4.4.1 Basic structure. 49

2 4.4.2 Effect of proto-coccolithsize and ellipticlty. 51 4.4.3 Nucleation related features. 55 4.4.4 Element growth variation. 56 4.5 EXTENSION OF THE MODEL TO OTHER GROUPS 57 4.5.1 Coccolithaceae. 58 4.5.2 Helicosphaeraceae. 58 4.5.3 Pontosphaeraceae. 59 4.5.4 Holococcoliths. 59 4.5.5 Other groups. 59 4.6 SUMMARY 60 5. SPECIES AND INTRASPECIFIC.VARIATION 5.1 INTRODUCTION 61 5.2 SPECIAL TAXONOMIC PROBLEMS OF COCCOLITHOPHORES 62 5.2.1 Polyphase life cycles. 62 5.2.2 Polymorphism. 63 5.2.3 Polyspecific coccospheres. 63 5.2.4 Discussion. 64 5.3 OTHER TYPES OF INTRASPECIFIC VARIATION 65 5.3.1 Structural variation related to element development. 65 5.3.2 Variation in degree of calcification. 66 5.3.3 Aberrant coccoliths. 66 5.3.4 Size variation. 68 5.4 DISCUSSION - INTRASPECIFIC VARIATION AND NANNOFOSSIL TAXONOMY 68 5.4.1 Conventional species concepts. 66 5.4.2 Relationship of species concepts to taxonomic models. 70 5.5 CONCLUSIONS 71

PART B. - BIQ5TRATI0RAEHX 6. NEOGENE NANNOFOSSIL BIQSTRATIGRAPHY 6.1 NANNOFOSSIL ZONATION SCHEMES 74 6.2 CHRONOMETRIC AGE OF NANNOFOSSIL EVENTS 77 6.3 CORRELATION OF BIOSTRATIGRAPHY AND CHRONOSTRATIGRAPHY 78 6.4 FURTHER DEVELOPMENT OF BIOSTRATIGRAPHY 80 7. RETICULQFENESTRID SIZE VARIATION 7.1 METHODOLOGY 82 7.2 DATA PRESENTATION 83 7.3 RESULTS 84 7.3.1 Main results; DSDP Sites 219,223, 232, and 242. 84 7.3.2 Red Sea Sites, 225 and 227. 88 7.3.3 South West Indian Ocean Sites, 249 and 251A. 88 7.4 COMPARABLE STUDIES 88 7.5 DISCUSSION OF THE SIZE VARIATION TRENDS 90 7.5.1 Validity and extent. 90

3 7.5.2 Possible causes. 90 7.5.3 Taxonomic Interpretation. 91 7.5.4 Biostratigraphic use. 94 7.6 PRINCIPAL CONCLUSIONS 94 9 - BIOSTRATIGRAPHY OF INDIAN OCEAN DEEP SEA DRILLING PROJECT MATERIAL 8.1 SITE 219 95 8.2 SITE 223 96 8.3 SITE 231 98 8.4 SITE 242 100 8.5 RED SEA SITES, 225 & 227. 102 8.5.1 Biostratigraphy 103 8.5.2 Nannofloras 104 8.5.3 Discussion 106 8.6 OTHER SAMPLES. 107 9 - APPLIED BIQSTRATIGRAPHY - THE MAKRAN OF PAKISTAN 9.1 REGIONAL GEOLOGY 109 9.1.1 Tectonic setting. 109 9.1.2 Tectonic divisions and History of Research. Ill 9.2 LITHOSTRATIGRAPHY OF THE COASTAL MAKRAN. 113 9.2.1 Panjgur Facies (basin-plain turbidites) 113 9.2.2 Parkini Facies (slope siltstones). 116 9.2.3 Talar Facies (cyclic shelf sediments). 117 9.2.4 Deformation in the Coastal Makran. 117 9.3 NANNOFOSSIL BIOSTRATIGRAPHY 120 9.3.1 Material and methods. 120 9.3.2 Zonation. 120 9.3.3 Reworking. 122 9.3.4 Results. 123 9.4 DISCUSSION 130 9.4.1 Further micropalaeontological work. 130 9.4.2 Reworking. 130 9.4.3 Stratigraphical synthesis. 131 9.4.4 Continuity of processes. 132 Nannofossil distribution tables. 134

PART C - SYSTEMATIC DESCRIPTION OF NEOGENE NANNOFQSSILS. 10. INTRODUCTION TO THE SYSTEMATIC SECTION 10.1 Taxonomic conventions adopted. 143 10.2 Systematic treatment of intraspecific variation. 143 10.2.1 Rationale 143 10.2.3 Formal usage 144 10.2.3 Informal usage 145 10.3 Taxa not included in the systematicchapters. 145

4 11. COCCOLITHACBAE 11.1 Structure and classification. 149 11.2 Biostratigraphy and size variation 150 11.2.1 Coccolithus pelagicus group 150 11.2.2 Calcidiscus leptoporus group 152 11.2.3 Other species 153 11.3 Systematics 153 12. PRINSIACEAE 170 13. HELICOSPHABRACEAE 13.1 Structure of Helicoliths 178 13.2 Taxonomic groups 180 13.3 Biostratigraphical use. 181 13.4 Systematics 182 13.4.1 Helicosphaera carter! group. 182 13.4.2 Helicosphaera obliqua group. 182 14. PONTOSPHAERACEAE 14.1 Structure 189 14.2 Taxonomic subdivision of Scyphosphaera. 191 14.3 Stratigraphical distribution of Scyphosphaera. 192 14.4 Systematics. 193 15. SPHBNOLITHACEAE 15.1 Structure 201 15.2 Biostratigraphical use. 202 15.3 Systematics. 203 16. DISCOASTERACEAE 16.1 Structure. 208 16.2 Subdivision, stratigraphical distribution and evolution. 211 16.3 Systematics. 212 16.3.1 Early Miocene, D.deflandrei group 212 16.3.2 Middle Miocene, D.exilisgroup 213 16.3.3 Late Miocene, D.variabilis group. 216 16.3.4 Other Neogene discoasters. 219 16.3.5 Discoaster aster and other preservational "species" 226 16.3.6 Catinastex. 227 17. CERATOLITHACEAE AND TRIQUBTQRHABDULACEAE 17.1 Structure. 233 17.1.1 Triquetorhabdulaceae 233 17.1.2 Ceratolithaceae 234 17.2 Possible phylogenetic relationships,and biostratigraphical use 235 17.3 Systematics. 236 17.3.1 Family Ceratolithaceae. 236 17.3.2 Family Triquetorhabdulaceae. 241 17.3.3 Possibly related ortholiths. 243

5 18. SUMMARY AND RECOMMENDATIONS 18.1 Summary of principal conclusions, and results. 248 18.2 Suggestions for future research. 249

BIBLIOGRAPHY 250 Initial Reports of the Deep Sea Drilling Program 263 APPENDICES APPENDIX 1 - Nannofossil distribution chartsfor DSDP material. 264 APPENDIX 2 - Computer programs. 270 PLATES. 276 IMIJCES. 286

LIST OF FIGURES PART. A<—PALAEQBIOLQQIt 1 - Higher classification of coccolithophores. 15 2 - Aspects of haptophyte biology. 18 3 - types. 20 4 - Coccoliths and organic scales. 21 5 - Haptophyte cytology. 30 6 - Protection and buffering adaptations. 33 7 - Complex haptophyte tests. 37 8 - Functional model for cqccosphere morphology. 41 9 - Coccolith rim geometry. 44 10 - Coccolith development. 47 11 - Reticulofenestrid coccolith element form. 47 12 - Reticulofenestrid coccoliths. 50 13 - Pseudoemiliania lacunosa variation. 53 14 - Ray width variation. 54 15 - Examples of intraspecific variation in livingnannoplankton. 67 16 - Species concepts and evolutionary patterns. 72

PART B. BIOSTRATIGRAPHY 17 - Neogene and Quaternary nannofossil zonations. 75 18 - Nannofossil events. 76 19 - Estimates of the ages of nannofossil datums. 79 20 - Reticulofenestrid size variation, Sites 219 &223. 85 21 - Reticulofenestrid size variation, Sites 231 &242. 86 22 - Reticulofenestrid size variation, Sites 225, 227, 249, & 251A. 87 23 - Taxonomic interpretations of reticulofenestridsize variation. 92 24 - Biostratigraphy of DSDP Site 219. 95 25 - Biostratigraphy of DSDP Site 223. 97 26 - Biostratigraphy of DSDP Site 231. 99

6 27 - Biostratigraphy of DSDP Site 242. 101 28 - Western Indian Ocean, map. 110 29 - Tectonic units of the Hakran Region, map. 114 30 - Schematic section across the Makran AccretionarySystem. 115 31 - Talar facies. 118 32 - Coastal Makran, sample location map. 119 33 - Zonation Scheme for Makran Material. 122 34 - Branguli area map. 125 35 - Sections through the Branguli area. 126 36 - S. Pasni Traverse and Talo Koh Sections. 127 37 - Stratigraphical summary. 134 38 - Sediment sources during the Late Miocene - EarlyPliocene. 134

PART C. SYSTEMATIC TAXONOMY. 39 - Coccolithaceae morphology 165 40 - Stratigraphical ranges of Coccolithaceae. 166 41 - Coccolithus pelagicus and Calcidiscus leptoporussize variation. 167 42 - Distribution of Coccolithus pelagicus, Clausicoccus pzinalis, and Cycloperfoil thus carlae. 168 43 - Coccolithus pelagicus and Umbilicosphaeza sibogae varieties. 169 44 - A. Reticulofenestra holotypes; B. Subdivision ofP.lacunosa group. 177 45 - Stratigraphical distribution of Helicosphaeraceae. 188 46 - Pontosphaeraceae and Helicosphaeraceae, morphology. 198 47 - Scyphosphaeza holotypes. 199 48 - Stratigraphical distribution of Scyphosphaera 200 49 - Sphenolith structure and morphology. 206 50 - Stratigraphical distribution of Sphenolithaceae. 207 51 - Morphological elements of discoasters. 229 52 - Crystallographic orientation of discoaster rays. 229 53 - Morphology of various Discoaster species. 230 54 - Stratigraphical distribution of Discoasteraceae. 231 55 - Cat inaster origins, development, and morphology. 232 56 - Triguetorhabdulaceae, morphology. 245 57 - Ceratolithaceae, morphology. 246 58 - Stratigraphical distribution of Triquetorhabdulaceae and Ceratolithaceae. 247

7 CHAPTER 1 - INTRODUCTION

Doctoral theses are unwieldy beasts whose primary purpose, presumably, is to document the students ability to conduct original research during a finite period. As a result they are liable to contain an heterogeneous assemblage of material. This is certainly the case here, with discussion of problems, from the biostratigraphy of accretionary prisms, to the functional adaptations of unicellular . Nonetheless I believe that the whole is more than the sum of the parts and that certain themes are apparent. In particular the topic of coccolith morphology is approached from a number of directions, including, analysis of functional morphology, investigation of developmental processes and their relationship to coccolith structure, computer modelling, study of intraspecific variation, taxonomic description, and biometric measurement. The result is hopefully a more coherent understanding of Neogene coccoliths, which are of considerable utility in solving geological problems.

1.1 BACKGROUND TO AND DEVELOPMENT OF THE PROJECT The research for this thesis was carried out at Imperial College during the tenure of a three year research assistantship, to Dr.J.K.Leggett, working on the biostratigraphy of the Makran of Pakistan. This area is a Neogene accretionary system, and although geologically fascinating the combined problems it presented of complex structure and poor nannofossil assemblages made biostratigraphy rather problematic, and unrewarding. In order to solve the problems arising from this work it soon became apparent that comparative material with better preserved nannofossils would be needed. For this purpose a suite of samples was obtained from Deep Sea Drilling Project cores recovered from sites in the Indian Ocean, and the Red Sea. Study of these occupied a relatively small amount of the total research time but, owing to the higher quality of the material, yielded a disproportionate amount of data. My initial intention had been to use this material for a palaeoecological study of nannofossil assemblage variation. In the event the sample set and time available proved inadequate for the objective, and my interests developed in other directions. Nonetheless study of this material formed the basis for the rest of my work. In particular I used it for an examination of size variation in the most abundant nannofossil group, the reticulofenestrids. This size variation is both biostratigraphically valuable and of considerable intrinsic interest. In conjunction with this practical work I devoted a considerable

8 amount of time to literature based research on both living and fossil nannoplankton. This was essential since after tventy years of intensive studies nannofossil research was approaching maturity; but when I commenced the project, there were no useful syntheses. In particular, little use had been made of information from studies of living haptophytes in nannofossil research and, perhaps as a consequence, large areas of taxonomy were in need of critical review. I became interested in this aspect initially as a result of undertaking a review of information on functional morphology, for the International Nannoplankton Association Meeting, in Vienna 1985 (YOUNG 1987). I found it a productive approach, and have made extensive use of information from biological studies. In this respect I was fortunate to be studying Neogene nannofossils, which have close living relatives.

1.2 STRUCTURE AND CONTENT OF THE THESIS. The structure of the thesis reflects the various research topics, and it is divided into three parts. In the first part palaeobiological themes arising predominantly from the literature based research are discussed. Chapters 2 and 3, deal with general biology and functional morphology, and are essentially literature reviews, although with a critical slant. The main theme in these is coccoliths as cell components. In the next two chapters, information on coccolith development, structure, and variation, from a range of sources is combined to produce the most important conclusions of the thesis. In Chapter 4 a model for heterococcolith morphology is developed from examination of coccolith geometry and.development. This is used as the basis of a computer program for drawing coccoliths, and is applied to structural analysis of reticulofenestrid coccolith morphology. In particular the relationship between growth processes and intraspecific variation in Pseudoemiliania lacunosa is discussed. In Chapter 5 intraspecific variation is approached more directly, from a consideration of patterns shown in living nannoplankton. It is suggested that biological species concepts should be applied to nannofossils; but that in order to do so many currently described species should be redesignated intraspecific varieties, as is done throughout the taxonomic section. In the second part the results of the practical biostratigraphy research are presented. Neogene nannofossil biostratigraphy is well established so the general discussion, Chapter 6, is brief, but hopefully pertinent. The most important new results come from the work on reticulofenestrid size variation, these are discussed in detail in Chapter 7. In particular an early Late Miocene "small Reticulofenestra interval", is identified. The results from the work on the material

9 from the Makran and from the DSDP cores are documented in detail in Chapters 8 and 9, these include some results of interest for regional geology, as well as nannofossil work. The third part, systematic taxonomy, represents the combination of ideas developed in the first part and results from the second part, together with synthesis of the extensive published information. In particular species level taxonomy is revised in terms of intraspecific variation. This allows a re-appraisal of the evolution of the families in terms of lineage development. Also the morphology of the nannofossils in each group is re-examined to see how far how far the concept of homologous development is applicable to their ultrastructure. Most important Neogene families are discussed, with emphasis on the Middle Miocene to Pliocene taxa (the period represented by material from the Makran).

10 1.3 ACKNOWLEDGEMENTS

I am genuinely grateful to my supervisor Jerry Leggett, for providing consistent support and encouragement in areas outside his expertise, for arranging trips to Pakistan, Japan, and the United States, and for allowing me the freedom to develop the project in somewhat obscure directions. For the Makran Project work I am also grateful to John Platt, and Shaji Alam for their interest, encouragement and free exchange of ideas. I am particularly pleased to acknowledge the friendship and advice of many fellow nannoplankton workers, especially those of the University College Micropalaeontology Unit - Alan Lord, Alaa Baky, Paul Bown, Kevin Cooper, Hilary Docker ill, and Liam Gallagher. I met other workers less frequently, but enjoyed and benefitted considerably from, the contacts, notably at the International Nannoplankton Association Meetings in Vienna and London, particularly significant influences were Jan Backman, Dave Bukry, Shiro Nishida, Katarina Perch-Nielsen, and Amos Winter. This research was funded by the Natural Environment Research Council, with some additional support from the Nuffield Foundation. The Deep Sea Drilling Project Panel made available the Indian Ocean material which formed the basis for a major part of the research. Goldsmith's College Geology department generously allowed use of their electron microscopes. Throughout the course of the project I have also been ably and materially supported by my family, who have indulged my gnomic habits, and kept me dry. Within the Imperial College Geology Department, and particularly the Stratigraphy - Palaeontology Section many people willingly gave help and advice whenever asked, particularly, Gwyn Thomas, Paul Grant, Dave Carter, and Dick Giddens. I am most grateful for the efforts of Nick Morton to produce plates from my mediocre efforts, and to the other members of the photographic unit, Grace Lau and Adolfo Cash, for help on other occasions. Many friends among the academic staff, secretaries, research students, MSc students, and under-graduates contributed to making life enjoyable and keeping me a geologist during four years at Imperial. It would be tricky to mention them all, but positively dishonest not to mention Sue Agar, Caroline Ellis, Mark Enfield, Jenny Garnham, Pat Gibson, Ralph Gillchrist, Paul Gillespie, Ian Hayden, Andy Hyett, Chris Izatt, John MacKenzie, Keith Myers, Martin Robinson, Kelvin Rowe, Andy Thickpenny, Yi Wang, and Joyce Watt. Also, of course, Tarquin, who would have been pleased that the project worked out well in the end.

11 F> A R T A CHAPTER 2 - INTRODUCTION TO NANNOPLANKTON BIOLOGY

2.1. INTRODUCTION There is an extensive body of literature on the living equivalents of calcareous nannofossils. For the roost part, however, it has been produced by specialist phycologists, biochemists, and marine biologists and rarely deals with topics of direct interest to palaeontologists. Conventional literature reviews tend to reflect these specialist interests and so have similar limitations. I have attempted instead to collect the scattered observations of relevance, from throughout the literature in order to approach three topics which are of major importance for palaeontology: functional morphology; intraspecific variation; and the structural development of coccoliths. In each case I hoped to be able to build up models from our knowledge of living, and fossil, nannoplankton that would be applicable to observations on nannofossils. In this I had varying success - the examination of possible functional significance of coccoliths produced a plausible synthesis, but not one which makes geologically useful predictions. On the other hand consideration of intraspecific variation and structural development has produced a number of conclusions which should help focus our understanding of nannofossil taxonomy and evolution, with significant implications for biostratigraphy. This chapter has two purposes, first to give background information on subjects of relevance to the subsequent chapters. Second to provide a guide for use with the literature sources that I have used, to help other workers to follow up or assess any points of interest in the subsequent chapters. Toward these ends I have given outlines of the taxonomic position of coccolithophores, of coccolith - cell relationships, of haptophyte life cycles, and of the relevant literature. There are many good descriptions of the cellular organisation and biology of haptophytes (eg HAQ 1978, SIEBURT 1979, HIBBERD 1980), and I have not attempted to give another here. The glossary at the end of this chapter, however, outlines the major features and the accompanying diagram illustrates a typical cell. Taxonomic references to living taxa mentioned are given in Chapter 10.

13 2.2 HIGHER CLASSIFICATION, THE NATURE OF COCCOLITHOPHORBS Coccoliths are calcareous plates secreted internally by plant of the division Haptophyta. This class is primarily distinguished from other classes by bearing in addition to two flagella a third -*like organ, the haptonema. In addition to coccolithophores, which bear coccoliths the group includes non-calcifying genera, such as Chrysochromulina. These are less abundant than coccolithophores, particularly in the open ocean environment, but studies of them have been very important in developing understanding of haptophyte biology. All coccolithophores belong to the marine nannoplankton ( <60microns) and together with a few other groups form the calcareous nannoplankton. The systematic position of haptophytes within the general classification of organisms is illustrated in the Figure 1. As it shows they constitute one of many divisions (ICBN equivalent of phylum) of unicellular algae within the kingdom Protista. The Protista includes all unicellular organisms. It can be divided into prokaryotic and eukaryotic groups, the former lacking organised cells. The algae all have as their dominant mode of nutrition. Photosynthetic pigment type has classically been the primary means of classifying algae, giving rise to terms such as, red-algae, and blue-. Increasingly additional biochemical, ultrastructural, and morphological criteria have been employed in classification, such as: food storage products; number and types of flagella; cell wall type; and the nature of organelles such as the golgi body. This has caused the proliferation of divisions seen on the chart. The golden- are an excellent example of this trend. All algae with such pigments are sometimes included in one division, the Chrysophyta, including coccolithophores and . Of these the diatoms have always been regarded as a distinctive group, class Bacillariophyceae. The other golden-browns were, however, for a long while regarded as a single class, the Chrysophyceae. PARKE (1961) showed that this group could be subdivided into two series on the basis of their appendages and scale structure. This lead to the establishment of the class Haptophyceae by CHRISTENSEN (1962), for coccolithophorids and related non-calcifying genera. Subsequent research (see eg HIBBERD 1976) has strongly supported this separation so that most authors now consider the haptophytes a separate division. Unfortunately there is a further, nomenclatural, complication - Haptophyceae is a descriptive name and so invalid, under the ICBN, as a class level name. Hence HIBBERD (1976) proposed the alternative typified names Prymnesiophyceae and Prymnesiophyta, from the genus Prymnesium. The former is widely used, the latter not so frequently, PROTISTA

PROKARYOTIC EUKARYOTIC

BACTERIA CYANOPHYTA FUNGI UNICELLULAR AL6AE (Blue-green algae) inc-Radiolaria Stromatolites, and - phytoplankton -

PYRROPHYTA CRYPTOPHYTA HAPTOPHYTA CHRYSOPHYTA BACILLARIOPHYTA XANTHOPHYTA CHLOROPHYTA EU8LEN0PHYTA =yellow-green algae

DINOPHYCEAE PRYNNESIOPHYCEAE CHRYSOPHYCEAE BACILLARIOPHYCEAE CHLOROPHYCEAE PRASINOPHYCEAE =Di no f1agel1ates, =golden-brown =Diatoms =green algae, inc. Thoracosphaera algae, inc. many multicellular si 1i cof1agel1ates

N.B. OF THE MAINLY MULTICELLULAR, ALGAE, ONLY THE CHLOROPHYTA ARE INCLUDED ABOVE, THE OTHERS ARE; PHAEOPHYTA: Brown algae. COCCOLITHOPHORES NON-CALCIFYING GENERA RHODOPHYTA: . e.g. Chrvsochromulina CHAROPHYTA: Stoneworts. Prymnesiui

CALCAREOUS NANNOPLANKTON

HOLOCOCCOLITHS HETEROCOCCOLITHS

CALYPTROSPHAERACEAE SYRACOSPHAERACEAE PONTOSPHAERACEAE PRINSIACEAE CERATOLITHACEAE BRAARUDOSPHAERACEAE

HYMENOMONADACEAE RHABDOSPHAERACEAE HELICOSPHAERACEAE COCCOLITHACEAE CALCIOSOLENACEAE

FIGURE 1 - HIGHER CLASSIFICATION OF COCCOLITHOPHORES. Chart showing the taxonomic position of coccolithophores, and the living families of Coccolithophores. N.B. The primary purpose of this diagram is to act as a key to the status of groups mentioned, it is not a rigorous classification scheme.

15 for the sake of clarity I have avoided both, and mainly used the informal term haptophyte. Typical haptophytes and chrysophytes are planktonic and , with two (golden-brown) chloroplasts, a prominent nucleus and well developed golgi-body. They differ in the following ways: A. The motile phases of haptophytes have smooth flagella, as opposed to the flimmer (hairy) flagella of chrysophytes. Also they have a unique third flagella-like appendage, the haptonema. This has a different fibrillar structure to that of flagella, is frequently coiled, and seems to be used for attachment rather than locomotion - it is best known from the non-calcifying genera. B. Haptophytes nearly always possess unmineralised organic scales, and often calcareous coccoliths. The scales of chrysophytes, when present are usually silicified, to varying degrees, and have quite distinct structures from haptophyte scales. [N.B. Prasinophyceae also have organic scales produced within the golgi body, and so are sometimes compared with chrysophytes and haptophytes1. C. Haptophytes have uniquely developed golgi bodies (in which the scales and coccoliths are formed), distinctive , slightly different pigments, and various ultrastructural and biochemical differences. D. Haptophytes are predominantly marine, whereas most chrysophytes are freshwater organisms.

Classification of the haptophytes is not well established. The division between calcifying and non-calcifying genera is probably best regarded as an informal one, since the ability to calcify is likely to have been repeatedly suppressed and re-exploited during the evolution of the group. Evidence for this includes: the similarity of some coccolithophores to particular non-calcifying genera; the ability of many coccolithophores to produce phases which do not bear coccoliths; the discovery of forms with weakly calcified scales (MANTON et al 1977); and cases of cryptic first occurrences of groups within the palaeontological record. The terms calcareous nannoplankton, calcareous nannofossils, and coccolithophores are favoured respectively by marine biologists, palaeontologists, and phycologists. They have much the same meaning but the first two are size and mineralogy defined groups and so can include non-haptophytes, such as thoracospheres, and groups of uncertain affinities, such as discoasters. Within the calcareous nannoplankton, and nannofo'ssils, a number of families are quite well established, based on their coccoliths, but groupings of these families are largely arbitrary. Much further work is needed on the phylogenetic histories of

16 these groups and on the cytology and life cycles of their living representatives, before a definitive classification can be produced.

2.3 HAPTOPHYTB LIFE CYCLES, AND REPRODUCTION The principal means of reproduction of haptophytes is simple fission, in which asexual division of the cell produces two, or more, similar daughter cells. In some cases the cell covering is retained during this process, and divided between the daughter cells (Fig.2/B). In other cases division is preceded by loss of the cell covering. Both styles of reproduction can occur in the same species (KLAVENESS 1972b). The life cycles of most haptophytes also involve phase changes with two or more discrete phases characterised by their cell covering, motility, and ecology. Life cycle studies require the culturing of pure strains, and so only a limited number of species have been studied. These species have shown remarkable diversity, nonetheless three basic phase types can be recognised: A. Motile: This is the characteristic haptophyte cell type, with two flagella, allowing active swimming, and usually an haptonema. All the species for which a life cycle is known include a motile phase. They have variable cell coverings, including examples with organic scales only (e.g. Chrysochromulina, Emiliania), with holococcoliths (e.g. Calyptzosphaera, Coccolithus), and with heterococcoliths (e.g. Calciosolenia, Helicosphaera, Pleurochrysis, and Syracosphaera). B. Non-motile planktonic: These phases lack flagella and haptonema, but still live within the water column, maintaining their position via passive bouyancy regulation. They characteristically bear heterococcoliths, although in the best investigated species ( and Coccolithus pelagicus) the ability to form coccoliths is often lost. C. Non-motile benthonic (apistonema phase): This phase is common in shallow marine and freshwater species. It probably is equivalent to the non-motile planktonic phase of open marine species, no species is yet known to produce both types of non-motile phase. The individual cells are sometimes organised into filamentous colonies. These were previously described as a separate genus, Apistonema, and this name is often applied informally to the phase. In other cases the benthonic phase is amoeboid - for instance in most Chrysochromulina species. Many non-calcifying species have been observed to possess a non-motile benthonic phase, as have the coccolithophorid genera Ochrosphaera and Pleurochrysis (including the much studied species P.carterae). Most benthonic phases bear organic scales. True coccoliths are not known from them, but at least one genus (Chrysotila) can produce less

17 FIGURE 2 - ASPECTS OF HAPTOPHYTE BIOLOGY. A. PHASE CHANGES - as documented in £our well studied genera. Additional phases have been suggested for Pleurochrysis. The Emiliania huxleyi phases are: S-cells, motile with scales; N-cells, non-motile naked; c-cells, non-motile with coccoliths. B. REPRODUCTION - binary fission as seen in B . h u x l e y i , based on a series of micrographs in KLAVENESS 1972a. C. COCCOLITHOGENESIS - intracellular development of coccoliths on base-plate in coccolith vesicle. Based on micrographs of Pleurochrysis carterae in OUTKA £ WILLIAMS (1971), coccolithosomes omitted. See Fig.5 for example of coccolith vesicles in cell. Scale bar 0.1 m icron s. organised calcareous sheathes (GREEN 1986).

The typical haptophyte life cycle thus involves alternation between a planktonic motile phase and a non-motile phase - which can be either planktonic or benthonic, (Fig.2/A). All the phases are self reproducing and physiologically active, there is no true equivalent of dinophyte resting cysts. Nor is there any consistent evidence of genetic significance to the phase changes. Additional complication is provided in most species, by one or both of the motility defined phases occurring with two different cell coverings. These variants are also often referred to as phases. Thus the motile phase of Pleurochrysis carterae can occur either with or without coccoliths (LEADBEATER 1970), as can the non-motile phase of Bmiliania huxleyi (KLAVENESS 1972a,b). In Coccolithus pelagicus two slightly different types of holococcoliths can be produced on the motile phase (ROWSON 1985). Similarly two different types of heterococcolith can occur on the non-motile phase of Umbilicosphaera sibogae (INOUYE & PIENAAR 1984). In each case these variants are true breeding, but it is not clear in any of them whether this variation is comparable to true phase changes, or whether it would be more meaningful to regard them as examples of ecophenotypic or genotypic variation.

2.4 CELL-WALL COMPONENTS AND ORGANISATION Figure 3 illustrates six different types of cell-wall, developed by planktonic phases of haptophytes, and Figure 4 illustrates typical components of them. These are discussed below.

2.4.1 Plasmalemma and columnar layer. The simplest type of cell-wall lacks any resistant components, producing a "naked” cell (Fig. 3/A). The living cell is covered only by its bounding membrane, the plasmalemma, and often an associated columnar layer. The plasmalemma is a universal feature of cells, and although direct information on it is scanty its basic functions are well known. It serves to bind the cell, to separate and protect it from the surrounding environment, and to regulate the flow of nutrients, waste products, gases, and food particles, into and out of the cell. The columnar layer seems to be a characteristic feature of haptophyte cells, little is known about it except its columnar fabric, but it presumably provides some reinforcement or protection of the plasmalemma. Few haptophytes have naked cells as their dominant phase, however many species which normally have more elaborate cell-walls can produce naked cells. For instance Chzysochromulina pringsheiraii can shed its IS3 Coccolith p igure 3 _ C e |, W a || Types ------Organic scale

Plasmalemma 20 FIGURE 4 - COCCOLITHS AMD ORGANIC SCALES. Diagrams illustrating the relation between organic scales and coccoliths. N.B. Fig. E is a reconstruction the others are tracings. Sources PARKE & MANTON 1962, MAMTON £ PARKE 1962, MAMTON £ LEEDALE 1969, LEADBEATER £ MORTON 1973, ROVSON e t a l 1986. S cale bars one m icron. A. Rimmed scale, Chrysochromulina pringsheimii, spine and struts usually present (compare with Fig. 7/C). B. Rimless large body scale from Coccolithus pelagicus. C. Coccolith of Pleurochrysis carterae with base-plate scale. N.B. The rim structure is more complex than is suggested in this view. D. Rhabdosphaera stylifera, partially formed coccolith on base-plate scale. E. Vigvamma arctica, the rim and struts have an organic framework which is present in uncalcified specimens. Distal concentric fibrils not shown. F. Small circular body scales of P.carterae showing typical irregular concentric fibrils. G. Chrysochromulina polylepis body scale, an example of a more elaborate scale structure. H.Coccolithus pelagicus holococcoliths, "Crystallolithus hyalinus" (left), and "Crystallolithus braarudii" (right). I. Coccolithus 21 pelagicus, heterococcolith developing in coccolith vesicle against base-plate scale. scale envelope (PARKE & MANTON 1962), and Bmiliania huxleyi has a naked phase (N-cells), (KLAVENBSS 1972a,b).

2.4.2 Organic scales. The majority of non-calcifying species do not have naked cells but possess an outer pellicle of resistant cellulosic scales (Fig 3/B,C). These organic scales are elaborate structures with a fibrillar fabric, examples are illustrated in Figure 4. Typically they show a radiating pattern on their proximal surface and an irregular concentric fabric on their distal surface. The simplest morphology is a disc a couple of microns across without marginal structures. Some species have tests composed solely of these scales (Fig. 3/B). More usually the cell-wall is made up of two different scale types (Fig. 3/C) - an inner layer of small scales, and an outer layer of larger, and more elaborate, scales. The larger scales commonly have raised rims, and may also carry central structures and spines. They can be strikingly similar to coccoliths. Additional specialised scales may occur at the flagellar pole, increasing the total number of scale types to three or four in some species, and producing pellicles similar in form to some of the more elaborate coccospheres (e.g. Chrysochromulina pzingsheimii, Fig.7/C).

2.4.3 Heterococcoliths. The typical cell-wall (Fig. 3/D) is similar,to that of the more elaborate organic scale bearing types, but with the larger scales replaced by heterococcoliths. Coccoliths are, of course, elaborate calcareous structures of variable form, but usually plate-like. In heterococcoliths the elements are of variable complex shapes, and are predominantly organised in regular cycles. Non-resistant organic matter has sometimes been detected forming a weak envelope around the coccoliths (eg BRAARUD ET AL 1952, KLAVENESS 1976). More importantly resistant organic base-plates have been observed beneath coccoliths in numerous species. These base-plates commonly have the radiating-concentric structure of organic scales, and seem to differ from scales only in occurring below coccoliths (Fig.4/C-D). This cell-wall type, with body scales, coccoliths and base-plates seems to be very widespread, occurring in the majority of coccolithophores examined in detail (the scales are not usually visible without TEM examination). These include Coccolithus pelagicus, and Pleurochrysis cazteiae (MANTON & LEEDALE 1969), Umbilicosphaera hulbertiana, Scyphosphaeza apsteinii (GAARDER 1970), Helicosphaeza caztezi (NORRIS 1971), and Syracosphaera pulchra (LEADBEATER & MORTON 1973). These examples span a wide range of coccolith families and it seems reasonable to regard this as the typical structure. A second,

22 apparently smaller, group of coccolithophorids, lack body scales; but still possess base-plates. Examples include Umbilicosphaeza sibogae (Pig.3/E, INOUYE & PIENAAR 1984), Calciosolenia, Anoplosolenia, and Michaelsazsia (MANTON ET AL 1984, 1985). WESTBROEK et al (1984) have shown that the, coccolith bearing, C-cells of Emillania huxleyi also have this type of cell-wall. The base-plates are, however, thin and were not recognised in earlier studies (e.g. KLAVENESS 1976). INOUYE & PIENAAR (1984) reported Gephyzocapsa oceanica as lacking obvious scales, it probably like E.huxleyi has nebulous ones.

2.4.4 Holococcoliths. Holococcoliths are distinguished from heterococcoliths by being formed entirely of minute rhombohedral crystallites variably arranged to form complex shaped coccoliths (Fig.3/F, 4/H). In contrast the elements of heterococcoliths are larger, of variable shape, and predominantly organised in regular cycles. Unfortunately the details of wall structure are only known for two holococcolith bearing phases, ”Czystallolithus hyalinus", the motile phase of C.pelagicus (PARKE & ADAMS 1960, MANTON & LEEDALE 1963, ROWSON et al 1986) and Calyptzosphaeza sphaeioidea (KLAVENESS 1973). In both cases the wall structure is similar to that of typical heterococcoliths; with body-scales and base-plates, but with an extra element, a "skin" or "envelope" which encloses the coccoliths. In Crystallolithus hyalinus the scale, holococcolith and skin layers can be repeated (MANTON & LEEDALE 1963).

2.4.5 Partially calcified plates. Partially calcified plates, intermediate between coccoliths and organic scales constitute a final type of cell wall component. MANTON e t al(1976, 1977) have described a number of Arctic genera of this type, including Calciarcus, Turrisphaera, and Vigvamma. An example, Wigvamma arctica is shown in Figure 4/E. The outer plates of this species have an organic base, of the normal type, with a raised rim from which four struts rise, forming a teepee-like structure. The rim and struts have an organic framework within which calcification, in the form of complex elements, occurs to varying degrees in different specimens. This is an unusual case since in normal coccoliths although calcification occurs on a base-plate, the morphology is not closely related to that of the base-plate, and partially developed specimens do not have additional organic frameworks (e.g. Fig.4/D). They appear to be intermediates between coccoliths and organic scales.

23 2.5 COCCOLITHOGENBSIS Calcification in haptophytes has been, and is, the object of extensive research (eg LAVINB & ISENBERG 1962, KLAVENESS & PAASCHE 1979, GREEN 1986, WESTBROEK et al 1986, de VRIND - de JONG et al 1986). The most immediately relevant results of this are, that coccoliths are predominantly formed within the cell, and that scale formation is a closely associated process. The structural development of coccoliths during coccolithogenesis is discussed in more detail later (Chapter 4).

2.5.1 Intracellular coccolithogenesis. Both organic scales and coccoliths have been widely illustrated occurring internally, within vesicles, usually associated with the golgi body, one is indicated in Figure 5, and a developmental series is shown in Figure 2/C. By such methods progressive stages of development have also been traced. Vesicles form initially at the base of the golgi body and then migrate outwards as the plate develops, releasing the fully formed scale or coccolith to the surface. In the case of coccoliths the organic base-plate is usually formed first, followed by the coccolith rim, and ending with central structures. With large coccoliths only one is formed at a time, and it can occupy a considerable amount of the cell before being released. Even lopadoliths of Scyphosphaera have been observed forming internally (GAARDER 1970). In species where the developing coccolith occupies a large proportion of the cell the position of the coccolith vesicle varies considerably between species. However, the coccolith vesicle is connected by the endoplasmic reticulum to the golgi body, the nucleus, and other relevant organelles. The location of the vesicle is hence probably a matter of convenient cellular organisation, rather than a feature of major physiological importance (KLAVENESS 1972a). The biochemistry of coccolithogenesis and its relationship to other cell processes is a complex subject and still not fully worked out (de VRIND - de JONG et al 1986, WESTBROEK et al 1986). It is, however, clear that the mechanisms for synthesizing the various organic and inorganic components of coccoliths, and for controlling their form, are highly complex. They occur through the interaction of most, if not all components of the cell - certainly involving the chloroplasts, endoplasmic reticulum, nucleus and mitochondria in addition to the golgi body. So, even though coccoliths are an optional component of the cell-wall, the ability to produce them is a highly specialised character of haptophytes.

24 2.5.2 Extracellular calcification. Intracellular coccolithogenesis is the best established calcification process, but there is also evidence for extracellular calcification. Extracellular in this context means outside the plasmalemma, but within the cell wall. Some benthonic haptophyte phases are unquestionably capable of extracellular calcification (PARKE 1971, GREEN 1986). In particular the species Chiysotila lamellosa has been closely observed. These initially form a thick mucillaginous cell wall without resistant components. Calcareous elements grow within this sheath - as radiating acicular crystallites which coalesce to form a continuous calcisphere (10 to 20 microns across). There is also evidence that the holococcolith of the motile (crystallolithus) phase of Coccolithus pelagicus may be produced by extracellular calcification. During detailed study of this phase MANTON & LEEDALE (1963) frequently observed scales developing internally, but never coccoliths. Also they observed partially formed coccoliths in the cell-wall. These observations lead them to suggest that the coccoliths might be calcified within the cell-wall. Recently ROWSON et al (1986) provided strong experimental support for this suggestion. They found that after decalcification, by bubbling COz through the culture, organic framework relics of the coccoliths were left. They then observed, and illustrated, extracellular recalcification of these frameworks. KLAVENESS (1973) also failed to observe intracellular coccoliths, during his study of Calyptrosphaera sphaeroidea. So extracellular calcification may occur in this species, and it is conceivable that it is the general means of calcification of holococcoliths. Finally INOUYE & PIENAAR (1984) have suggested that in Umbilicospbaera sibogae var. foliosa growth continues after extrusion, since intracellular coccoliths are invariably smaller than those on the coccosphere. Their suggestion is supported by the morphology of the distal shield of this variety, the rays of which have unusual faceted inner parts and counter-clockwise imbricated overgrowths on the outer parts (illustrated in Figs.15 & 43). This morphology is, however, unique - at least within the Cenozoic - which can perhaps be taken as evidence that the process is also unique. In any case the critical crystal nucleation phase occurs prior to extrusion. The implications of this evidence for extracellular calcification are not clear, but a few points are worth making. First, intracellular and extracellular deposition could occur through the same biochemical pathways, although with a change in regulatory mechanisms.

25 Second, it seems to be generally accepted (e.g. GREEN 1986) that the very precise regulation required for regular heterococcolith formation is unlikely to occur outside the cell. Third, whatever the mechanism, an implication of extracellular calcification is that sophisticated interaction can occur between the cell and cell-wall.

2.6 LITERATURE SOURCES The following list outlines the main bodies of literature available. A. Phycology: As small open-marine organisms, coccolithophores have not been particularly popular for biological studies. Nonetheless during the last thirty years or so, a useful amount of work has been done on them, and more particularly on closely related, but non-calcifying, haptophytes. Such work has mostly been observational, with study of cultural isolates of easily bred species, by light microscope observations of living cells and TEM studies of ultra-thin sections. This work is responsible for most of our knowledge of haptophyte cytology, ecology and reproduction, and for the taxonomy of non-coccolith bearing haptophytes. The most important single research group has been based on Leeds University and the Plymouth Marine Biological Station, under Professors Parke and Manton. Reviews include PARKE (1961), BONEY (1970), DODGE (1973), and HIBBERD (1976, 1980). Original research papers such as PARKE & ADAMS (1960) or MANTON & OATES (1985) are, however, often more immediately comprehensible than these reviews. B. Biochemical research: In addition to the phycological research on haptophytes there has been biochemical research using the group as suitable material. In particular, Erailiania huxleyi and Pleurochrysis carterae have been intensively studied, since they are very easy to culture. This work is mostly experimental and has been directed toward elucidating the biochemical and genetic mechanisms and pathways of coccolithogenesis. This is a difficult subject for the non-specialist, reviews include PAASCHE (1968) and KLAVENESS & PAASCHE (1979), there are a number of valuable synthetic papers in LEADBEATER & RIDING (eds, 1986). C. Marine biology: Studies of living nannoplankton populations have a respectable pedigree, commencing with LOHMANN (1902). Findings of the early workers are synthesised by SCHILLER (1930). Modern research centres around SEM studies of ocean water samples, concentrated by filtration. Taxonomic research is still very important, along with studies of the distribution of nannoplankton with depth and latitude. Many workers, particularly of the Norwegian school (Braarud, Gaarder, Heimdal) have been closely associated with the phycological

26 work. There is also a useful degree of interaction with palaeontologists, and several workers have studied both fossil and live material (e.g. Okada, Nishida, Verbeek), studies of ocean bottom sediments provide a further link. Major beautifully illustrated papers include OKADA & McINTYRE (1977), NISHIDA (1979a), HEIMDAL & GAARDER (1980, 1981). There are as yet no comprehensive reviews, but various authors have produced taxonomic syntheses of particular groups (eg BURNS 1973 - Pontosphaera, NORRIS 1985 - holococcoliths). D. Palaeontology: The palaeontological literature is vast, mostly dealing with taxonomy and biostratigraphy. The most valuable studies in terms of palaeobiological information have been descriptive studies of exceptionally preserved material (e.g. PERCH-NIELSEN 1971, GOY 1981, COVINGTON 1985, LAMBERT 1986). PERCH-NIELSEN (1985a,b) provides an invaluable general review. E. Non-specialist literature: For each of the topics less specialist literature dealing with general concepts and several biological groups are frequently valuable, for instance HUTCHINSON (1967), MORRIS (ed, 1980), RAYMONT (1980), HOLM HANSEN (1984). TAPPAN (1980) gives a comprehensive review of most topics, and a very extensive bibliography.

27 2.7 GLOSSARY OP TERMS USED IN THE BIOLOGICAL DESCRIPTION OF HAPTOPHYTES.

N.B. Most features are illustrated in Figure 5. The notes are based on information from various papers cited elsewhere, and the Penguin Dictionary of Botany (ed. BLACKMOORE & TOOTHILL, 1984).

BASE-PLATE: Organic scales occurring on the proximal surface of coccoliths. BODY SCALE: Organic scales occurring independently of coccoliths. CHLOROPLAST: Organelle containing photosynthetic pigments and bounded by a double membrane. These are the photosynthetic organelles in most algae. In haptophytes there are usually two chloroplasts, on either side of the nucleus. They have a golden-brown colour due to the presence of significant quantities of p-carotene in addition to chlorophyll (a and c). They have an internal structure of lamellae, each composed of three thylakoids. COCCOLITH: Calcareous plates occurring around the outside of haptophytes. COCCOLITH VESICLE: Intracellular vesicle in which a coccolith develops. They usually originate in the golgi body and migrate through the cell during coccolithogenesis. COLUMNAR LAYER: The lowermost layer of the haptophyte cell wall, lying directly above the plasmalemma. It has a distinctive columnar texture but little else is known about it. CYTOPLASM: The fluid or gel of the protoplasm occurring around the organelles. ENDOPLASMIC RETICULUM: System of flattened membrane bounded channels and vesicles within the cytoplasm. It interconnects with the nuclear envelope, plasmalemma, chloroplast envelope, and golgi body, and separates and transports newly synthesised products. FLAGELLUM: Thread-like extension of the protoplasm beyond the main body of the cell, used primarily for locomotion. Motile haptophytes are distinguished by having two smooth subequal flagella and a haptonema (qv). Some non-motile haptophyte phases still possess complex flagellar base structures. GOLGI BODY (, dictyosome): This organelle consists of a stack of saucer shaped cisternae, and spherical vesicles which arise from the ends of the cisternae. The vesicles transfer material synthesised in the cisternae to the cell surface, in this way the golgi body participates in the formation of the plasmalemma and cell wall. In haptophytes the golgi body is strongly developed, and occupies the region between the flagella bases and the nucleus. The coccoliths, scales, and columnar layer are normally produced within

28 it. HAPTONEMA: Flagellum-like appendage occurring only in haptophytes. It arises between the flagella, but is normally of markedly different length, and is often coiled. It has a unique internal structure, whereas flagella have similar structures in all groups. The main observed function is anchorage, they do not seem to be prominent in oceanic coccolithophores. MITOCHONDRION: Mitochondria are organelles found in all eukaryotic cells, they produce the energy-transferring molecule ATP. The inner bounding membrane is complexly folded giving rise to internal tubule like structures, cristae - these make the mitochondria easy to identify on TEM sections. NUCLEAR ENVELOPE: Double membrane surrounding the nucleus, part of the endoplasmic reticulum (q.v.). NUCLEUS: The part of the cell containing the genetic material. ORGANELLES: Specialised membrane enclosed structures within the cell. Include the nucleus, mitochondria, chloroplasts and golgi body. PLASMALEMMA (plasma membrane): Bounding membrane of the cell, separating the cell wall and the protoplast. It is semi-permeable, allowing and probably regulating, the passage of nutrients, water, etc. PLASTID: General term for chloroplasts (qv) and other organelles with a similar structure but lacking photosynthetic pigments. PROTOPLAST: The living cell, including the cytoplasm, organelles and plasmalemma, but not the cell wall. : Proteinaceous bodies within chloroplasts. In haptophytes there is usually one per chloroplast, forming a bulge on the inner surface, lamellae pass through the pyrenoid, but are wider spaced than in the body of the chloroplast. SCALES: Haptophytes, with very few exceptions, have a layer of organic scales as a major part of the cell wall. These typically are simple plates, constructed of resistant cellulosic fibrils with a radiating pattern on the proximal side and concentric pattern on the distal side. THYLAKOIDS: Sublayers of the lamellae within chloroplasts. VACUOLE: Fluid filled cavity within the cytoplasm.

29 FIGURE 5 - HAPTOPHYTE CYTOLOGY. Sectional diagram of a motile-phase coccolith bearing cell of Pleurochrysis carterae. Based on published transmission electron micrographs, primarily Figure 1 of OUTKA & WILLIAMS (1971). Most labelled features are discussed briefly in the glossary. Body scales and coccoliths of P.carterae are illustrated in Figure 4/C,F, and intracellular coccolithogenesis in Figure 2/C. Scale bar one micron.

30 CHAPTER 3 - FUNCTIONAL MORPHOLOGY

Virtually nothing is known for certain about the functions of coccoliths, or scales, but a certain amount can be inferred from their morphology, and from what is known of haptophyte ecology. Scattered speculations based on this approach have been made by BRAARUD et al (1952), HAY (1968), GARTNER & BUKRY (1969), and HONJO (1976). I adopted it on a more systematic basis (YOUNG 1987) in order to see if palaeoecological information could be derived from the distribution of coccoliths, as has successfully been done with planktonic foriminifera (HART 1980). Subsequently MANTON (1986) has published a short paper on the functional significance of scales in haptophytes and other phytoplankton groups. This is highly relevant since, as she has argued in a succession of papers, coccoliths are in many respects best regarded as calcified scales rather than as distinct entities. This argument is outlined below, then the main possible functions are discussed. The emphasis and some conclusions are slightly different to those in my previous discussion (YOUNG 1987).

3.1 RELATIONSHIP OF COCCOLITHS AND ORGANIC SCALES Several lines of evidence from the preceding chapter suggest that organic scales and coccoliths are closely related structures including; 1. Larger scales and coccoliths occur in similar positions on the cells of non-calcifying genera and coccolithophores, [Fig.31. 2. The morphology of scales and coccoliths, and of the tests that they form are closely similar, [Figs.4,131. 3. Partially calcified plates are known from some genera, these appear to be intermediate between scales and coccoliths [Figs.4,131. 4. Coccoliths are formed on organic base-plates [Figs.3,41. 5. Coccoliths and scales are produced by similar processes, in the golgi body [Fig.2/C1. 6. Although the most common condition is for cells to bear both coccoliths and scales, either or both may be absent [Fig.3].

These observations in turn have suggested the following conclusions. A. Coccoliths and organic scales are homologous, interchangeable and optional components of the cell-wall. B. Whatever functions they perform cannot be essential to the cell [since they are optional components].

31 C. Coccoliths and scales, in part at least, perform the same functions [from 1-3 above]. (D. Calcification has no functional significance, but occurs as a byproduct of other cellular biochemistry).

Conclusions A-C are inevitable deductions from the evidence. D is an extreme development from this line of reasoning, which has been tentatively suggested in the literature (eg MANTON & LBEDALE 1969, MANTON et al 1977). It does not appear rational to me, particularly since many species of haptophytes, and even phases of coccolithophores do not calcify. Calcification may both be functionally adaptive, by enhancing the primary functions of plates, or permitting secondary ones. Hence the need to calcify may be controlling the cellular biochemistry, rather than vice versa.

3.2 PROTECTION Two separate protective functions are possible for coccoliths. First, support for and protection of the plasmalemraa. This must be a delicate membrane since the flagella, ingested food particles, and of course scales, are able to pass through it. Hence it may be vulnerable to damage, making physical protection of it important. A variant of this might be protection against . As MANTON (1986) has argued this almost certainly is the type of function performed by scale coverings. By extension it is likely to be the basic function of coccoliths. Calcification may have occurred initially in order to enhance this function. However, in many species the coccoliths join together to form robust coccospheres. This is particularly obvious for placoliths such as Coccolithus pelagicus and Emiliania huxleyi, or for the helicoliths of Helicosphaera(Fig.6/A-B). For these a second protective function, inhibiting predation looks possible. Cropping is one of the main controls on phytoplankton (FROST 1980, SMAYDA 1980), and there are many different phytoplankton and groups. So Darwinian selection should rapidly occur if even a mild degree of protection is afforded by coccoliths. Coccospheres do not appear very effective at stopping predation, since laboratory experiments have shown that coccolithophores form good foodstocks for a range of zooplankton (BONEY 1970, HONJO 1976), and coccospheres have repeatedly been observed in the guts and fecal pellets of salps, copepods and other zooplankton (NORRIS 1971, ROTH et al 1975). Furthermore most zooplankton seem to be rather unselective grazers ingesting all particles within a given size range. Calcification may,

32 FIGURE 6 - PROTECTION AND BUFFERING ADAPTATIONS. A-Bs Coccospheres, showing two different possible adaptations of coccolith form for providing a rigid . C-E: Coccoliths and sections through segments of coccospheres, showing three different possible adaptations of coccolith form for providing an extended cell wall, of low density and strength, but of possible significance in regulating interaction between the protoplast and environment. N.B. Figure B is a tracing (from NISHIDA 1979a pl.l/la), the others are reconstructions. Scale bars one micron, representative only. Main sources, GAARDER & HEIMDA1 1977, NISHIDA 1979a, REID 1980, NORRIS 1985. A. Helicosphaera cartezl. B. Emllianla huxleyi. C. Calyptrollthlna multipoza. D. Umbellosphaera sp. E. Syracosphaera sp. 33 however, still be adaptive by making grazing more difficult, much as silica deposition in grass is adaptive by making digestion difficult. It is my opinion that the elaborate adaptation of coccolith form toward producing a rigid test can best be interpreted in terms of such a protective function. Spines may have a separate function in this respect, raising the effective diameter of the test and so reducing predation by certain grazers.

3.3 FLOTATION The two most important controls on phytoplankton are light and nutrients. The distribution of both of these varies vertically, hence regulation of position within the water-column is vital for phytoplankton. The various topics involved here - the distribution and effects of light and nutrients, and the hydrodynamics of phytoplankton - have been widely researched and good syntheses are available, notably in MORRIS (Ed., 1980). The principal nutrients are nitrates and phosphates, whilst Si, Mo, Mn, Co, B, Zn, and Cu are also important. Phytoplankton need to obtain all these from the sea-water. The principal nutrients are generally present only in limited amounts and so exert a major influence on phytoplankton distribution (eg MCCARTHY 1980, NALEWAJKO & LEAN 1980). An obvious example of this effect is the increased productivity of upwelling zones. Generally surface waters are depleted in nutrients due to extraction by phytoplankton, hence there are advantages in occupying the deeper nutrient rich waters. In addition MUNK & RILEY (1952) have postulated, that when nutrients are in limited supply they will rapidly be depleted in the immediate vicinity of a phytoplankton cell, unless there is a continuous flow of water around it. For this flow to occur the cell has to move relative to the water; by actively swimming, or by passively rising or sinking. This model has been generally accepted among planktologists (eg HUTCHINSON 1967, TAYLOR 1980). A simple consideration of volumes indicates its applicabilty to coccolithophores. Peak bloom abundances are around 107 cells per litre (HOLLIGAN et al 1983), and an individual coccosphere - with a diameter of 10 to 20 microns - has a volume of the order of 10:'z litres. So even during blooms each cell occupies a volume of water 10^ times its own volume. Exploitation of the nutrients requires motion of cells through this volume, 10 metres of movement would be needed for the cell to pass through the entire volume. The effects of light on phytoplankton are usefully reviewed by YENTSCH (1980) and HOLM HANSEN (1984). Light decreases rapidly with

34 depth but effective photosynthesis appears to be possible down to depths where the light intensity is only 0.1% of that at the surface, the light compensation intensity. At these depths light is the prime regulator of phytoplankton activity, but this ceases to be the case at light levels above about 20% of surface light, the saturating light intensity. Above this level phytoplankton crops are largely unrelated to light levels, with nutrients being a more important control. Toward the surface ultraviolet radiation may be responsible for reducing primary production. Light intensity and nutrient supply thus combine to produce a vertically variable water column, which is reflected in phytoplankton distribution. Plainly mechanisms for regulating position in the water column are potentially valuable. Such mechanisms include swimming, by means of flagella, whilst turbulence and differential reproduction rates will also be important. Bouyancy has to be a further important factor. Coccoliths and coccospheres can affect cellular bouyancy in a number of ways as discussed below. A. Ballasting: On the basis of the nutrient deficiency model of MUNK & RILEY (1952) non-motile cells, in particular, will benefit from either sinking or rising. Rising is an unstable strategy, as it is liable to result in the entire population ending up as a scum on the surface, an environment of limited extent and nutrient supply. Slow sinking, combined with turbulent mixing can result in a more useful depth distribution. Many workers have suggested that this is a function, or at least an effect of (eg HUTCHINSON 1967, TAYLOR 1980), and it is also probably applicable to at least the more massive coccoliths, as suggested by HAY (1968) and HONJO (1976). The greatest ballasting effect should be provided by heavily calcified placoliths such as those of Coccolithus pelagicus, and Calcidiscus leptoporus. These are also usually borne by non-motile phases. B. Bouyancy control: An extension of the ballasting concept was suggested by HONJO (1976), on the basis that isolated coccoliths are anomalously abundant in surface waters, and some species have multi-layered tests. He suggested that the cells might be able to vary the number of coccoliths on them, by producing and shedding coccoliths, thus adjusting their density and rate of sinking. This mechanism can only apply to a few species, since most coccospheres are not multi-layered, the best candidates are certain placolith bearing species, particularly Emiliania huxleyi (Fig.6/B). Also although variations in coccolith production rate can easily be envisaged it is harder to see how coccolith shedding could be regulated. C. Surface roughness: Coccospheres vary greatly in surface

35 roughness. This might be interpreted as another means of varying rate of descent - by lowering frictional resistance to sinking; MANTON & OATES (1985) have suggested such a role for the elegant shaping of the scapholiths of Calciosolenia. However, owing to the high effective viscosity of water for small bodies, laminar flow should occur around coccoliths, in which case the effect of surface roughness should be negligible (HUTCHINSON 1967). So this is one putative function which can reasonably be discarded. D. Aspherical cells: Diatoms usually are discoidal or elongate in form, and this is widely interpreted as being a sinking related adaptation. Shear forces tend to rotate aspherical bodies into the orientation of maximum form resistance [a falling sheet of paper demonstrates this effect well]. This slows descent, and can also aid photosynthesis by presenting a large surface area, (HUTCHINSON 1967, TAYLOR 1980). Many coccolithophores have distinctly aspherical coccospheres. Fusiform and cylindrical shapes are most common, occurring for instance in Calciosolenia, Acanthoica, some Syzacosphaeia species ,and to a lesser degree in Helicosphaeza (Fig.7). Discoidal forms are much less common, Scyphosphaeza is one of the few reasonably convincing examples. Apart from Scyphosphaera all these coccolithophores are motile so the flagella may assist in cell orientation. E. Spines: Spines provide an alternative way of reducing sinking rates, without altering the protoplast shape, and can contribute to cell orientation. This seems very likely to be the function of some spines, such as those of Acanthoica and Calciosolenia, which enhance the shape of an already asymmetrical test. A more extreme case is provided by the parachute-like circum-flagella whorl coccoliths of Michaelsazsia, Halopappus and Calciopappus. These virtually certainly are an adaptation for slow sinking (Fig.7/A).

FIGURE 7 (overleaf) - COMPLEX HAPTOPHYTE TESTS. Drawings of motile phase tests of four unrelated haptophyte species, showing analogous adaptations of test form for flotation, and consequent periplast polymorphism. N.B. These are reconstructions rather than tracings, and so may contain errors of interpretation. Scale bars one micron. Main sources, PARKE & MANTON 1962, BORSETTI & CATI 1972, NISHIDA 1979a, HEIMDAL & GAARDER 1981, MANTON & OATES 1985, MANTON et al 1984. A. Halopappus adriaticus (Syracosphaeraceae). B. Calciosolenia sp. (Calciosolenaceae). C. Chzysochzomulina pzingsheimii (non-calcifying genus). D. Acanthoica guattrospina(Rhabdosphaeraceae).

36 Figure7 37 In a number of other species spines occur distributed over the cell surface, these are most characteristic of the rhabdosphaeraceae, but there are examples in numerous different families. These could function to retard sinking, by increasing effective cell diameter without greatly changing its mass; but other functions are equally possible, particularly enhanced protection. Summary: There are good reasons for bouyancy regulation to be important for phytoplankton and certain adaptations of coccolithophores can clearly be related to this end, in particular aspherical coccospheres, and grouped spines. It is difficult to tell whether the influence is more pervasive.

3.4 LIGHT REGULATION Since phytoplankton are dependant on photosynthesis, light regulation is an attractive possible function for coccoliths. Two separate light regulatory functions have been proposed. BRAARUD et al (1952), suggested that coccoliths could reflect ultraviolet light thus enabling them to live higher in the water column than other phytoplankton. Possible support for this comes from recent remote-sensing work which has identified oceanic patches with high reflectance of ultraviolet light caused by blooms of E.huxleyi (HOLLIGAN et al 1983). It is by no means clear, however, that this effect is strong enough to benefit individual cells. GARTNER & BUKRY (1969) suggested the reverse role, that coccoliths might act as light gatherers, thus enabling coccolithophores to live deeper in the water column than other phytoplankton. The main supports for this are the lens-like form of placoliths, and the fact that calcite has a higher refractive index than water and so will tend to refract light into the cell. There is no support for either of these suggestions from the distribution of coccolithophores, and the wide range of shapes and crystallographies of coccoliths hardly suggests optical design. It seems highly unlikely that these effects can be anything more than minor incidental advantages of calcification. More generally it may be noted that, as discussed above, there is an excess of light through much of the water column. Light regulatory devices have not been identified in other phytoplankton groups.

38 3.5 ENVIRONMENTAL BUFFERING The plasmalemma as well as bounding the cell carries out a number of regulatory functions. It controls the passage of nutrients, waste products, and dissolved gases into and out of the cell. It is possible that some of these functions in addition to protection, may in part be adopted by the cell-wall. This concept has not been developed much for either haptophytes or other phytoplankton, but MANTON (1986) suggested that water trapped in multiple-scale layers, or by basin or dome shaped scales may act as chemical buffer layers or even as filters. This is perhaps the least likely sounding possible function for coccoliths, but it provides a powerful explanation for the form of numerous otherwise anomalous coccoliths. It is most attractive as an explanation for forms which produce a two layered coccosphere with a space between. Three separate types of coccolith have this effect (Fig.6). (1) Numerous holococcoliths have a domal form (see NORRIS 1985 for a review), including Calyptzolithina, Calyptrolithophora, Calyptrosphaera, Daktylethza, and Sphaezocalyptza. (2) Various Syracosphaeraceae have two layered coccospheres consisting of an inner layer of normal coccoliths with a low basin form, and an outer layer of discoidal or domal coccoliths, e.g. Syzacosphaeza, Deutschlandia. (3) Several Rhabdosphaeraceae have broadly flaring spine ends which unite to form a continuous outer sphere. This is separated by the length of the spines from the inner sphere formed by the basal discs. The main examples are Discosphaera, and Umbellosphaeza, an analogous structure is shown by Papposphaeza lepida, an unrelated living species. The fact that these three groups have independently evolved two layered tests, by separate devices is rather strong evidence that this structure has adaptive significance. Also in each case the coccoliths appear to be too delicate to be of value for physical protection or for adjusting cell-density. Other structures could also be related to protoplast-hydrosphere interactions; basin shaped coccoliths such as those of Pontosphaeza may perform a crude type of water trapping function. Porosity may also be related, as various authors have speculated, including SCHILLER (1930) and HAY (1968). Almost without exception coccoliths are perforated and frequently these perforations are finely subdivided, as for instance in Reticulofenestra and Cycloperfolithus. This is another feature which has been evolved independently in different coccolith families.

39 3.6 DISCUSSION Coccoliths are essentially calcified scales, and so have their origins as structures for the re-inforcement and protection of the plasmalemma. However, calcification has become an integral part of coccolithophorid biology, thus it can be adapted to perform a number of different functions including: protection of the cell as a whole from predation; control of cell density and cell shape for flotation control; production of a buffer zone of trapped water; and possibly light regulation. These functions are probably developed to varying degrees in different families, genera and species, in response to different ecological strategies. So for instance, holococcoliths appear to be primarily adapted for water trapping, and helicoliths for protection. Most coccolith types, however, could perform a range of different functions, for instance distributed spines serve to increase the cell diameter but this could be an adaptation for either slower sinking or discouraging predation by size selective zooplankton. Similarly the placolith form may be adapted for any, or all, of ballasting, protection, or light concentration. Figure 8 gives what is in my opinion the simplest likely model, in terms of coccosphere morphology. Three basic types of function are included: Physical protection; environmental buffering; and flotation control. These are responsible respectively for: Rigid cell-walls; extended delicate and porous cell-walls; and modification of coccosphere shape. Each of these adaptations can be achieved by various coccolith shapes, thus they seem adequate as functional explanations for the range of form shown by coccoliths. Other aspects of coccolith morphology - such as the details of their structure - are probably better explained in terms of the influence of developmental processes.

40 FLOTATION ADAPTATIONS

FIGURE 8 - FUNCTIONAL MODEL FOR COCCOSPHERE MORPHOLOGY. A. Naked cell; e.g. Brailiania huxleyi N-cells. B. Organic scales only, elementary protection; e.g. Chrysochromullna and E.huxleyl S-ce11s. C. Single non-imbricate layer of coccoliths, raised rims give rigidity and precise location; e.g. Hymenomonadaceae and Pontosphaeraceae. D. Single imbricate layer of placoliths, placolith rims provide more rigid and thicker protective test; e.g. most Coccolithaceae and Prinsiaceae. E. Multiple imbricate layer of placoliths, extra coccoliths may further enhance protection, or allow bouyancy regulation; e.g. E.huxleyi C-cells. F. Dithecatism, delicate outer layer of specialised coccoliths produces extended cell wall, possibly for regulation of cell - water interaction; most Syracosphaeraceae. G. Domal coccoliths, alternative means of producing extended cell wall; most Calyptrosphaeraceae (holococcoliths). H. Elongate cell with specialised whorl coccoliths for orientation; Calciopappus, Halopappus, and Michaelsarsia, also analogous forms with apical spines (Fig.7). I. Coccosphere with extended equatorial coccoliths, probably for orientation and retarded sinking; Scyphosphaera. J. Distributed spines, probably for retarded sinking; e.g. Rhabdosphaeraceae, Noelaezhabdus.

41 CHAPTER 4 - COCCOLITH STRUCTURE AND DEVELOPMENT

INTRODUCTION Since the application of electron microscopy to nannoplankton research it has been possible to observe accurately the fine structure of coccoliths. This has become a standard descriptive technique with invaluable results for the development of taxonomy. I have used some of the accumulated data in a slightly different way here, investigating certain geometrical aspects of coccolith morphology, and combining this with information from studies of the development of living coccoliths. These approaches have been used to develop an enhanced understanding of the structure of coccoliths. A prime objective of this is to identify those aspects of the fine structure of coccoliths which are the result of architectural control, and so separate other aspects which are of taxonomic significance, or which may be of adaptive significance. A more direct impetus was provided by an attempt to write a computer program to generate mathematically illustrations of coccoliths. This is a valuable end in itself, since accurate illustration is an essential means of communication in palaeontology. Also writing the program provided both a need for a model of coccolith geometry, and a means of testing the predictions made by such a model. The program I have developed has enabled me to produce diagrams of a number of species, as used in this thesis (e.g.Fig.12). At least equally accurate illustrations have, of course, been produced by conventional techniques (eg GRUN et al 1974, THEODORIDIS 1984, VAN HECK & PRINS 1987). Computer generated diagrams have the advantage, however, that they can be readily redrawn, updated, modified, and experimented with. Also they can be produced by workers with modest artistic skills, such as myself. They are not, however, necessarily faster to produce, each diagram set still took several hours to produce. . There is considerable potential for further development of the programs, to produce perspective illustrations, and to model birefringence patterns. This chapter deals with development of a model for heterococcolith development and geometry, and its application to reticulofenestrid coccoliths. This model forms the theoretical basis for the computer program. I have not discussed the computing aspects here, outline details are given in Appendix 2.

4.1 SOME GEOMETRICAL ASPECTS OF COCCOLITH MORPHOLOGY It is difficult to determine the underlying controls on coccolith morphology from circular coccoliths, because the high degree of symmetry they possess leaves many possibilities open. Indeed it is possible to model the morphology of circular coccoliths, and discoasters, without

42 any reference to their mode of development. The same is not true for elliptical coccoliths, the elements of which inevitably vary around the rim. Investigation of their geometry is essential for modelling, and of interest in terms of structural control. The simplest way of mathematically deriving an elliptical form from a circular one is via simple linear deformation. This type of transformation is illustrated in Figure 9/B-C. Unfortunately the resultant form is not very realistic, it differs from real coccoliths in a number of important ways, as discussed below.

4.1.1 Rim width, and ellipticity variation, on single coccoliths. On the deformed coccolith the uniform stretching means that all circles become true ellipses of similar elongation (1.3 in Fig.9/C). As a consequence the rim width varies. It is greater at the ends of the coccolith (i.e. along the long axis) than at the sides, by a factor equivalent to the elongation. Examination of electron micrographs of real coccoliths shows that a quite different relationship is the normal case. Characteristically the rim width is constant, and the ellipticity varies, decreasing outwards. The rim width constancy in turn implies that rays are of similar length all round the coccolith. The rim width constancy and ellipticity variation can be seen qualitatively by observing electron micrographs, and is apparent even with light microscopy. It can also be demonstrated by measuring the length and width of the inner and outer margins of the rims of individual specimens. Figure 9/1-L gives data of this type, from my own and published micrographs. Rim width and ellipticity variation for fifty placolith coccoliths, are plotted on two graphs (Figs.9/1,J). As these show rim width is constant, and so outer ellipticity is consistently lower than inner ellipticity. This relationship is also shown by data from a single species, Pseudoemiliania lacunosa (s.l.), Figure 9/K. It appears to be a basic feature of coccolith geometry. This has not always been appreciated, as demonstrated in Figure 9/L, in which similar data from published drawings of coccoliths is plotted. In most cases the rim width at the ends has been drawn as greater than that at the sides.

4.1.2 Element orientation. If the distorted coccolith is compared with the circular coccolith from which it was derived (Figs.9/B,C), it can be seen that the angular separation of the elements is increased along the sides of the distorted coccolith and decreased at the ends. Thus the elements fan strongly at the sides and only slightly at the ends. This again is the reverse of

43 FIGURE 9 - COCCOLITH RIM GEOMETRY. A-C. Computer generated diagrams. A. Elliptical coccolith modelled by my program, based on E; B. Circular coccolith derived from A; C. Elliptical "coccolith" formed by stretching B. D-H. Tracings of real coccoliths, scale bars one micron. Sources ROTH 1970, BUKRY 1971c, 1974a, PERCH-NIELSEN 1971, 1977. D. Emiliania huxleyi, proximal; E. Coccolithus pelagicus; F. C.crater; G. Calcldlscus leptoporus; H. BUlpsollthus macellus, proximal. I-L. Biometric data illustrating rim width constancy. I,J. Data from 50 miscellaneous specimens; K. Data from 40 Pseudoeailianla lacunosa specimens, as in figure 13; L. Comparative results from measurements of drawings of coccoliths. the normally observed relationship on real coccoliths. As shown by the tracings, Figure 9/D-H, maximum fanning occurs around the ends, and the elements along the sides are often sub-parallel. The fanning effect is shown particularly clearly by elongated coccoliths such as Bllipsolithus macellus (Fig.9/H). A related effect is shown by central area elements, in real coccoliths (e.g. Emiliania huxleyi, Fig 9/D) these tend to be bunched at the ends, instead of converging uniformly on the centre. Both these effects suggest that the orientation of elements is more closely related to the orientation of the rim (i.e. the local tangent), than to radial directions from the centre of the coccolith. This and other aspects of the geometry of single coccoliths are hard to measure reliably, without very high quality enlarged micrographs. They can, however, be qualitatively observed in published micrographs of Mesozoic or Tertiary coccoliths. Among many possible sources BLACK (1972-1975) is worth mentioning, since he also gives a valuable discussion of various aspects of coccolith geometry.

4.1.3 Element spacing. In the distorted coccolith the element spacing is directly related to the angular divergence, and so is maximum along the edges and minimum at the ends. This again is the reverse of the generally observed relationship, usually elements are broader at the ends of elliptical coccoliths than along their sides (Figs.9/D-H). Closer examination shows that this is predominantly an effect of the outer margin. By contrast, around the inner margin of coccolith rims element spacing is usually more or less uniform. The variation in the width of the ends of the elements is thus a product of the variation in angular divergence, or ray fanning.

4.1.4 Discussion. This succession of mismatches between the coccolith morphology derived by distorting a circular coccolith and that of real coccoliths makes it plain that coccolith morphology is not controlled in this manner. An attractive alternative suggested particularly by the constancy of rim width, with its implication of uniform element length is to consider coccoliths as being formed from a cycle of elements of constant morphology - with final form determined by the shape of elements, their orientation and spacing. Analogous structures can be created with a deck of playing cards arranged in an elliptical pattern.

45 4.2 COCCOLITH DEVELOPMENT DURING COCCOLITHOGENESIS An alternative approach to looking at the final form of coccoliths is to examine how they actually develop. As discussed above (Section 2.5), coccolithogenesis in heterococcolith producing phases is an intracellular process closely associated with organic scale formation. The process has been followed in detail in three species, Coccolithus pelagicus (PARKE & ADAMS 1960, MANTON & LEEDALE 1969), Emiliania huxleyi (WILBUR & WATABE 1962, KLAVENESS 1976, WESTBROEK et al 1984), and Pleuzochzysis caztezae (MANTON & LEEDALE 1969, OUTKA & WILLIAMS 1971). Although there are important differences between coccolithogenesis in these species they are primarily biochemical and the sequence of morphological development is similar in all three. So a general pattern can be deduced, as shown in Figure 10. In each case an organic base-plate scale has been observed to form before calcification starts (Fig.lO/A). In the case of E.huxleyi this was not identified until the study of WESTBROEK et al (1984), although it was illustrated by WILBUR & WATABE (1962). Initial calcification occurs around the rim of this base-plate, producing a proto-coccolith ring of simple elements (Fig.lO/B). Good examples are illustrated in KLAVENESS (1976) and LEADBEATER ,& MORTON (1973, my Fig.4/D), in them the elements appear to be uniformly spaced around the ring. Various workers have suggested that the base-plate is important at least in providing a frame of reference for the nucleation sites (MANTON & LEEDALE 1969, OUTKA & WILLIAMS 1971, WESTBROEK et al 1984). During subsequent coccolith growth (Fig.lO/C-D) the elements remain attached to the base-plate, so it is unlikely that they move relative to each other. Hence the element spacings and orientations determined by the initial nucleation should be retained during coccolith growth. In E.huxleyi the base-plate continues to grow after nucleation of the proto-coccolith ring, and so its margin does not correspond with the margin of the proto-coccolith ring (WESTBROEK et al 1984). In the other species base-plate growth terminates with crystal nucleation. Coccolith growth can only be crudely followed in sectional material, and the regulatory mechanism is not clearly understood (WESTBROEK et al 1986). Essentially though an organic membrane appears to closely surround the growing elements and to gradually extend with calcification occurring within it. The coccolith vesicle also extends during this process, and migrates toward the outside of the cell. In E.huxleyi, growth occurs in upward, outward and inward directions so that the proto-coccolith ring approximates to the base of the proximal shield. In Umbilicosphaeza sibogae var. foliosa, for which this stage of coccolithogenesis has been described by INOUYE & PIENAAR

46 J - . -Sj ' J L \ V V , 1 l •* ■,-::!Kl h x> : \ m m r n t 0

n liliiii B V ’Si :>v - t.Txy> r /—i 1 i r t r > i'-'l , ''V f r ,4 . / ) \\ / / _ _ -••<■:./'•■:■■■■.■.■:,[ .• • i ; t-;c- «• ••:• vV ,-

i, V/,y7:V!i !! ;i i■■•■■:■■•.;:'!■ 7 /■ r-rj'T.:;-rJ \ \ Y'

D ■t\\ \\\ U r

£ I c.<\rt.C\

Series of diagrams showing intracellular coccolith development. Organic base-plate scale shown by dotted lines. Left - sections, right - plan views. A. Base-plate only; B. Base-plate and proto-coccolith ring; C. partially formed coccolith; D. Final form.

FIGURE 11 (bottom) - RETICULOFENESTRID COCCOLITH ELEMENT FORM. Simplified drawing of a single reticulofenestrid segment, showing its complex multi-part construction. N.B. For clarity the elements are shown as of similar thickness, in fact the , tube elements are rather thicker. Also kinks in the rays, grill complexities, etc., are omitted.

47 (1984), growth occurs in a very similar manner. Pleurochrysis carterae is also similar, although there is little or no outward growth, as the coccolith remains a ring shaped cricolith rather than developing into a placolith. In Coccolithus pelagicus the coccolith development stages were not so readily discernible. It appears likely, however, from the illustrations of MANTON & LEEDALE (1969) that in this species too, growth occurs outward and upward from the proto-coccolith ring (their figs 22-28) so that it corresponds to the base of the tube, and that in the mature coccolith the base-plate is attached to the inside edge of the proximal shield (their figs.16-17 & 31).

4.3 A BASIC MODEL FOR HETEROCOCCOLITH DEVELOPMENT It appears that the coccolith development process in these species can be divided into three discrete stages. These stages combined with the information from coccolith geometry constitute a reasonable model for the development of heterococcoliths. The first stage is formation of the base-plate scale. This is a critical precursor to coccolith development since the base-plate acts as a template for crystallite nucleation. Its shape and size is thus a major influence on the ultimate form of the coccolith. The second stage is crystal nucleation around the base-plate. It is at this stage that the number, spacing and orientation of the elements is fixed. The observations from coccolith geometry suggest that the elements should be uniformly spaced at this stage and should have a constant orientation relative to the base-plate margin. The final stage is growth of the elements. This occurs in various directions but leaves the proto-coccolith ring on the proximal surface of the coccolith. The constant width of coccolith rims suggests that elements develop to uniform lengths during this process. All the elements of a cycle also tend to have similar forms; apart from width variation due to differences in curvature of the ring. The asymmetry introduced by elliptical form appears to be compensated for by a slight expansion of elements without any alteration in their basic form. An interesting aspect of the process is that calcification does not occur according to a fixed model, or within a predetermined template, but rather is a growth process of variable pattern. This point can be illustrated by comparison with other groups. In the formation of a foraminiferal chamber the shape is initially defined by an organic membrane and this acts as a template for subsequent calcification. Analogous processes occur in the formation of ostracod carapaces, diatom frustules, and silicoflagellate . In each case an organic

48 precursor closely defines the form of the final mineralized . This analogy can only be applied to the nucleation phase of heterococcoliths, although precursor base-plate growth and subsequent element growth considerably influence final coccolith form. As a result the final coccolith form can vary greatly in such obvious features, as size, ellipticity, number of elements, and central area to rim ratio.

4.4 RETICULOFENESTRID STRUCTURE To test whether the developmental model discussed above has any validity or utility it needs to be applied to real coccoliths. This inevitably involves a considerable digression into taxonomy, so only a single example group is discussed in detail here, the reticulofenestrids. They are suitable since they form a coherent, abundant, and well known group, and since coccolithogenesis has been studied in the principal living species, Emiliania huxleyi. Also they show an interesting range of structures. In this section, and elsewhere, I use the term reticulofenestrid for those coccoliths with the same basic rim structure as Reticulofenestra. This includes all the Late Eocene to Recent Prinsiaceae, but excludes the earlier genera Toweius and Prinsius, which have a significantly different rim structure (see e.g. PERCH-NIELSEN 1985b). The most important Neogene species are illustrated in Figure 12. Basic reticulofenestrid structure is discussed first, to show how it can be related to growth from a proto-coccolith ring. Then each of the three basic steps of coccolith development suggested above is looked at in turn to see how it affects aspects of morphology and variation in the group.

4.4.1 Basic structure Host reticulofenestrids consist on the proximal side of a shield constructed of a single cycle of rays and a grill flooring the central opening. On the distal side two cycles of elements are visible, an outer cycle forming the distal shield and a discrete inner cycle of smaller "cover plates" overlying the distal shield. A central tube connects the proximal and distal shields, this consists of two cycles, an outer cycle with clockwise imbrication of the elements and an inner cycle (termed the wall) with anticlockwise imbrication (i.e. the tops of the elements are offset in an anticlockwise sense from the bottom of the elements). Thus a total of six cycles can be identified (proximal and distal shield, wall, , : , outer tube, cover plates, and grill). However, closer examination shows that the various cycles are connected. Thus the cover plates arise from the elements of the wall, which in turn are

49 FIGURE 12 - RETICULOFENESTRID COCCOLITHS. Computer generated diagrams of eight species. N.B. 1. Diagrams are not drawn to a common scale, and the cross-sections are at larger scales than the plan views. 2. Central area details and cross sections are schematic. 3. All diagrams are based on information from many sources rather than single specimens. A. D^r^ococcm-s daviesii; B. Dictyococcites scissuza; C. Cyclicargolithus floridanus; D. Reticulofenestra pseudoumbilica; E. Pseudoemiliania lacunosa; F. Noelaezhabdus bozinovicae; G. Emiliania huxleyi; H. Gephyzocapsa oceanica. 50 connected to the elements of the proximal shield, and so to the grill. These relationships are readily apparent in suitable micrographs and have been noted by many authors (e.g. HAY et al 1966, HAQ 1968, PERCH-NIELSEN 1971, EDWARDS 1973, ROMBIN 1979). Similarly the distal shield rays continue into the outer tube cycle, and this again merges with the proximal shield. This relationship is not so obvious, since it requires unusual specimen orientation to observe it, it is, however, clearly shown in micrographs of PERCH-NIELSEN (1971, pi.24/3, and noted p.31), BRAMLETTE & WILCOXON (1967, pi.1/3), and STEINMETZ & STRADNER (1984, pi.26/4). Thus all the various apparently distinct elements are interconnected at the base of the tube, and so are pants of a single cycle of unuts, of remarkably elaborate form (Fig.11). (N.B. This structure was also noted by Gallagher and Driever, oral comms., INA Meeting London 1987). Light microscopy also suggests this interconnection of elements - all the components of the coccolith show similar optical behaviour in both plan and side view. The central grill is rarely visible in cross-polarised light but this is almost certainly a result of its extreme thinness rather than different optical orientation. This structural interpretation, with all the elements traceable to a common origin at the base of the tube, provides strong support for the concept of them growing from a proto-coccolith ring corresponding to the base of the tube. This is a critical prediction of the model outlined above. Conversely the model provides a convincing explanation for the element form.

4.4.2 Effect of proto-coccolith size and ellipticity. An obvious consequence of the growth model is that one of the primary controls on coccolith morphology is the size and shape of the proto-coccolith ring, which in turn is related to the base-plate size and shape. In general shape, or ellipticity, seems to show a low degree of variation within species, and size rather more (see also Chapter 6). An interesting example of the effects of combined variation in these factors is provided by the Pseudoemiliania lacunosa - Reticulofenestza dozonicoides group. This group is usually regarded as consisting of two or more species, with R.doronicoides a typical small elliptical reticulofenestrid, and P.lacunosa a slightly larger circular to sub-circular form with awide central area, numerous elements and slitting between the rays of the distal shield (Fig.l2/E). There is, however, a continuous range of morphotypes between these end-members, with correlated variation in size, number of elements, central area openness, roundness, and slitting.

51 Almost certainly only a single variable species is involved. This interpretation was based initially on my own, mainly qualitative, light microscope observations, but the range and style of variation is also clear from published illustrations and descriptions (e.g. McINTYRE et al 1967, SAMTLEBEN 1978, NISHIDA 1979b, 1980, 1981, PUJOS 1985a). Similar interpretations of the taxonomy of the group have been made by McINTYRE et al (1967) and SAMTLEBEN (1978). The large number of morphological variables makes this a suitable group for biometric study. I have conducted a crude study of this type, using about forty electron micrographs of P.lacunosa, mainly from the literature. Graphs from this data set are given in Figures 9,13, & 14, and a synthetic series of diagrams based on it in Figure 13. Unfortunately it is not possible to identify precisely the proto-coccolith ring location, particularly since distal views have had to be used to get a reasonable sized data set. However, since element growth is predominantly outward, aperture shape forms a useful approximation. As shown in Figure 13/F aperture length and width are strongly correlated, but not directly proportional; ellipticity decreases with increasing size (see also Fig.l4/E). As a result aperture width, which is related to both increasing aperture size and decreasing ellipticity, shows greater variation than aperture length. It proved the most discriminating parameter to plot other data against. The parameters which are most nearly independent of aperture width are rim width (Fig.14/F), and element spacing around the aperture edge (Fig.l4/D). The variation in these parameters is low and apparently random, or related to other controls. This is of considerable interest since it suggests that variation can occur in one phase of the coccolith development process with little effect on the other phases. An inevitable consequence of the low rim variation is that central area size increases relative to coccolith size,as the aperture width increases (Fig.l3/G). Similarly since initial element spacing is approximately constant the number of rays increases with aperture width (Fig.l3/H). The number of slits between rays increases very markedly (from 0 to 50), with aperture width. This is partially an effect of the increasing number of rays, but in addition the percentage of rays with slits increases (Fig.13/1). The cause of this is not clear, but slitting and the change to broader more open coccolith forms might both be mechanisms for minimising calcite usage, and so could be parallel responses to an external factor. The distribution of slits around the coccoliths appears to be random, as modelled in Figures 13/A-E. Summary: A large number of superficially independent changes in coccolith form can be seen to be related to the single control of

52 fil

Rn

88 70 o 80 + + 50 40 30

20 + + + + + 0 + + 10

0.5 1.0 1.5 2.0 2,5 3.0 Percentage of rays with Slits vs Aperture width (nitrons)

FIGURE 13 - PSEUD0EMIL1ANIA LACUNOSA VARIATION. A-E. Computer generated series of coccoliths, illustrating correlated variation in aperture shape, aperture width, number of rays, and degree of slitting. N.B. these diagrams are to constant scale, scale bar one micron (based on mean rim width of 1.4 microns). F-I. Biometric data, from measurements of photographs (mainly published) of P.lacunosa (s.l.) specimens, illustrating co-variation of various parameters with aperture width. Circles represent the computer generated specimens of diagrams A-E. Dotted lines in F show ellipticity (length:width ratio). There are further graphs from this data set in Figs.9 & 14. 53 Rw 0 , 25-

0.2 0-

0 . 1 5 -

0.18-

0.05-

A 2 4 6 8 10 12 14 16 18 20 Cl 8.5 1.0 1.5 2.0 2.5 3.0 fiw Initial ray width vs Coccolith length (nicrons) - RETICULOFENESTRIDS Initial ray width vs flperturo width (microns) - P.lacunosa

RTw ftl/fiw 0.60 n 1.8 -1

8.50 1.7 -

1.6 - 0 . 4 0 - 1.5 - + ** * + 0.30 - 1.4 -

1.3 - 0.20 -

1.2 - 0.10- + + 1,1 -

B 2 4 6 8 10 12 14 16 18 20 Cl 0.5 1.8 1.5 2.8 2.5 3.0 A« Ray tip width vs Coccolith 1 an91h <«icrons> - RETICUL0FEHESTR1DS Aperture ellipticity vs Aperture width (microns)

Rn <

120 - ] 3.0 -1

100 - 2.5

80 2.0 -

60 + *•'+ + 1.5 - / * 40 - i. , 1.0 -

20 - 0,5

0 2 4 6 8 10 12 14 16 18 20 Cl 0.5 1,8 1.5 2.0 2.5 3. Aw Nunbfr of rays vs Coccolith lonyth (microns) - RETICULOFENESTRIDS Rim width (microns) vs Aperture width (microns)

G FIGURE 14 - RAY WIDTH VARIATION. A-C: Biometric data from measurements on miscellaneous Eocene to Recent BIOMETRIC PARAMETERS reticulofenestrids. Cl - Coccolith length D-F: Biometric data from P.lacunosa Cw - Coccolith width specimens (same data set as in A1 - Aperture length Fig.13, circles represent diagrams of Aw - Aperture width Fig.l3/A-E). A, D. Direct illustration of the low, Rw - Initial Ray width and independent, variation of initial RTw - Ray tip width ray width. B-C. Indirect effects of Rn - Number of rays this, on ray tip width and number of Sn - Number of slits rays; dotted line in C represents data of BACKMAN (1980). E-F Further aspects of P.lacunosa N.B. These are the measured geometry: related variation of features, additional parameters ellipticity and aperture width; and such as width and ellipticity independence of rim width. are calculated from them.

54 increase in base-plate size, with faster increase in width than length. This is demonstrated in the diagrams, Figure 13/A-E, the only parameters changed in the program were aperture width, ellipticity, and degree of slitting. Interestingly there are analogous cases of correlation in these features, except for slitting, in other groups, notably Reticulofenestra pseudoumbilica - rotazia, and Calcidiscus leptoporus- macintyrei, variation in the Coccolithus pelagicus group also shows some such features. More generally variation in base-plate development appears to be an important process in producing inter-specific and intraspecific variation.

4.4.3. Nucleation related features. The second phase of coccolith development is crystal nucleation, when the spacing and orientation of elements are determined. It is readily apparent from electron micrographs that there is little variation in the orientation of the elements, between the various reticulofenestrid species. Similarly light microscopy shows that the elements have constant crystallographic orientation. More surprisingly element spacing is also remarkably constant within the group. This can be demonstrated by direct measurement of the width of the rays at the inner margin of the proximal shield (assuming again that this approximates to the proto-coccolith ring). Measurements of ray width in a range of species, mainly from published micrographs, are given in Figure 14/A. The ray widths only show a two-fold variation (0.1 to 0.2 microns), in comparison to a ten-fold variation in coccolith length. A similar pattern is shown by the results from within the single species P.lacunosa (Fig.l4/D). The accuracy of the measurements is not high - with errors likely from inaccurate quotation of magnifications, specimen tilting, and choice of location to measure elements. These problems should, however, increase rather than decrease the variability. Ray tip width is more variable since it also depends on the degree of element fanning, and on rim width, it consequently shows a general correlation with coccolith size (Fig.l4/B). A secondary effect of the constancy of initial ray spacing is that the number of rays increases with proto-coccolith ring size. Within the reticulofenestrids the rim width to central area width ratio is usually approximately constant so the number of rays also approximately correlates with total coccolith length. The relationship between number of rays and coccolith size was investigated by BACKMAN (1980), using members of the Reticulofenestra pseudoumbilica group. He showed that the number of rays was linearly related to coccolith length within this

55 group. His trend is shown in Figure 14/C, together with data from other elliptical reticulofenestrids (same specimens as in Fig.l4/A). Despite the range of species, coccolith sizes, and central area sizes, there is still a strong trend to the data. This low variation in both element orientation and spacing suggests that the nucleation processes are a rather constant factor during evolution of the group. As such they are of importance as higher level taxonomic features.

4.4.4 Element growth variation. Apart from the slitting mentioned above, reticulofenestrid rim cycle elements are remarkably similar in all species, with the basic structure illustrated above (Figure 11). Constant features include the four part form, the imbrication directions, and kinks in both the proximal and distal shield rays (Fig.12). Also most reticulofenestrid coccoliths have pointed ray tips, this is probably related to the crystallography of the elements, and may in part be of diagenetic origin. The grill and particularly the wall cycles show considerably more variation in mode and degree of development, and so are important in subdivision of the group. In Cyclicargolithus flozidanus and other Early Miocene and Oligocene reticulofenestrids the wall elements, or a limited number of them, overlap onto the distal shield as cover plates (Fig.l2/C). In most later reticulofenestrids there are no discrete cover plates, instead a raised collar occurs around the central area (Fig.l2/E). This is formed jointly by the wall cycle elements and the distal shield elements, which butt against them. A third type of development is for the wall to terminate flush with, or slightly below the edge of the distal shield, without development of a collar or cover plates (Fig.l2/G). This morphology is most characteristic of Erailiania huxleyi, but is also seen in some Gephyzocapsa specimens. In all three cases the wall cycle can be thickened inwards to close the central area. Forms with this development are often placed in a separate genus, Dictyococcites. However specimens of Emiliania huxleyi with closed central areas have been illustrated by several authors (e.g. BLACK 1968, PUJOS-LAMY 1977, NISHIDA 1979a, HEIMDAL & GAARDER 1981, my Fig.l5/D), and this type of variation seems more likely to be of ecological than taxonomic significance. Probable exceptions are Dictyococcites scissuza (Figure 12/B) in which the central area is closed by a plug of inward growing cover plates; and Dictyococcites daviesii in which a limited number of wall elements contribute to an expanded grill. The general variability of wall cycle development is continued by

56 two more radical structures. In Gephyzocapsa wall elements on opposite sides of the distal shield become elongated and arch over to form a bridge (Fig.l2/H). The bridges always show the same diagonal orientation (NE-SW), and frequently the elements fail to meet precisely, with anti-clockwise offset. Etched specimens show that each half of the bridge is formed from two to five wall elements. Noelaerhabdus (a late Miocene Paratethyan genus) has a single asymmetrically placed spine similarly formed from a few wall elements. The development of different structures from the wall is thus a consistent theme in reticulofenestrid variation. The development of new features in this way does not involve the addition of new crystal elements to the structure, or even much modification of the old ones, but only the addition of an extra phase of development. In this respect the different developments fit well with the model proposed above. On the other hand the spine of Noelaerhabdus and the bridge of Gephyzocapsa show that elements in one part of the cycle can be grown differently to those in other parts of the cycle. This is an important modification of the general pattern of similar development of all elements, here final form appears to be regulating element development, rather than vice versa. Some degree of interplay is nonetheless evident, in the offset of the bridge halves in Gephyzocapsa, and in the eccentricity of the spine in Noelaezhabdus. Also malformed Gephyrocapsa specimens with three or more bridge plates grown from different parts of the tube do occur (e.g. OKADA & McINTYRE 1977, pi.3). In all these respects the final structure is distorted from symmetrical perfection by the developmental process. To summarize it appears that basic element morphology is constant in reticulofenestrids. Major modifications are mainly confined to addition of extra features and are usually taken as generic level features. Intraspecific variation can produce changes in the degree of development, causing variations in rim width, in degree of slitting, and in central area closure.

4.5 EXTENSION OF THE MODEL TO OTHER GROUPS The following notes briefly outline the extent to which the three-phase / proto-coccolith ring model can be applied in other Neogene coccolith families. In each case their structure is described and illustrated in more detail in the taxonomic sections.

57 4.5.1 Coccolithaceae The Coccolithaceae, like the reticulofenestrids, are placoliths. Coccolithogenesis has been studied in two species, Coccolithus pelagicus and Umbilicosphaeza sibogae. These seem to develop in the same fashion as B.buxleyi (see above, section 4.2). However, they have at least two discrete cycles of elements, since the proximal and distal shields show contrasting optical behaviour, and the proximal shield is bicyclic. My interpretation of the structure is that there are two cycles (discussion and illustrations in the taxonomic section, Chapter 11). An upper cycle (non-birefringent) forms the distal shield, tube, and upper part of the proximal shield. The other cycle (birefringent) forms the lower part of the proximal shield. This obviously requires two separate sets of crystal nuclei. The plates of MANTON & LEEDALE (1969, their £igs.28 & 31, my Fig.4/1) provide striking evidence for this. They show the edge of the base-plate curving up during proto-coccolith formation, and separate crystals developing on either side of it, thus forming a double proto-coccolith ring. Further complexity is provided in Coccolithus, Chiasmolithus, and Cruciplacolithus by the presence of a wall (i.e. inner tube elements), and by central area structures. These are birefringent like the proximal shield, but on the basis of the interpretation given above are separated from it by the tube - distal shield cycle. Hence these also require separate nucleation, which considerably increases the amount of crystal nucleation, although nucleation still seems likely to occur on the base-plate.

4.5.2 Helicosphaeraceae. The coccoliths, or helicoliths, of Helicosphaera consist of three m®,r| /parts, proximal plate, flange and blanket (cf. THEODORIDIS 1984, and see taxonomy). Helicosphaera has not been cultured, so there is no information on coccolithogenesis in the genus. NORRIS (1971) has, however, illustrated a base-plate scale on H.carteri, entirely covering the proximal plate; and GAARDER (1970) recorded similar observations. Hence this is a likely position for the proto-coccolith ring. Strong support for this is provided by the form of the elements of the proximal plate, which show ray bunching and wedging inward suggesting inward growth (Plate 4/26,27). The proximal plate can thus be interpreted as a normal cycle, but grown inwards, like the grills in reticulofenestrids. The flange is composed of the same number of elements as the proximal plate, and they have similar similar optical orientations. It is, however, a strongly modified cycle, with considerable variation in element form and length, to produce the spiral effect. This is a sharp contrast to the very uniform rim development deduced for typical

58 coccoliths, and indicates a significant modification of the coccolith development process. A further problem is provided by the blanket. The elements of this unit have different optical orientation from those of the flange and proximal plate. However, they are floored by the proximal plate, and so must have nucleated on this rather than the organic base-plate.

4.5.3 Pontosphaeraceae. The Pontosphaeraceae have an analogous structure, with proximal plate, flange and inner/upper cycles of smaller elements. GAARDER (1970) has illustrated a base-plate scale entirely covering the proximal plate, on an ordinary coccolith of Scyphosphaera apsteini i. The flanges in both the Helicosphaeraceae and Pontosphaeraceae display anti-clockwise element imbrication, but otherwise have entirely different forms. Direct relationship between the groups is by no means certain, but in terms of the coccolith development they appear to be similar. There are also distinct similarities between the blanket elements in these two groups and the wall elements of the coccolithaceae. In all cases they are small, sub-vertically orientated, lath-like, elements arranged in numerous irregular concentric cycles (Plate 4/24). They show strong birefringence in plan view, with an oblique extinction cross. It would appear that they are the product of a different type, or fourth phase, of development, characterised by formation of numerous small elements, with less precise distribution in cycles, and less precise regulation of growth.

4.5.4 Holococcoliths. The diagnostic feature of holococcoliths is that they are formed of numerous minute identical crystallites. In addition holococcolith elements are not arranged in uniform cycles, few elements are in contact with the base-plate, and elements show similar orientations over large zones, forming irregular pseudo-crystals. Plainly the concept of development from a proto-coccolith ring is not applicable to holococcoliths. A quite different developmental model is needed to interpret their morphology, this is also of course suggested by the evidence that they form by extra-cellular calcification (Section 2.5).

4.5.5 Other groups. The model of development from a proto-coccolith ring is probably applicable to Syracosphaeraceae, Calciosoleniaceae, Zygodiscaceae and most Mesozoic heterococcoliths, but I have not examined them in detail. Discoasters and sphenoliths can readily be envisaged developing in an

59 analogous manner, but with little or no base-plate development. Groups with structures which appear to reflect different development processes include: Rhabdosphaeraceae - spirally arranged cycles (but outer rim is analogous); Braarudosphaeraceae - layered ultrastructure; Ceratolithaceae and Triquetorhabdulaceae - totally different form, and rodded ultrastructure.

4.6 SUMMARY 1. On the basis of observations of coccolith geometry and of studies of coccolithogenesis a three phase model for heterococcolith cycle development can be proposed. The phases are: base-plate formation; crystal nucleation; and element growth. The first two stages produce a proto-coccolith ring, the third develops it into a coccolith.

2. Element nucleation and growth processes are usually uniform around cycles, except where elements interfere with each other. Departures from this pattern, such as flange development in Helicosphaera and bridge formation in Gephyrocapsa probably require functional explanations.

3. Structural complexity is primarily a result of elaborate element form, rather than production of numerous cycles. Variation, including evolution, can occur during any of the developmental phases with varying effects on final form, but some aspects are more stable than others. Most stable are nucleation related features - element spacing, orientation and number of cycles. Element form is only slightly less stable, but element length and degree of calcification are both, and independently, liable to intraspecific variation. Base-plate, and so proto-coccolith ring, diameter is similarly variable, and in some species is accompanied by shape variation.

4. This model is applicable to the development of typical rim, flange, and proximal plate cycles in Neogene heterococcoliths. It does not appear directly applicable to some central area structures, to blanket elements, or to holococcoliths.

5. The model is directly applicable to the problem of mathematically describing coccoliths, and forms the theoretical basis of the computer program used to produce illustrations of coccoliths.

60 CHAPTER 5 - SPECIES AND INTRASPECIFIC VARIATION

5.1 INTRODUCTION "The species concept Species of calcareous nannofossils are defined typologically. As a consequence of this species concept two morphologically different, fossil coccoliths will be assigned to two species. Morphologically distinct specimens may, however, well have belonged to a single coccosphere, as it is known that Recent coccospheres may bear several types of coccoliths and that the types of coccoliths may change considerably in successive phases of the life cycle of a species (Parke and Adams, 1960). In addition, morphological differences in coccoliths of Recent species may be the result of differences in temperature (Watabe and Wilbur, 1966). Thus, a fossil species and a living species are different entities.n ROMEIN 1979, "Lineages in early Palaeogene calcareous nannoplanktonn, p.13.

"It is possible to distinguish circular as well as slightly elliptical specimens of Pseudoemiliania lacunosa. One or the other shape may dominate in a given interval over the range of the species, with circular forms being less common in the Pliocene, however, to date no consistent relationship has been established between the shape of this species and the age of the sediments in which that shape predominates. The distinction of an elliptical species, although it can be made readily, does not seem to be useful." GARTNER 1977, "Calcareous nannofossil biostratigraphy and revised zonation of the Pleistocene", p.17.

The first quotation above shows how various worries about special biological features of coccolithophores have resulted in a cautious approach to the interpretation of fossil species in biological terms. The principal of these problems are polyphase life histories, and polymorphic coccospheres (Romein's comments on typological definitions are not strictly valid). The second quotation illustrates the practical approach to species which, largely as a consequence of these worries, has been adopted by many nannofossil workers. Species are erected on the criteria of recognisability, and of biostratigraphic utility, rather than biological validity. In the particular case discussed it is likely that only one, rather variable, species is involved (see discussion above, Section 4.4.2). Gartner does not split the species, but only on the grounds

61 that this is not stratigraphically useful. In this chapter I first discuss the various problems posed by Haptophyte biology, and try to show that they do not present any very extraordinary difficulties for the palaeontologist. Then I discuss intraspecific variation in living coccolithophores, and the conclusions which can be drawn from it. Application of this approach, to Neogene nannofossils, forms a major theme of the taxonomic section of the thesis.

5.2 SPECIAL TAXONOMIC PROBLEMS OP COCCOLITHOPHORES 5.2.1 Polyphase life cycles. As described above (section 2.3) most haptophytes have life histories with discrete motile and non-motile phases. The most important example for taxonomy of such polyphase life cycles was the discovery of PARKE & ADAMS (1960) that the species Coccolithus pelagicus had, in addition to the non-motile placolith bearing phase, a motile holococcolith bearing phase. This phase had previously been described as a different species, Crystallolithus hyalinus, and placed in a separate family to the non-motile phase. The discovery that it was con-specific with C.pelagicus raised the spectre of many seemingly unrelated species being merely different phases of single species. This seemed to suggest that the classic approach to haptophyte taxonomy, of using coccoliths and scales as the primary classificatory criteria, could be highly misleading, and hence that protoplast characteristics should be used preferentially for taxonomy. As a result it looked unlikely that classification of fossil coccoliths would ever be very meaningful. Fortunately work in the subsequent quarter century has done little to substantiate this pessimistic attitude. In particular: A. Phycologists and marine biologists have continued to find scales and coccoliths the most reliable cell components for taxonomy. B. Only motile to non-motile phase changes have been shown to involve substitution of cell-wall components (as opposed to their modification or omission). C. Other patterns of motile to non-motile phase change are known in coccolithophorids which do not involve two coccolith types. In particular the motile phase of Emiliania huxleyi bears scales, not coccoliths. D. Although it has often been suggested, from the C.pelagicus - C.hyalinus case, that many heterococcolith species may have holococcolith bearing motile phases, this suggestion is not supported by any evidence of pairing in terms of geographical or geological

62 distribution. Also many heterococcoliths are motile, including most, if not all, members of the Syracosphaeraceae and Helicosphaeraceae. It is probably reasonable to conclude from these observations that the C.pelagicus type of life-cycle is not widespread, and that it is unlikely that many haptophyte species have more than one phase capable of producing fossilisable coccoliths.

5.2.2 Polymorphism. It is common for single cells to bear coccoliths of two or more consistently different types, representing some degree of functional specialisation, a phenomenon generally referred to as polymorphism. The most widespread type of polymorphism is related to the asymmetry of motile cells. This frequently causes variation of coccolith shape and size over the test; illustrated examples include Calciosolenia (Fig.7/F, p .37), and Helicosphaera (Fig.l5/A, p.67). Also specialised coccoliths are often developed around the flagellar opening. Typically such coccoliths have the same structure as the others, but differ in size or shape, or bear additional spines or flanges (Fig.15/A,C). In some cases these structures are very strongly developed, probably to aid orientation or retard sinking and similar structures are often developed at the opposite end of the coccosphere (Fig.7). A few modern taxa show stronger polymorphism, with two radically different coccolith types on a single cell. The coccospheres of Scyphosphaera bear both elevated and low rimmed lopadoliths (Fig.l5/C). Most Syracosphaeraceae have a two layered test with quite different coccoliths in the two layers (Fig.6/E). Calciopappus, and Michaelsarsia have whorl coccoliths of extraordinary elongated form, as well as dimorphic normal coccoliths (Fig.7/A).

5.2.3 Polyspecific coccospheres. Various workers have produced electron micrographs purportedly showing coccospheres, bearing coccoliths of more than one species. The main published examples are in CLOCHIATTI (1971), WINTER et al (1978), and GARD (1987). The first two are coccospheres of Gephyrocapsa oceanica and Fmiliania huxleyi, the third is a coccosphere of Calcidiscus leptoporus and Coccolithus pelagicus. Due to the phylogenetic affinities of G.oceanica and E.huxleyi both Clochiatti and Winter assumed their examples were genuine, and speculated that they represented some type of dimorphism. Gard considered this less likely for her specimen but was unable to make an alternative suggestion. If they were genuine then they would indicate that single coccolithophores were able to produce quite different coccolith types at random, and as such would constitute proof of the most pessimistic assessments of

63 coccolith utility as a taxonomic criterion. I have, however, also been shown several examples of similar coccospheres by S.Nishida (pers. comm. 1984), from Pacific Ocean water samples. These consisted mainly of Coccolithus pelagicus and Calcidiscus leptoporus, but a few contained three or more unrelated species. Unlike true dimorphic coccospheres there was no regular arrangement of the coccolith types. The interlocking of the coccoliths was also usually less regular, and in a couple of cases coccoliths were even upside-down. The published examples are not so obviously artificial, but they all include coccoliths with a wider range of sizes and preservation states than is normal on single coccospheres. These observations make it extremely unlikely that they are genuine coccospheres. Instead the coccoliths have probably been collected by another protist, and used to form an agglutinated wall, analogous to those of textulariid foraminifera. Tintinnids are known to form tests from coccoliths in this way (e.g. WINTER et al 1986), but are significantly larger, and cup shaped rather than spherical.

5.2.4 Discussion These three problems (polyphase life cycles, polymorphism, and polyspecific coccospheres) are the only ones which could be considered to create taxonomic problems of such a magnitude that normal species concepts cannot be applied to fossil coccoliths. Of them polyspecific coccospheres are almost certainly artefacts, and so not a problem, whilst polyphase life-cycles are unlikely to produce two separate fossilisable coccolith types. Polymorphism is unquestionably a real problem, but it is not insurmountable. The milder types of polymorphism, where the polymorphs are variants of the main coccolith types, can readily be recognised once it is known that this sort of variation affects the family in question. Even using observations on isolated fossil coccoliths this has often been successfully achieved. For instance, in Helicosphaeza wing development has long been recognised as a variable feature, even though one of the main causes of this, circum-flagella variation, does not seem to have been realised. Similarly the presence or absence of spines, and their size, is well known to be a feature which requires careful utilisation. Where polymorphism produces genuinely different coccolith types it is more of a problem. However, such species are rare,' and their specialised coccoliths tend to be delicate, and so even rarer in the fossil record. There are very few records of fossilised exothecal coccoliths, and none of whorl coccoliths. Scyphosphaera is probably the only genus which presents significant problems of this type.

64 Furthermore, although only the most robust coccospheres are usually preserved, under exceptional conditions even very delicate coccospheres can be fossilised. The most important facies for this type of preservation appears to be laminated anoxic shale (e.g. NOEL 1973, GOY 1981). Even a few coccospheres from such deposits can provide critical evidence on polymorphism, as shown by the recent work of COVINGTON (1985) and LAMBERT (1986). Most other fossil groups also have taxonomic problems, caused by such factors as sexual dimorphism, ontogenetic variation, and indeed polymorphism. In comparison those of coccolithophores are not exceptionally serious. So there are in my opinion no special reasons why, to paraphrase Romein, a fossil species and a living species should be different entities.

5.3 OTHER TYPES OF INTRASPECIFIC VARIATION SHOWN BY LIVING COCCOLITHOPHORES In addition to the special problems already discussed a considerable amount of extra variation is shown by living species. As described in the previous chapter coccoliths are produced by a developmental process, and slight variations in degree of development at the various stages can produce quite significant changes in final form. Observations on coccospheres are particularly valuable since any variation between the coccoliths on a single coccosphere has to be intraspecific, in this respect the composite nature of haptophyte skeletons is an unusual advantage for taxonomic study of the group, rather than a special problem.

5.3.1 Structural variation related to element development. In Gephyrocapsa the distinctive central area bridge is almost certainly developed during the final stages of coccolith growth. Termination of development prior to this stage seems sometimes to occur as witnessed by the occurrence of coccospheres of Gephyrocapsa in which some specimens lack bridges (e.g. OKADA & McINTYRE 1977, NISHIDA 1979a, HEIMDAL & GAARDER 1981). Analogous examples include: A. Helicosphaera cartezi central area development: The helicoliths of this species show variation from forms with two large pores, via specimens with two small pores, to ones with only a single small pore. Illustrated coccospheres (e.g. in BORSETTI & CATI 1972, NISHIDA 1979a) typically show a portion of this total variation. In addition somewhat more random variation can occur (Fig.l5/A), possibly a reflection of the low regularity of blanket element nucleation. Similar variation in degree of development of blanket elements is shown by Pontosphaera, and

65 the normal coccoliths of Scyphosphaera (Fig.l5/B). B. Presence of a bridge in Coccolithus pelagicus: As in Gephyrocapsa the central area of Coccolithus pelagicus can be either open, or spanned by a bridge. Both forms are occasionally present on single coccospheres, and MANTON & LEEDALE (1969) observed specimens with and without bridges developing in a single cultural isolate. C. Rim development in Umbillcoaphaeza sibogae: Two varieties of U.sibogae occur (Fig.l5/E-F, taxonomy), with much greater development of the distal shield in U.sihogae var. foliosa.

5.3.2 Variation in degree of calcification. The rim elements in Bmiliania huxleyi are characteristically I-shaped, with slits between them. Variation in this feature is, however, very well documented, with discussions in WATABE & WILBUR (1966), McINTYRE & BE (1967), and BURNS (1977), and further illustration in most other papers on living nannoplankton. There is a continuum from specimens with virtually no slitting, via specimens with slitting in the distal shield only, to specimens with slitting in both shields. This trend is accompanied by a general reduction in degree of calcification. This variation is, at least in part, temperature related. Although all assemblages show a range of morphotypes, there is a shift toward higher calcification in assemblages: from lower latitudes (McINTYRE & BE 1967, BURNS 1977); produced during experimental growth in cold water (WATABE & WILBUR 1966); and in winter plankton samples (WINTER et al 1978). Analogous variation has been illustrated in Gephyrocapsa (BURNS 1977) although without apparent temperature control. These changes are not accompanied by consistent change in element length so perhaps this is a separate effect from element development as discussed above. That is element development may be related to one set of controls, such as rate of coccolith production, and calcification to another, such as supply. Alternatively it may be that element development in these species occurs in two stages the first being growth of skeletal elements and the second being further deposition of calcite on these elements.

5.3.3 Aberrant coccoliths A very common observation, is that with time an isolated strain loses its ability to produce coccoliths, and that there is an intermediate stage during which weakly calcified and malformed coccoliths are frequently formed (MANTON & LEEDALE 1969, PAASCHE & KLAVENESS 1970). Similar aberrant coccoliths have also been observed in studies of wild material. Examples have been illustrated by OKADA & HONJO (1975), WINTER et al (1978), NISHIDA (1979a).

66 FIGURE 15 - EXAMPLES OF INTRASPECIFIC VARIATION IN LIVING NANNOPLANKTON. Drawings of coccospheres and coccoliths of various species. Sources BORSETTI & CATI 1972, NISHIDA 1979a, HEIMDAL & GAARDER 1981. Scale bars one micron. A. Helicosphaera carteri coccosphere, showing: variable central areas; increase in coccolith size toward the flagellar opening; and expanded wings on the coccoliths around the flagellar opening. B. Scyphosphaera apsteinii, showing dimorphism, with elevated and normal coccoliths, also variable development of the central areas in the normal coccoliths. C. Syracosphaera pulchra, showing mild dimorphism, with spines developed on the circum-flagella coccoliths. The species also has an outer layer of coccoliths of radically different morphology, lost here (see Fig.6/F). D. Emiliania huxleyi coccolith with strongly calcified wall cycle (compare H,I). E. Umbilicosphaera sibogae var. foliosa. F. U.sibogae var. sibogae, the coccoliths of these varieties differ primarily in distal shield size, proximal shields are of similar size (see also taxonomy). McINTYRE & BE (1967) suggested U.s.sibogae was a warm water form. G. Emiliania huxleyi subsp. corona, variant of E.huxleyi with wall cycle extended distally. H. E.huxleyi, cold water heavily calcified variety. I. E.huxleyi, normal variety. J & K. Neosphaera coccolithomorpha, coccospheres with small and large varieties of coccoliths, variation is primarily in central area, rim width is more constant (about 1.5-2.5 microns). Intermediates occur and both varieties can occur on single coccospheres; according to McINTYRE & BE (1967) the smaller form is indicative of colder waters. 67 5.3.4 Size variation. Unfortunately there has been little work on size variation in living species. It appears, however, that on single cells and within cultural isolates the size of coccoliths (particularly placoliths), and scales is rather stable. A 20-30% size variation seems typical judging from illustrations and text discussions. Broader variation is evident, however, in wild populations, for instance E.huxleyi varies in size at least from 1.5 to 4 microns. Little more is to known about the causes of size variation. There do not, however, seem to be any ontogenetic causes for size variation, neither reproduction nor cell growth need be accompanied by coccolith size change. Ecologically controlled size variation has been suggested for Neosphaeza coccolithomozpha {McINTYRE & BE 1967), but not documented in detail. Possibly genetic control is dominant, which would be consistent with the biostratigraphic utility of size. In terms of coccolith development size variation can be caused by variation of proto-coccolith ring diameter, or by variation in element length. Cases can be seen of various combinations of these possibilities. Thus in U.sibogae size variation appears to be primarily a result of element length variation, in the larger variety U.s.foliosa the central area is if anything narrower than in the smaller variety U.s.sibogae. In Neospbaeza coccolithomozpha by contrast rim width is constant (c.2microns) whereas central area width varies sharply (1 to 4 microns) (e.g. McINTYRE & BE 1967, OKADA & McINTYRE 1977, NISHIDA 1979a, Fig.l5/J-K). More generally both factors vary approximately in harmony, as in the Reticulofenestza pseudoumbilica group (Plate.3). In each case the effect on final morphology can be considerable - even when proto-coccolith ring and rim dimensions vary together the ray width tends to remain constant and so the number of rays varies. Hence small and large coccoliths of one species almost always differ from each other in structure as well as in size.

5.4 DISCUSSION - INTRASPECIFIC VARIATION AND NANNOFOSSIL TAXONOMY 5.4.1 Conventional species concepts. As the preceding sections have shown, polyphase life-cycles, and strong polymorphism, which have been considered the primary problems for nannofossil taxonomy, are unlikely to be sufficiently widespread to cause palaeontologists more than interesting marginal problems. By contrast the less spectacular mechanisms for producing intraspecific variation have generally been discounted, but are likely to be highly significant. These mechanisms include: the milder types of

68 polymorphism, related to position on the cell; variable coccolith development, which particularly affects central structures; variable degree of calcification, which can strongly affect the rim; and size variation, which can also affect the general morphology of the coccolith. The variation these mechanisms produce is in all cases gradational. So there is continuous variation between the end-members, even though there may be a radical difference in form between these end-members. As a result one might expect intraspecific variation to cause nannofossil workers relatively few problems. However, as taxonomists dealing with other fossil groups have found, there are numerous ways to mistakenly split a continuum, including particularly, basing species on isolated specimens, or on small populations. Since nannofossil workers frequently have thousands of specimens this is not always a problem, but the extremely small size of nannofossils does pose another problem. Unlike most fossils, including planktonic foraminifera, coccoliths cannot be manoeuvred, manipulated, or systematically sorted. Hence specimens can only be closely compared when they fortuitously fall near one another on a slide. This makes it rather difficult to estimate the variability of species, and to spot intergradational morphotypes. Ways around the problems exist, by photographing, measuring, or even drawing, large numbers of specimens. These approaches are, however, time consuming, and consequently unpopular. Furthermore for biostratigraphic work it is not convenient to identify every specimen. Instead assemblages are scanned to establish which species are present, and their approximate abundances. With this technique it is entirely possible to ignore ambiguous specimens - including crucially those which are intermediate in morphology between two "species". Hence artificial species definitions are at least possible, if not attractive. The species concepts of many workers thus make little attempt to distinguish between true species and intraspecific variants, but simply strive to recognise distinctive morphotypes with discrete stratigraphical ranges. This has been a successful approach in terms of biostratigraphy, but has almost certainly lead to the artificial division of many species. For example Helicosphaera sellii is separated from H.carteri on the basis of rather tenuous variation in central area development, and modern coccospheres with both "species" have been illustrated (NISHIDA 1979a), but H.sellii has a discrete stratigraphical range, it first occurs in the Early Pliocene and becomes virtually extinct in the Early Pleistocene, an event which has been used for biostratigraphy. H.carteri by contrast occurs from the Early Miocene to the present. Similarly as described above Pseudoemiliania lacunosa is

69 linked via a continuous series of intermediates to undistinguished small reticulofenestrids so it is most unlikely that the two are distinct species. However the P.lacunosa morphotype is confined to the Late Pliocene to Middle Pleistocene, whereas small reticulofenestrids indistinguishable from those which it grades into occur throughout the Neogene. Other such examples occur in virtually every Neogene nannofossil group, as discussed in the taxonomy section. This is something of a paradox, the biostratigraphic distribution suggests that fine subdivision is meaningful, whereas the morphological intergradation suggests that such subdivision is spurious. Explanation of this paradox requires a brief consideration of evolutionary models.

5.4.2 Relationship of species concepts to evolutionary models. In Figure 15 the two well known evolutionary models of phyletic gradualism and punctuated equilibrium are compared. As indicated, both models suggest that the operational concept of species as distinctive morphotypes with discrete ranges should be valid. Of the two punctuated equilibrium seems to fit the observed patterns better since phyletic gradualism predicts that related species should form lineages with individual species successively replacing one another. The more generally observed pattern is for closely related species to overlap greatly in range, as predicted by the punctuated equilibrium model. Diagrams illustrating nannofossil evolution have thus tended to suggest punctuated equilibrium as the dominant process (e.g. figures of HAQ 1973, ROMEIN 1979, THEODORIDIS 1984, PERCH-NIELSEN 1985a,b). Also the biostratigraphical concept of abrupt first and last occurrences is better predicted by the punctuated model. In effect this model is the implicit basis, and theoretical justification, of conventional nannofossil taxonomy. However, it completely fails to predict the occurrence of intergradational taxa with discrete ranges. Instead a third pattern, shown in Figure 16/C, seems likely. This is essentially phyletic gradualism, but with intraspecific variation made much more important; and with the assumption that such variation can occur on a systematic stratigraphical basis. This type of pattern will cause genuine taxonomic problems, and requires care in identifying species. If broad species definitions are adopted then the majority of the biostratigraphic information is lost. Conversely if species are narrowly defined then they will bear little relation to the true living species, and will intergrade hopelessly. My impression from work on Neogene nannofossils is that this type of pattern does hold, and that the species definitions which have conventionally been adopted fall between the two extremes. Thus they are too narrow to reflect the true species, but still too broad to

70 extract the maximum of biostratigraphic resolution.

5.5 CONCLUSIONS 1. Polymorphism and polyphase life cycles are unlikely to be significant problems for nannofossil taxonomy, and should not be used as reasons for adopting an artificial taxonomy.

2. "Polyspecific coccospheres" are probably artefacts, and of no taxonomic significance.

3. Most living species do display significant degrees of intraspecific variation, this typically involves variation in degree of development of the elements or in size or shape of the proto-coccolith ring. Causes for variation include adaptation to aspherical cells, circum-flagellar specialisation, ecotypic variation, and probably genotypic variation.

4. Nannofossil taxonomy has been produced primarily to serve biostratigraphy and has tended to ignore the problems caused by intraspecific variation.

5. Nannofossil taxonomy needs to be adjusted taking intraspecific variation into account if it is to approximate to a natural taxonomy. For evolutionary studies and palaeoecological work this is essential. It requires lumping of existing species.

6. Intraspecific variation in nannofossils can be biostratigraphically useful, hence intraspecific taxa need to be investigated and defined. This may require the recognition of additional more finely defined taxa than in the conventional "species” only system.

71 A. CLASSIC PHYLETIC GRADUALISM. Boxes indicate one possible subdivision of the lineages into species, of discrete morphology and range. This model suggests that closely related species should not have significantly overlapping ranges and so does not seem immediately applicable to nannofossil taxonomy.

B. PUNCTUATED EQUILIBRIUM. Model of ELDREDGE & GOULD (1972), based on direct reading of stratigraphic distribution of species, with the interpretation that evolution occurs primarily in abrupt events within isolated populations. It appears to predict well the documented stratigraphic distribution patterns of nannofossil species, but implies that they are genuinely distinct.

V

C. PHYLETIC GRADUALISM WITH SIGNIFICANT INTRASPECIFIC VARIATION. Boxes indicate possible taxonomic subdivisions of the evolving plexus. Since these are intergradational they cannot all be considered separate species, but rather subspecies or varieties. In this case branching of the plexus means that two or three species can usefully be distinguished. This pattern is applicable to much of nannofossil taxonomy.

FIGURE 16 - SPECIES CONCEPTS AND EVOLUTIONARY PATTERNS. Three diagrams illustrating the relationship between possible evolutionary patterns and taxonomy. In each case the vertical axis is time and the horizontal axis morphology.

72 IP ART B B I O S TRi^TI O R A P H Y

73 CHAPTER 6 - NEOGENE NANNOFOSSIL BIOSTRATIGRAPHY

INTRODUCTION Neogene nannofossil biostratigraphy is well established and has been comprehensively discussed by many authors, e.g. ELLIS (1982), JIANG & GARTNER (1984), THEODORIDIS (1984), PERCH-NIELSEN (1985b), MARTINI & MULLER (1986). Consequently I give only a brief outline here, and discussion of certain problems, followed by a description of the approach I adopted. Subsequent chapters discuss in detail biostratigraphic use of reticulofenestrid size variation in the Neogene, and biostratigraphy of the studied material from Indian Ocean DSDP sites and the Makran of Pakistan.

6.1 NANNOFOSSIL ZONATION SCHEMES Figure 17 shows four separate nannofossil zonations that have been proposed for parts of the Late Cenozoic. The two main schemes are the Standard Calcareous Nannofossil Zonation of MARTINI & WORSLEY (1970), MARTINI (1971) and the Low-Latitude Coccolith Zonation of BUKRY (1973a, 1975, 1981), OKADA & BUKRY (1980). Both these zonations cover the entire Tertiary and, despite their names, are based on virtually the same succession of events. This reflects their similar historical origins, and the fact that they both primarily use large species and non-coccoliths. The Pleistocene presents special problems for biostratigraphy, mainly due to the absence of non-coccolith nannofossils. Effective zonation of this period is not a feature of the main schemes but has been developed subsequently, following the work of GARTNER (1977). The Integrated Miocene Zonation of THEODORIDIS (1984) is a significant attempt to incorporate information from other groups, arising from work on Mediterranean sections, where many of the conventional open-ocean markers are missing. In particular, Helicosphaera species are extensively used. Since the scheme incorporates most of the events of the older schemes there is no problem in correlating them, despite renaming and re-organisation of the zones. The succession of events from these zonations, together with a few others, is shown in Figure 18. In total there are about 45 events for the Neogene alone, compared with only 18 zones in the Standard Zonation. Many of the additional events are, however, based on species which are only sporadically present (eg most Helicosphaera spp.), or are gradual rather than abrupt events (eg first occurrences of D.pentaradiatus and G.rotula, last occurrence of T.rugosus). As a result the relative

74 MARTINI l HORSLEY 1970 OKADA t BUKRY 1980 GARTNER 1977, RAFFI 1. RIO 1979 E.huxleyi acme A a E . h x 21 E.huxleyi 15 E.huxleyi E . h u x l e y i A E . h u x A D . q n q A E.hxl 20 G.oceanica 14b C.cristatus G . o c e a n i c a xP.lac v P . l a c y P . l a c 14 G.oceanica 14a E.ovata P . l a c u n o s a 'lG. o c n A G . o c n small Gephyrocapsa yH.sel H . s e l l i i (C.mac 13b G.caribbeanica P.lacunosa C.macintyrei X G . c r b A G.c r b 19 P.lacunosa 13 C.doronicoides 13a E.annula C.doronicoides v D . b r w v D ,br w >(D.brw 18 D.brouweri 12d C.macintyrei C.pelagicus D . b r o u w e r i (R.rot R . r o t a r i a A R . rot 9b A.primus A . p r i m u s A A . p r * A A . s p p C.pelagicus r M . c n v 11 D.quinqueramus 9 D.quinqueramus 9a D.berggrenii G . r o t u l a B XD .q n q A D . q n q '‘'D.qnq 3b D.neorectus A D . nrc G . r o t u l a A •‘D.pnt 10 D.calcaris 8 D.neohamatus Sa D.bellus E.pentaradiatus ( D . h a m ■(D.hara •

FIGURE 17 - NEOGENE AND QUATERNARY NANNOFOSSIL ZONATIONS. Comparison of four calcareous nannofossil zonations proposed for the Late Tertiary. 1. Standard Calcareous Nannofossil Zonatiori of MARTINI & HORSLEY (1970), MARTINI (1971). Zone numbers conventionally prefixed NN (Neogene Nannoplankton). 2. Low-Latitude Coccolith Zonation of BUKRY (1973a, 1975, 1981) with zonal numbers of OKADA & BUKRY (1980), prefix CN (Coccolith N e o g e n e ) . 3. Pleistocene zonation of GARTNER (1977), with slight alterations after RAFFI Si RIO (1979). 4. Integrated Miocene Zonation of THEGDORIDIS (1984). In each column zones are given on the left, separated by solid lines, and subzones on the right separated by dotted lines. The zonal events are given in the margins. Species names are abbreviated and only the main events are noted, the other events and full names are given in Figure 18. In cases of variable taxonomic assignment (eg Eu-diseoasterldiscoaster, Aiaurolithus/Ceratolithus) the author's assignment is used for zone names, but a standardised system for events. A First occurrence; v Last occurrence; Aa acme. QLI60CENE I MIOCENE I PLIOCENE I QUATERNARY v---- A *NN17 v *NN 18 v A A *NN14 v ^NP 25 A v A A \ v A v *NN 3 v A *NP 24 *NP 22 ■i NN19 NN 20 ______NN 21 NN15 NN16 --- NN12 --- ______NN13 ______NN 11 --- NN S NN 10 --- NN 8 ______NN 7 NN 5 --- ______NN 6 ______NN 4 --- NN 2 --- NN 1 --- ______--- NP 23 o r g a n i s e d h o r i z o n t a l l y t o s h o w t h e i r u s e i n t h e n a n n o f o s s i lz o n a t i o n s o f M a r t i n iI t H o r s l e y a n dBu o f k r y , IUE 8 NNOOSL EVENTS Listing NANNOFOSSIL - 18 principal of FIGURE Oligocene Recent to nannofossil events used in biostratigraphy. Events are g i v e n i ntin h t e h dnu e e x i r b ifo e g r h s r t th h a e n s d e c z o o l n u e a s n a . r e g i vEv e n e . n t s f r o a o t h e r s c h e a e s a n d s o a e a d d i t i o n a le v e n t s a r e P - a d dT i t - i o e n v a e l nGe t v - s e f n e r t v s o e a c n o t t a s h a f e o r n M l o i y a o c u t e s h n e e e d Pf z l o o e n r i a s tl t i o o o w c n e l n o a e t f i z T t o H u n E d a 0 e t D i 0 b o R i n I o D o s I t f S r ( 6 a 1 A t R 9 i T 8 g N 4 r E a R ) p ( . h 1 y 9 ,f 7 r 7 o ) a . P E R C H - N I E L S E N ( 1 9 8 5 b ) . A f i r s t o c c u r r e n c e ; ATN 17 OAA t UR 18 OHR EVENTS OTHER 1980 It BUKRY OKADA 1971 MARTINI E.huxleyi- P.lacunosa R.pseudouabilica- D.pentaradiatus-- D.brouveri D.surculus . rugosus C. A.tricorniculatus D.asyaaetricus- - D.quinqueranus quinqueraaus D. D.hanatus C . c o a l i t u s . kugleri D. S.heteroaorphosus D.haaatus T. carinatus S. beleBnos- A------S.heteroaorphosus' - - - - - H.aapliaperta D.druggii H.recta R.uabilica------S.ciperoensis S.distentus------C.foraosus ------1 ---- l a s t o c c u r r e n c e ; A a s t a r t o f a c a e ;* a e n d o f a c a e . ------A— ---- S.ciperoensis,*------D.bisectus— 'CN I a i CNI4b CN15 CNI3b CNI4a NI2cCN 12dCN CN NI2bCN CN NlOdCN NUa CN CN10a CN10c . CN 9a CN 9b CN10b * CN 7b CN 8b CN 8a CN 6 ___ CN 7a CN 5b CN 5a CN 4 CN 2 CN 3 CN la CNlb CN1c CP19a CP19b a CP16c CP17 CP18 ------13a 12a lib O.kugleri------*a D.deflandrei A- ■*- O.serratus------R.hillae S.beleanos C.abisectus S.distentus G.oceanica- - - G.caribbeanica S p h e n o l i t h s D.taaalis, asyaaetricus A.priaus C.acutus . c t s T.rugosusC.acutus, D.taaalis, Aa D.asyaaetricus- D . n e o r e c t u s + DD.berggrenii, . l D.neorectus o e b l i c h i A.priaus------i — C.calyculus ------* a s a a l lG e p h y r o c a p s a - A a E . h u x l e y i Aa D.b.triradiatus A- large • Gephyrocapsa j - C . a a c i n t y r e i <- H.sellii---Aa saall Gephyrocapsa A- P.lacunosa, Gephyrocapsa r_ A- A- M.convallis A - D . p e n t a r a d i a t u s ,* - D . p ’ p e n t - T y- H.valbersdorfensisA- D.calcaris, bellus *a C.Tloridanus -r H.waltrans A- H.orientalis A- H.stalis- A- A- D.signus A- C.abisectus, H.recta ■i~ A- H.aapliapertaA- G.rotula A- P.enorais - Ir c r u P y-I.recurvus D.prepentaradiatus D.exilis Z.bijugatus D . v a r i a b i l i s ' R.rotaria M.convallisA.tricorniculatus, delicatus D.neohanatus, D.loeblichii— P R.rotaria C.floridanus C.aiopelagicus H.perch-nielsenae ------.... ------... -- — T ■— T T P T T T T T T T G G 6 G G G P P P T T T T P T T P P P

position o£ these events within the overall sequence of events is liable to vary between sections, with varying preservation, or palaeoenvironment. A similar problem is presented by events based on gradualistic transitions within evolving lineages (eg first occurrences of D.exilis, D.calcar is, and C.calyculus). Such events can be used successfully; for instance in the Oligocene the principal zonal marker events are provided by arbitrary subdivisions of the continuously evolving Sphenolithus predistentus - distentus - ciperoensis lineage. However, as this example showed, the use of gradualistic transitions depends on careful research and the adoption of unambiguous, if arbitrary, taxonomic criteria. This is a consequence of these lineages showing gradualistic evolution of the type discussed above (Section 5.4), with the morphotypes occurring together over long intervals and intergrading. For most Neogene lineages detailed work is still needed, and so only limited biostratigraphic use can be made of them. The zonation of Martini and Worsley was based mainly on distinctive species with abrupt first or last occurrences and so avoided these problems. Of the existing schemes it needs the fewest modifications to make it acceptable, is the most straightforward and best known scheme, and provides a widely applicable frame of reference. Consequently I have used it throughout this study. The other events have been used as appropriate, to aid in zonal assignments, and for supplementary information. The necessary modifications to the Standard Zonation are; (1) Combination of zones NN13 and 14 as zone NN13/14, this is common practice since the first occurrence of D.asymmetricus is a very unsatisfactory event. (2) Subdivision of NN11 into early (eNNll) and late (1NN11) parts, using the first occurrence of Amaurolithus. This event is widely used, if not as a formal component of the Standard Zonation. (3) Use of the last occurrence of S.ciperoensis rather than of H.recta for the NP25-NN1 boundary, as recommended by MARTINI (1981). Also I have generally used index numbers rather than zonal names, since many names have been used in two or more different senses in the various schemes.

6.2 CHRONOMETRIC AGE OF NANNOFOSSIL EVENTS For many purposes it is useful to have estimates of the durations of individual zones, and of the ages of their boundary events. A very considerable amount of work has been done on providing such estimates, as part of the general process of integrating Tertiary stratigraphy. Two major syntheses have recently been produced, with extensive bibliographies, BERGGREN et al (1985) and HAQ et al (in press).

77 Biostratigraphical events have mainly been dated by an indirect process involving a number of steps. The first of these steps is correlation of biostratigraphic events to the magnetostratigraphic sequence, using cores or sections for which both magnetostratigraphy and biostratigraphy are available. The next step is calibration of the magnetostratigraphic sequence against estimates of the ages of chronostratigraphic boundaries. The boundary ages in turn are based on statistical treatment of geochronological dates from around the boundary age. Obviously errors and uncertainties are inevitable at each stage of the process, not least the initial biostratigraphy. To a certain extent, however, systematic errors have been eliminated by checking from other sources, such as dating of tuffs within marine sequences. As an indirect assessment of the reliability of the resultant ages I have compared, in Figure 19, the estimated ages of the zonal boundaries from a number of papers, and also MARTINI'S (1971) estimation of the relative duration of the nannofossil zones. This diagram suggests that uncertainties on the order of .5-2 Ma still exist in the ages of events, and of around 30% in the duration of zones. This variation is considerable, and shows that if data from nannofossil biostratigraphy is presented in terms of "absolute" ages rather than nannofossil zones then it is more ambiguous and liable to revision. Similarly for discussions of nannofossil evolution and nannofloral development although the chronometric time scale provides an invaluable scaling, the essential frame of reference must be the nannofossil zonation. The direct use of chronometric ages (eg HAQ 1980, PUJOS 1987) is still premature. Nonetheless changes in estimates have not been too drastic. In particular events closely related to system boundaries (NN18/19, NN11/12, NP25/NN1) have almost constant ages. This reflects the emphasis on system boundary ages in the production and calibration of these schemes. The single most important change is the pronounced recalibration of Mid and Late Miocene ages in the schemes of BERGGREN et al (1985) and HAQ et al (in press). This is the result of a major revision in the palaeomagnetic time scale occasioned by new magnetobiostratigraphic data (BERGGREN et al 1985). The wide variation in estimates of the age of the NN10-11 boundary is, in part at least, an associated effect. Finally it is noteworthy that the most recent scheme, that of Haq et al. is also the one that most closely follows the original estimates of the relative durations of the zones of Martini. I have used this set of age estimates in all relevant diagrams.

78 HART 71 P-H 72 BERG+ 74 RVAN+ 74 BUKRV 75 HflQ 84 BERG+ 85 HAQ+ 87

PLEISTOCENE

PLIOCENE

LATE MIOCENE

MIDDLE MIOCENE

EARLY MIOCENE

FIGURE 19 - ESTIMATES OF THE AGES OF NANNOFOSSIL DATUMS. Graph showing the estimates froi several papers of the ages of the nannofossil datuas separating the Standard Nannofossil Zones of MARTINI It HORSLEY (1970). Each sub-horizontal line represents one datui. The vertical scale is in lillion years. Zonal index nuabers are given on the right-hand side of the graph. N.8. (1) Abbreviations: MART 71 - MARTINI 1971; P-N 72 - PERCH-NIELSEN 1972; BERG+ 74 - BERGGREN It VAN C0UVERIN6 1974; RYAN+ 74 - RYAN et al 1974; BER6+ 85 - BERGGREN et al 1985; HAQ+ 87 - HAQ et al, in press. (2) MARTINI (1971) gives estimated relative lengths of the zones rather than absolute ages, these have been scaled by assigning an age of 24Ma to the 01igo-Miocene boundary, as in conteaporary papers. (3) The dotted line within NNU represents first occurrence of fit auralithus (eNNll-lNNll boundary). (4) The NN13-NN14 boundary has been oiitted. (5) PERCH-NIELSEN (1972) does not give an estiaate of the NN1/NN2 boundary age, so the saae value has been used as in Martini (1971).

79 6.3 CORRELATION OP BIOSTRATIGRAPHY AND CHRONOSTRATIGRAPHY. Correlation of biostratigraphic zones to chronostratigraphic units is a more straightforward process. Ideally it only a requires a study of the biostratigraphy of stratotype sections. In practice however most stage stratotypes, and particularly their boundaries have been poorly defined, at least until recently. Also most type sections are in shallow marine sequences with rather poor nannofloras. Hence correlation of the two systems is not in fact very precise and is liable to revision. This is particularly true at the stage level; I have avoided using stages for this reason, and also because they are unfamiliar to British geologists. Instead I have used series and sub-series, correlations at this level have been reasonably stable. The most important exception is the Middle - Late Miocene boundary. This was initially placed by Martini near the top of zone NN9 but more recent syntheses (eg BERGGREN et al 1985, HAQ et al in press) place it within zone NN8. As a consequence zone NN9 has changed from being mainly Middle Miocene to entirely Late Miocene in age, the usage adopted here.

6.4 FURTHER DEVELOPMENT OF BIOSTRATIGRAPHY I have not tried to find further isolated events to fit into a zonal sequence but rather concentrated on the evolutionary development of the separate family level groups, each of which can be used separately to provide information on the age of samples or sequences. I have adopted this approach partly because the DSDP sample set I used was not really suitable for refinement of the high resolution zonal schemes, but also in part for a number of inter-related positive reasons. 1. The first and last occurrences of many common species are too diachronous to use as biostratigraphic events, but nonetheless these species do have usefully restricted ranges. This applies for instance to the first occurrences of Discoaster surculus, D.pentaradiatus, Coccolithus pelagicus var. pontus, and Calcidiscus leptoporus. 2. In poorly preserved material the main marker species are often missing or rare and so it is important to be able to extract all the biostratigraphic information possible from those species that are present. 3. Many species with usefully restricted ranges are either, too rare, or too sporadically distributed, to be used in a standard zonal scheme. This applies for instance to Discoaster altus and Discoaster exilis var. petaliformis. Both these species are highly distinctive, and have short, if still poorly defined, stratigraphical ranges, but they are present in by no means all sections. On a broader scale it

80 also applies to the Helicosphaeraceae as a group, their distribution is very sporadic, almost certainly as a result of ecological control. Consequently, although they evolve rapidly during the Neogene, events based on them are unreliable as components of the standard zonal scheme. 4. Gradualistic evolutionary developments, as opposed to cryptic first and last occurrences, need to be studied in terms of lineage development. 5. To a considerable extent preservational and ecological factors act uniformly on family level groups thus the presence of one member of a group can cast extra significance on the absence of another. For instance the absence of Triguetorhabdulus carinatus from Lower Miocene sediments could be due to a number of factors, but if T.ailowii is present then it is most likely that the sediments were deposited after the last occurrence of T.carinatus (end zone NN2). The group I studied in most detail were the reticulofenestrid coccoliths, and these are discussed in detail below. For other groups discussions are given separately in the systematics section, since the taxonomy and biostratigraphic use of species are closely related. For most groups a range chart - evolutionary scheme is given, based on a synthesis of my own and published data.

81 CHAPTER 7 - RETICULOFENESTRID SIZE VARIATION

INTRODUCTION The single most abundant species during the Middle Miocene to Early Pliocene interval is Reticulofenestra pseudoumbi2ica, which usually constitutes over half of even well preserved assemblages. During this period despite the structural complexity described above (Chapter 4) they show rather little variation in form. The wall cycle varies in degree of development, which affects the width of the central area, but there is no specialised development of parts of it. The coccoliths do, however, vary strongly in size; they range from about two to twelve microns long. This is nearly an order of magnitude variation, and more in terms of area or volume. The combination of high abundance, regular morphology and wide size variation makes biometric investigation attractive. Moreover it was readily apparent from reconnaissance work on the DSDP material, and the Makran material that their size range varied through the Neogene. In particular it was clear that the larger forms, R.pseudoumbi1ica sensu stricto, were not universally present from their first occurrence (Early Miocene) to their extinction (NN15). For these reasons I systematically recorded size distribution of reticulofenestrids as part of my routine study of samples.

7.1 METHODOLOGY All work was carried out using conventional light microscopy, and so size was measured using an eyepiece graticule. At the highest magnification (x2000) the graticule divisions corresponded to 0.64 microns on the specimen, which provided the limit of resolution. This is adequate for recording gross size variation, given the wide total size variation. It is inadequate, however, for investigating variations in form, since other parameters, such as coccolith width, central area length or rim width correlate strongly with length. Electron microscopy, or image analysis equipment, would be needed to quantitatively investigate the residual variation in these independent of length. Hence, and in order to make measuring fast enough for routine work, only the single parameter of coccolith length was measured. For each sample I measured the length of about one hundred randomly selected specimens. My technique was to measure about twenty specimens in one area of a slide, moving from one specimen to the nearest unmeasured specimen, then moving to another area of the slide. In all cases smear slides were used. For this type of work slides need to be

82 fairly thinly prepared (less than about forty specimens per field of view). In some samples there is a skewed distribution with small specimens predominating, in these cases I did a further count of large specimens only. This process of separately counting different size fractions makes data reduction slightly more complex but allows rapid investigation of total size variation. For this work I adopted a broad definition of Reticulofenestra. I included specimens with open and closed central areas, since specimens with closed central areas ("Dictyococcites") were rare and probably mainly of diagenetic origin in my material. At high latitudes these forms are undoubtedly more important, and primary in origin (BURNS 1975, BACKMAN 1980, 1984, HAQ 1981). For comparative purposes I continued the study into the late Pliocene including all members of the Reticulofenestra doronicoides - Pseudoemiliania lacunosa plexus, but not Gephyrocapsa spp.

7.2 DATA PRESENTATION The data is presented in Figures 20-22, with minimum possible manipulation. Each diagram gives the data from a single site as a series of compressed size / frequency histograms for individual samples. On these histograms the horizontal scale is coccolith length and the vertical scale relative frequency. Both scales are linear, and the size intervals are those recorded (ie increments of 0.64 microns). Only two adjustments have been made to the data, both primarily to cope with the occasional very abundant small specimens. First, abundances have been calculated relative to the abundance of specimens over two microns, rather than of all specimens. Thus the presence of abundant small specimens in a sample does not depress the indicated abundances of the other specimens. Second, any relative abundances over 0.4 are rounded down to 0.4, this prevents histograms overlapping. The vertical positioning of the histograms within the diagrams is based on age as determined by conventional biostratigraphy (details for each site in Chapter 8), with scaling from the time scale of HAQ et al (in press). Nannofossil zones, chronometric ages and sample numbers are given beside the vertical axes of the diagrams.

7.3 RESULTS 7.3.1 Main Results, DSDP Sites 219, 223, 231, and 242. The results from the four sites with reasonably continuous successions are given in Figures 20-21. These are discussed together since the results are very similar in each case; even though the sites

83 are separated by thousands o£ kilometres (Fig.28, p.110). They can be subdivided into four intervals. A. NN6 to early NN10: The size distribution patterns in all the samples from this interval show similar features. They are unimodal with a broad spread of sizes, typically 2-10 microns. Despite the general similarity of all the size distributions, a general trend can also be detected of size increase through this interval. This is shown particularly by variation in the modal length. Thus although all samples contain significant numbers of specimens over 5 microns long these only constitute the majority of the assemblages in the later samples. B. Late NN10 to early NN11 - Small Reticulofenestra interval: Within zone NN10 there is a dramatic decrease in maximum, minimum, and modal size with coccoliths over five microns long virtually disappearing, and smaller specimens increasing in abundance (in both relative and absolute terms). This interval is the most distinctive part of the pattern, for convenience I have termed it the "small Reticulofenestra interval”, by analogy with the Pleistocene small Gephyrocapsa interval (GARTNER 1977, RIO 1982). C. Late NNll to NN15: During this interval the reticulofenestrids again show a wide range of size variation, as in the first interval. The size distribution patterns, however, are strongly skewed toward the smaller specimens, so the bulk of specimens are always less than 5microns long, and large specimens (>c.7microns) are always rare. Within the smaller size range several samples show evidence of bimodality with a frequency minimum around 3-3.5microns. It is thus possible to describe the reticulofenestrid assemblage in this interval as being formed of three components: very small coccoliths (<3microns): small coccoliths (3-5microns); and larger coccoliths (>5microns). Apart from size the coccoliths of the three groups are, however, identical. There is some evidence of size increase through this interval, particularly in maximum length but owing to the low relative abundance of these larger specimens a separate study would really be necessary to test this. D. NN16 to NN18, Late Pliocene: The larger specimens disappear entirely above NN15 although as they are rare in any case this does not have a very marked effect on the appearance of the assemblages. In addition the subsequent assemblages are almost entirely unimodal, with the dominant size, 2.5-4 microns approximately corresponding to the frequency minimum in the previous bimodal distributions. Also, of course, larger specimens show Pseudoemiliania lacunosa morphology, at least in well preserved material.

84 SITE 219 (ARABIAN SEA) - RETICULOFENESTRID SIZE UARIATION SITE 223 (ARABIAN SEA) - RETICULOFEHESTRIO SIZE UARIATION

FIGURE 20 - Reticulofenestrid size variation, Sites 219 and 223.

NN11

T 2 3 4 5 6 7 8 9 10 TT COCCOLITH LENGTH (MICRONS)

UERTICAL SCALE Abundance, relative to the total nunber of specimens nore.than tvo nicrons lono.

10.4 (0.05

COCCOLITH LENGTH (MICRONS) MIOCENE I PLIOCENE S I T23(G E 1 U LO FAD F ERE N- T ) I C O L O F E N E S T RSI I 0 ZUR E R I A T I O N -i 2 4 6 8 1 11 10 9 8 7 6 5 4 3 2 l ------1 ------OCLT LNT (MICRONS) LENGTH COCCOLITH 1 ------1 ------i ------1 ------1 ------1 ------1 ------i ------1—

MIOCENE 1 PLIOCENE S I T24(M E 2 O Z A M B I Q UCH E A N N ERE L- T ) I C U L 0 F E N E 5 T R ISI D Z UR E R I R T I O N

FIGURE 21 - Reticulofenestrid size variation, Sites 231 and 242. SITE 225 (RED SEfl) - RETICULOFENESTRID SIZE UfiRIflTIOH SITES 249 t 251A

I PLIOCENE FIGURE 22 Reticulofenestrid - size variation, Sites 225, 227, 249 and 251A. O cn ZD G-4 U G-4 M M f > > ZD cn tn tn cn 3 MIOCENE ! I I

COCCOLITH LENGTH (HICROHS) 7.3.2 Red Sea Sites, 225 and 227. The results from these sites (Fig.22) are of some interest since the largest Reticulofenestra specimens are missing, possibly an ecological effect related the restricted nature of the Red Sea at this time. Nonetheless the results are plainly comparable to those from the other sites and the last occurrence of specimens over five microns is still apparent.

7.3.3 South Vest Indian Ocean Sites, 249 and 251A. The data from these sites (Fig.22) is limited and of rather poor quality, in particular the biostratigraphy is not as good as for the other sites (Chapter 8), and preservation is rather worse. I did not want, however, to exclude data which was contradictory to the general model presented. The results from the NN6-NN9 interval (samples 251A-26-2 to 251A-15-1) are similar to those from the other sites. The small Reticulofenestra interval is arguably represented by samples 251A-13-6 to 249-13-3 but since there are some large specimens still present in these samples it is much less distinct than at the other sites. The three samples above these (251A-13-2 to 251A-13-5) have lower modal size than in the early samples, but again the presence of large specimens reduces this contrast. The youngest assemblages (Samples 251A-4-5 to 251A-10-CC) are similar to those of this age from the other sites. The poor development of the small Reticulofenestra zone may be an artefact related to the problems noted above. Alternatively it may be related to the higher latitude of these sites (35-40° as opposed to <25°).

7.4 COMPARABLE STUDIES The last occurrence of large Reticulofenestra at the end of the Early Pliocene is well established as a global event, it forms an important datum in all Pliocene nannofossil zonations. Beyond this the literature contains few explicit indications that Neogene reticulofenestrids show systematic size variation patterns. There are, however, several studies which can be used to investigate whether the patterns described above also hold outside the Western Indian Ocean. A. Other biometric studies: The only extensive biometric study I know of on Miocene reticulofenestrids is that of BACKMAN (1980). He did detailed work mainly on the, high latitude, N.Atlantic DSDP Site 116. His work was, however, primarily taxonomic and concentrated on a few samples. The limited data presented on size variation through time (his figs.5 and 24) could be interpreted as giving weak evidence for a late

88 Miocene small Reticulofenestra interval, but is not directly comparable with mine leg size data is given in terms of "placolith area”, this parameter is undefined but appears to equal 2-2.5xlength*l B. Quantitative studies using detailed taxonomy: From his biometric work BACKMAN (1980) proposed a rationalized subdivision of the reticulofenestrids based on size and central area opening (summarised in Fig.23/D, p.92). This scheme has been adopted by several authors, with only minor changes to the species definitions. Some of these have presented quantitative information on the species abundances which can be compared with my data. HAQ (1981) studied variations in nannofossil assemblages in 444 samples from 23 South Atlantic DSDP sites. The information is given in terms of principal components analysis so details are obscure, but major trends can be seen. The R.pseudoumbilica - R.haqii component shows a marked drop around lOMa (probably within NN10, he uses an undefined time scale), with a correlative increase in the D.minutus component (his figs.6 & 7). This could correspond to the start of the small Reticulofenestra interval. LOHMANN & CARLSON (1981) conducted a similar study of Pacific material, with arguably similar results, which they interpreted in terms of climatic control. Again use of multivariate analysis and an undefined time-scale makes data interpretation tenuous. PUJOS (1985c, 1987) recorded quantitatively variations in the abundance of reticulofenestrid species in samples from Central Pacific DSDP Sites 571-575, Leg 85. This is the most readily comparable study to mine and the results are very similar. There is a distinct small Reticulofenestra interval (R.minuta dominant, and R.pseudoumbilica absent) at all sites. Also the assemblages below are dominated by large to medium forms (R.pseudoumbilica, R.minutula, D.perplexus and D.hesslandii) whilst those above are characterised by small to medium forms (R.minuta, and R.minutula with rare R.pseudoumbilica). Unfortunately the relevant interval (NN9-eNNll) is represented by sediments which are stratigraphically condensed and in which the marker discoaster species are rather rare. As a result of these factors the timing of events is rather uncertain but the Reticulofenestra assemblage development is at least broadly correlative with that seen in the Indian Ocean. FLORES & SIERRO (1987) give similar information but from a restricted time period (within NN11), and so their study is of little use in this context. C. Routine studies: Events as marked as the small Reticulofenestra interval might be expected to be detected during routine studies, however, in most cases only one or two divisions of the Reticulofenestra group are made and even semi-quantitative abundance information is not

89 always given. Thus in the Indian Ocean DSDP reports the small Reticulofenestra interval can be discerned in the data given by BOUDREAUX (1974) for Site 223, but not in the data of ROTH (1974) for Site 231 or of MULLER (1974a) for Site 242. Of sites I did not study there is also evidence for it at Site 214 (GARTNER 1974), and possibly Site 220 (BOUDREAUX 1974). Outside the Indian Ocean the interval is discernible in the data of JIANG & GARTNER (1984, DSDP Leg 74, S.Atlantic), and more ambiguously MARTINI (1981, DSDP Leg 59, Central Pacific); but few of the reports I have checked contain suitable information from the relevant interval/ I have not attempted an exhaustive search.

7.5 DISCUSSION OP THB SIZE VARIATION TRENDS 7.5.1 Validity and extent. If the size variation trends had only been documented from one site then they could have been interpreted as an preservational artefact, or even a random effect. However, since three well separated sites independently show virtually the same pattern the most reasonable explanation is that these patterns represent primary variation in the nannofossil population within the study area, i.e. low latitude Western Indian Ocean. The evidence from published work, discussed above, is not conclusive but does suggest that the small Reticulo£enestra interval is also discernible in the Eastern Indian Ocean (Site 214), the Central Pacific and the South Atlantic. The contrast between the NN6-9 interval (large Reticulofenestza dominated) and the NN11-15 interval (small Reticulofenestra dominated) is harder to detect in published records, but is shown in the data of PUJOS (1985c). At high latitudes the record is complicated by poorer dating, and by the presence of forms with closed central areas, nonetheless since several authors have looked closely at reticulofenestrids from high latitude material it is unlikely that an event as obvious as the small Reticulofenestra interval would have gone unnoticed. In conclusion it seems quite possible, however surprisingly, that the size variation patterns documented here may hold more or less universally at low latitudes (c.0-30°), but it is likely that they are at least modified at higher latitudes.

7.5.2 Possible causes. The size variation patterns can be interpreted in a range of ways. The two extreme possibilities are, first, that they represent the effect of change in ecological factors such as climate on a genetically unchanging population. Second, that they reflect evolutionary change

90 without environmental change. It is not possible to decide conclusively between these, and the spectrum of intermediate possibilities, without information on a range of other topics including: how widespread these changes are, particularly at high latitudes; how coccolith size is regulated by reticulofenestrid coccolithophores; and how climate changed during the Neogene. At present, however, there is too much uncertainty in all these topics for conclusions based on them to be meaningful. Arguments could be constructed from the present limited knowledge to support either of the two extreme possibilities. Nonetheless, since size range and modal size vary independently it seems reasonable to conclude that the patterns should not be interpreted in terms of a simple response to a single external factor, such as temperature. Also a simple ecological control might be expected to produce much less stable patterns. My suspicion is that the direct control is evolutionary, although an ecological influence (as opposed to control) is extremely likely. This was also the interpretation tentatively suggested by workers on living nannoplankton to whom I showed the data (Westbroek 1986 pers. comm., Gayral 1987 pers. comm., Winter 1987 pers. comm.).

7.5.3 Taxonomic Interpretation. In the preceding discussion I have avoided taxonomic subdivision of the group, however, for many purposes taxonomy is essential. Unfortunately there^many different possible ways of dividing the group depending on both taxonomic philosophy and the causal interpretations discussed above. Three possibilities are illustrated in Figure 23. In Figure 23/A a strictly morphological classification is shown based on convenient size divisions. This has the advantage of being objective and can be used to convey observations on size variation concisely and consistently. It is, however, an explicitly arbitrary classification which does not reflect any possible evolutionary trends. Figure 23/B shows the very different result of using an age related classification based on the assumption that the events at the end of NN15 and within NN10 constitute natural breaks within an evolving lineage. This approach has the considerable advantage that all members of a single taxon of a single age are given the same name. It has the practical disadvantage that identical specimens of different ages may be assigned to different taxa and so can only be named when the age of the sample has been determined. Figure 23/C shows a third possibility, based on the assumption that the variation is due to a number of separate taxa being present at most intervals. On this basis it is most important to describe the typical form of each taxon, and to trace how this varies through time. The

91 DEGREE OF CENTRAL AREA CLOSURE open saall pore closed c9 R.p.gelida R.pseudouabilica “O cz m D.perplexus in (D.antarcticus) in a t -C SIZE R.ainutula R.haqii 4 pa 3 E R.ainuta D.productus

FIGURE 23 - Taxonomic interpretations of Reticulofenestrid size variation. A-C: Illustrations of three possible subdivisions of H.Miocene to Pliocene keticulofenestn, within the described framework of size variation, on the basis of different taxonoaic concepts. See text for discussion D: Adopted coaproaise scheae. E: Generally used subdivision of the group, adopted scheme uses the sane size definitions. ( 4 . sAtKfW i# 8e?, Paros tq*7). 92 definitions of boundaries between taxa are less important, since it is to be expected that some overlap in the size of taxa may occur. This model is only valid if the taxa are genuine species, which is unlikely since despite occasional bimodality there is a continuous range of variation in size and other criteria do not provide independent support for division into these taxa. Each of these schemes has some validity, and is attractive in some respects, and for some purposes; but a single scheme needs to be adopted in order that nomenclature has any meaning - note the very different extents of R.pseudoumbilica in the three diagrams. Separation of the Late Pliocene group is almost certainly justifiable, since associated with the change in size variation is a parallel change in the structural variability of the group. In addition to simple reticulofenestrids Pseudoemiliania lacunosa morphotypes appear, with slitted shields, lower ellipticity and wider central areas. Apparently an evolutionary change in the variation potential of the group occurred at this interval and so it seems reasonable to divide the lineage at this level, as on Figure 23/B. Identical coccoliths nonetheless do occur above and below the boundary. In the NN11-NN15 interval the three-fold minuta - hagii pseudoumbilica division proposed by Backman seems to be justified by the data, although due to the presence of intermediates between these groups and the lack of independent evidence for them being separate I am inclined to regard them as varieties rather than species. In the NN6-NN9 interval these divisions have less justification. The five micron division between R.haqii and R.pseudoumbilica often corresponds to the peak in the unimodal size frequency distributions of this interval. Thus it is tempting to use different taxonomy for this interval, as shown on Figure 23/C. However, their is no independent structural evidence that the small Reticulofenestra interval is an evolutionary event, so this approach which would require the creation of new taxa or revision of the definitions of old ones would be premature. Instead I prefer to apply, artificially, the Late Miocene - Pliocene taxa to this interval since they do permit objective description of the assemblages, and can be used to investigate the changes in them. If further research proves that the small Reticulofenestra interval is an evolutionary break, as suggested on Figures 23/B,C . it would then be appropriate to revise the taxonomy.

93 7.5.4 Biostratigraphic use. Biostratlgraphlc application is more straightforward than taxonomic or causal analysis, since the patterns can be applied independently of any interpretations, although it is preferable to have a theoretical basis. For high resolution biostratigraphy the most interesting new feature in the size variation pattern is the sharp drop in size at the beginning of the small Reticulofenestra interval. This seems to occur consistently within zone NN10. If it proves to be widely recognisable it should form a valuable additionto the standard nannofossil zonations. Theend of the small Reticulofenestra interval is almost certainly gradational and of little use as a biostratigraphic event. For low resolution work the contrast in assemblages above and below this interval is useful, since they provide a first order indication of age in even the most impoverished samples. In the Makran material I found this a valuable additional tool. Poor preservation results in some skewing of the size distribution toward larger specimens, but not so much as to obliterate the trends. On the other hand as my data show the contrast between Early and Late Pliocene assemblages although real is not as dramatic as might be hoped. Since the large specimens are usually rare in the Early Pliocene, reworking can very easily confuse this boundary.

7.6 PRINCIPAL CONCLUSIONS 1. There are consistent variations in the size-frequency distribution patterns of reticulofenestrid coccoliths in the Neogene of the Western Indian Ocean.

2. The most prominent feature of this variation, the small Reticulofenestra interval can be detected in the data of other workers and is probably a widespread low-latitude event.

3. The reticulofenestrids from above and below this interval show different size distribution patterns, but otherwise seem to be identical.

4. The start of the small Reticulofenestra interval, and the contrast in assemblages above and below it, have real biostratigraphic use.

5. Further work on this topic is needed to determine the extent, causes, and taxonomic significance of the changes.

6. The success of this study suggests that equivalent studies on other groups may be worthwhile.

94 PLANKTONIC FORAMINIFERA (Fleisher 1974) S.belenos H.npliiperta S.heteroiorpbosas R. rotaria B.batatas o o B. qainqaerams 0-0- J.ragosas C. artatas n n n n n C.cristatas Btanrolithas B.asfttetricas 0_2- n n P o S. abies (+torif) n n £> R. pseadoatbilica n O__Q □_Q_ B.sarcalas o o o o o n r. n n o B.pettaradiatas a_o___o___ o o n n n n o B.brouteri S. oceatica P.Iacetosa E.haxle/i

o o 0 00000 00 00 Ny Saiples

CORES

INTERPRETATION

FIGURE 24 - BIOSTRATIGRAPHY OF DSDP SITE 219. Ranges of key biostratigraphical species as reported by BOUDREAUX 1974 (solid lines), THEODORIDIS 1984 (dotted lines), and BUKRY 1974c (arrows, based on his core assignments), also my data (circles), planktonic foram zonation (see fig.27 for correlation with nannofossil zonation), and adopted interpretation. Drawn to 150m (Core 14), drilling continued 410m (L.Palaeocene).

95 CHAPTER 8 - BIOSTRATIGRAPHY OP INDIAN OCEAN DEEP SEA DRILLING PROJECT MATERIAL

INTRODUCTION Approximately 200 samples were examined from Deep Sea Drilling Project Sites located in the Western Indian Ocean, and Red Sea. These were used for taxonomic research, including the study of reticulofenestrid size variation discussed in Chapter 7, and to provide information on the stratigraphical distribution of nannofossils in the region. Work on this material also forms the basis of the taxonomic section. The biostratigraphic assignments of the samples and certain features of interest of individual sites are briefly discussed below. Nannofossil distribution charts for each of the sites are given in Appendix 1. The geographical locations of the sites are shown on Figure 28 (p.110). For each sample a detailed count was made of the nannofossil assemblage, by light microscopy using smear slides. All specimens were identified and counted from twenty fields of view. Also further counts of rare taxa were made from two centimetre traverses of the slides. Data reduction was performed with a computer spreadsheet package: This was used to mathematically combine the data from the separate counts; to calculate percentage abundances; to convert these into logarithmic abundance categories; and to create a database for the entire sample set. This database was used to generate the distribution charts for each site, and summary diagrams of the distribution of individual taxa (eg Fig.42).

8.1 SITE 219 DSDP Site 219 was located in the Arabian Sea on the northern end of the Chagos - Laccadive Ridge, (9°2'N, 72°53'E), about 400km west of the southern tip of India). A rather condensed sequence was recovered, some 420m of Upper Palaeocene to Recent sediments. The top 130m of these are Neogene and Pleistocene nannofossil rich sediments. In the Initial Reports of Leg 23 (WHITMARSH, et al 1974), the nannofossils were reported on by BOUDREAUX (1974) and by BUKRY (1974c), and the planktonic foraminifera by FLEISHER (1974a). Subsequently THEODORIDIS (1984) examined the Miocene cores. The biostratigraphic results of these authors are combined with mine in Figure 24. The Neogene succession is divided into two parts with a major break between them. Cores 13 and 14 are Lower Miocene, there is then a 15m gap in recovery followed by Cores 12 to 1, which contain an apparently continuous Late Miocene to Recent

9 5 A succession. Boudreaux interpreted the sequence as nearly continuous succession, with mid-Miocene sediments in Cores 9 to 12. This is not tenable since Amaurolithus spp., D.guingueramus, and Late Miocene planktonic foraminifera occur throughout this interval. This and similar discrepancies elsewhere between his results and those of other workers seem to be primarily differences of interpretation due to his emphasising last occurrences. His observations agree well with those of other workers; with a few exceptions such as his record of D.hamatus in Cores 9-12, this must be a misidentification. The discontinuous interpretation is supported by the lithology of the cores - Cores 1 to 12 are composed of clay rich nannofossil oozes, whereas the Lower Miocene sediments are white planktonic forara - nannofossil oozes, without detrital clays. Curiously, however, this lithological break does not exactly coincide with the palaeontological break. The top section of Core 13 contains an Early Miocene nannofossil assemblage, but is lithologically similar to the Upper Miocene sediments in the cores above, and contrasts sharply with the pure oozes directly below. One preparation from this material contained rare Late Miocene discoasters. These may have been a result of contamination, alternatively they may indicate that this section is Late Miocene in age but largely composed of reworked Lower Miocene sediment. On this interpretation the 30m uncored interval would be a continuation of the Upper Miocene, not a condensed Middle Miocene sequence. The Early Miocene assemblages all contain common to abundant Sphenolithus heteromorphosus, and so plainly come from zones NN4 or 5. Rare S.belemnos occurs at the base of the sequence; Helicospbaeza ampliaperta has been recorded sporadically; and H.obliqua occurs. All these suggest NN4 rather than NN5. Preservation in these cores is highly variable, I only examined in detail my best preserved samples. Biostratigraphy of the higher cores is reasonably straightforward. All samples from them contained abundant well preserved nannofossils. A point of interest is the occurrence of H.inversa Gartner 1977 in Core 1. This species does not seem otherwise to have been reported since its description.

8.2 SITE 223 Site 223 was located on the Owen - Murray Ridge, the offshore continuation of the Indus suture (18°45'N, 60°8'E, about 300km of the Oman coast). It was discontinuously drilled to basement, at 730metres, recovering variable Upper Palaeocene to Recent sediments. These include thick Pleistocene and Upper Miocene successions, and more condensed

96 FIGURE 25 - BIOSTRATIGRAPHY OF DSDP SITE 223. Ranges of key biostratigraphical species as reported by BOUDREAUX 1974 (solid lines), and BUKRY 1974c (arrows), with my data (circles), and adopted interpretation.v Drawn to 500m (Core 28), drilling continued to basement at 730m.

97 Middle Miocene and Pliocene sediments. I looked at several samples from this site since it was close to the Makran, and had a nearly complete Middle Miocene section. The preservation is not, however, very good. There is a marked lithification front in Core 22, and samples from below this had rather overgrown nannofloras. Above this preservation is better, either very good or slightly etched, but siliceous are often abundant and dilute the preparations. BOUDREAUX (1974) and BUKRY (1974c) reported on the nannofossils. Their results agree closely throughout the Neogene, and mine were similar again. This data is summarised in Figure 25, which also shows my adopted age assignments. The only significant point of disagreement is in the age of Core 27. Boudreaux recorded rare Sphenolithus belemnos, Triguetorhabdulus carinatus, Discoaster druggii, and Helicosphaera ampliaperta from this core. He therefore assigned it to the Early Miocene T.carinatus zone of BRAMLETTE & WILCOXON (1967) (=NN1 + NN2 of MARTINI & WORSLEY 1970). However, I did not find any of these species in my samples from this core, but did find S.heteromorphosus, so, like Bukry I have assigned this core to a rather later interval, zones NN4 - 5. The anomalous specimens of Boudreaux may have been reworked. A consequence of this reinterpretation is that a major hiatus, representing most of the Early Miocene must be present at this site. There is no good evidence for hiatuses in the rest of the Neogene, although the Upper Pliocene succession is condensed.

8.3 SITE 231 Site 231 was one of three sites drilled in the Gulf of Aden during Leg 24. All three were continuously cored to basement, but Site 231 yielded the longest succession, probably because it was situated furthest from the spreading axis. 580 metres of Middle Miocene to Recent nannofossil ooze were recovered, with intercalated sands in the Pliocene and Pleistocene. The Neogene oozes have a uniform green-grey marly appearance. Preservation of nannofossils is generally very good in them, and reworking is negligible. The Pleistocene and top Pliocene samples are rather worse than the lower material, due to the input of clastic carbonate and silt, and associated reworking. I consequently did not examine in detail samples from this interval. The nannofossils from this site have been studied by ROTH (1974), BUKRY (1974d), and THEODORIDIS (1984 ). In general their observations agree well with each other and with mine, as summarised in Figure 26. The derived biostratigraphy is thus unusually reliable, and samples can be dated without much trouble. This is particularly true in the Miocene.

98 =3 SS •-< • ~i r n •—» ■ m •l •—• OI

0 0 jo f|o o jo 0 0 ' ° 1 1 N N 19

1 2 0 0 1 0 ° 13 14 NN18

0 0 0 0 ° 15

• 0 0 0 o 16 o o o 0 0 1 7 NN17 0 * 0 o o 1 / 1 0 0 0 ° 18

0 0 0 O 0 19 NN16

8 8 S § 2 0 o o 6 o 0 0 0 o o o ll 2 2 NN15 0 0 0 0 o ° 23

24

0 3 0 3 o ° 25 3 0 0 0 0 ° 26 N N 1 3 /1 4 3 0 3 3 o o 27

0 0 0 0 o o 28

0 l 0 l 0 2 0 * o o 2 9 NN12

d ; 0 l 3 | 0 * o ° 30

31

32

33

o | o : 0 i 3 ; o o 34 35 1

36

37

38

39 NN 11

40

41

0 ! o : o l o I 3 o 42 —

43

44'

45 e

3 • 3 0 I 3 o 47

o : 3 O ° 48 3 o l 0 ° 49 NN10

. • 0 o ° 50

51

o ; 3 = o 52

53

54 NN9 o l 3 ° 55

0 o 56 NN8 3 • 3 ° 57 0 3 ° 58 NN7

3 3 o 59 t» o 3 0 3 ° 60

0 0 o 61 NN6

62

6 3 ?N N 5

64

FIGURE 26 - BIOSTRATIGRAPHY OF DSOP SITE 231. Ranges of key biostratigraphical species as reported by ROTH 1974 (solid lines), THEODORIDIS (1984), dotted lines, and BUKRY 1974d (arrows), also my data (circles), and adopted interpretation. Coring continuous to basement at 575m, basalt Cores 63 4 64.

99 8.4 SITE 242 Site 242 was located on the Davie Ridge (15°51*S, 41°49'E) , a submarine high of uncertain origin, lying between Madagascar and the African mainland. It was drilled discontinuously to 675metres, without reaching basement. Eighteen cores of Recent to Late Eocene clayey foram bearing nannofossil oozes and were recovered. It is remarkable in two respects, firstly all the sediments contain exceptionally well preserved and diverse nannofossil assemblages; secondly the Miocene cores exhibit extraordinary condensation. A result of the condensation is that although only 9% of the Miocene thickness was cored, these cores contain all the Miocene Standard Nannofossil Zones, except NN8. They appear to represent about 70% of the Miocene. This curious result is well documented by the biostratigraphic studies in the Initial Reports (SIMPSON et al 1974) of MULLER (1974a), BUKRY (1974e) and ZOBEL (1974, planktonic foraminifera). It is also supported by my observations. Neither of the nannofossil workers gave range charts but their biostratigraphic assignments should be reliable in light of the quality of the material. Also, as shown on Figure 27, their results agree well with each other, with the planktonic foram data, and with my data. This means that the condensation has to be accepted as a genuine feature of the cores, rather than an observational artefact. Condensed sequences are most obvious in Cores 6, 7, and 8. These are each nominally 9 metres long, but span respectively, eNNll to mid NN9, mid NN7 to early NN5, and early NN4 to mid(?) NN1. BUKRY <1974e) also suggested that Core 5 was condensed, containing the entirety of his A.primus zone, an interval usually represented by thick sequences. Cores 1, 2 and 4 have tightly constrained ages and seem to represent time periods proportional to their thicknesses. Core 3 is more ambiguous, Bukry records C.acutus, C.rugosus and A.primus from Section 3-6, this is suggestive of the NN12/13 and in conjunction with other evidence implies that the core extends from zone NN15 to NN12. It also suggests that it overlaps in age slightly with Core 4. The Palaeogene cores do not contain such well preserved microfossils, and so their biostratigraphy is less certain. Cores 12 to 18 record a thin but approximately continuous section of Late Eocene - Earliest Oligocene age. Core 9 spans the Oligo - Miocene boundary as indicated by planktonic foraminifera and by my work - last occurrence of Sphenolithus ciperoensis, occurrence of S.delphix and S.capricornutus. The bulk of the Oligocene is represented by the 60m interval between Cores 9 and 12. It is thus either incomplete or condensed.

100 FIGURE 27 - BIOSTRATIGRAPHY OF DSDP SITE 242. Summary of information on the biostratigraphy of Site 242, with suggested interpretation. C - Chronostratigraphy; P - Planktonic foram zones; N - Nannofossil Zones; M - Data of MULLER 1974< B - Data of BUKRY 1974e; Z - Data, of ZOBEL 1974 (planktonic forams); Y - My data; R - Reported Recovery, cores in black, depth in metres. Chronostratigraphy and zones scaled from HAQ et al (in press). For each data set vertical lines represent indicated age of core, horizontal lines datum planes. Thin lines from cores show adopted interpretation, see text for discussion. 101 As discussed above the palaeontological evidence for the dating of the cores is very strong, the cause of the condensation is not, however, clear. Three classes of explanation are possible. 1. Sedimentological: Repeated fluctuations in sedimentation rate might conceivably create the column as indicated. This is highly unlikely, particularly as the uncored intervals, such as NN8, which on this basis would need to be represented by expanded sections are not normally represented by thick sequences. It would also require incredibly fortuitous drilling. 2. Tectonic: Thrusting, possibly related to formation of the Davie Ridge, could produce repetition of the sequence. This is geologically more plausible, but again would require fortuitous drilling. Also the cores do not show obvious signs of tectonic disturbance. 3. Drilling artefact: This was the solution adopted in the Initial Reports, citing the discussion of drilling problems in MOORE (1972). On this basis each core should contain sediments recovered from an interval 30-60m thick, instead of the nominal 9m. This would be very unusual, and the relevant cores are composed of semi-lithified chalks which show little sign of drilling disturbance (SIMPSON et al 1974, my observations). The discussion of MOORE (1972) does not help explain the problem since it was concerned with mechanisms for causing incomplete recovery of intervals rather than excessive recovery. His discussion does, however, show how drilling on the early DSDP legs was a rather erratic, and poorly understood, technology. A drilling artefact is thus the least unlikely solution and so, in the absence of other evidence, must be tentatively adopted. Despite this problem the high quality of the nannofloras, and the independent planktonic foram dating, mean that the site is an invaluable source of information.

8.5 RED SEA SITES, 225 AND 227 During the latter part of DSDP Leg 23 drilling was carried out at six sites in the Red Sea (WHITMARSH et al 1974). Two of these, Sites 225 and 227, penetrated Lower Pliocene and possible Upper Miocene sediments before encountering evaporites. I looked at samples from both these sites since they provided possible examples of assemblages with lowered diversity due to environmental restriction and so were of palaeoecological interest. The two sites were drilled about half-way along the Red Sea on the East flank of the axial trough, with Site 225 (21°19'N, 38°15'E) rather further from the trough than Site 227 (21°20'N, 380|8'E). They have similar successions, although detailed lithostratigraphic correlations

102 do not seem possible. They bottom in evaporitic halite and anhydrite, seismic profiles reveal that these are several hundred metres thick, but only the top hundred metres or so were drilled in each case. BOUDREAUX (1974) reported Late Miocene, NN11, nannofloras in sediments associated with the evaporites from both sites. The evaporites are followed by Pliocene and Pleistocene marine sediments - 170m thick at Site 225 and 220m thick at Site 227. These are composed predominantly of nannofossils, detrital silt, and clay in varying proportions with subsidiary rhombic calcite, dolomite, and gypsum, probably of evaporitic origin.

8.5.1 Biostratigraphy. As a result of the rather restricted nature of the nannofloras, particularly from the Lower Pliocene sediments, there are a number of biostratigraphic problems. The first of these is the Late Miocene age for the top of the evaporite sequence. Boudreaux based this on his recording of Discoaster guingueramus and Tziquetorhabdulus rugosus. This would correlate well with the ending of Messinian evaporite deposition in the Mediterranean. I was unable to repeat his observations in samples I examined, but I did find Ceratolithus armatus in a sample from Site 225, Core 21. This is good evidence for an age of NN12 (base Pliocene). In Site 227 there appears to be a distinct first occurrence datum for Ceratolithus cristatus, at about the same level as the last records of T.rugosus, this again suggests the presence of zone NN12. BUKRY (1974c) assigned much of Site 227 (Cores 17 to 31) to his D.berggrenii / C.primus zones (=NN11). This is one of the rare cases where I significantly disagreed with his assignments - C.cristatus definitely occurs within this interval, so his age is plainly untenable for much of the interval. My suspicion is that he was misled by the fact that Helicosphaera intermedia, which normally can be taken as a good indicator of Miocene age, occurs within this interval. It seems reasonable to conclude that, marine conditions were established in the Red Sea in 1NN11 / early NN12, ie within the Messinian stage. Hence evaporite deposition here and in the Mediterranean, probably overlapped in time, although it is not possible to tell if they ended synchronously. The time of commencement of Red Sea evaporites is poorly constrained. In the main part of the sequences the only clear datums are the Early/Late Pliocene (NN15/16) and Plio/Pleistocene (NN18/19) boundaries. The other zonal boundaries can only be tentatively placed, as indicated on the range charts (Appendix 1).

103 8.5.2 Nannofloras The Early Pliocene nannofloras of these two sites show a number of unusual features, as described below. These are probably primarily the result of ecological restriction in general, and raised salinity in particular, since: Salinity variations are well known to affect nannofloras (eg BUKRY 1974a, WINTER et al 1978); benthonic and planktonic foraminifera also show the effects of ecological restriction (FLEISHER 1974b); the Red Sea presently has slightly raised salinities; evaporites directly underlie the Pliocene sediments; gypsum and dolomite occur within the marine sediments; the nannofossil assemblages contain delicate forms such as Pontosphaera, Scyphosphaeza, and Discoastez pentazadiatus, so preservation cannot be the only cause of the nannofloral aberrations. The variations are discussed below for each relevant family. Sphenolithaceae: S.abies is very abundant in all samples up to its extinction level, and frequently is more abundant than Reticulofenestra pseudoumbilica. This is unusual and suggests an ecological control. Prinsiaceae: Reticulofenestza pseudoumbilica is common to abundant in most samples, as usual in sediments of this age. Larger specimens are, however, rarer than usual (Fig.22, p.87), and become completely absent slightly before the last occurrence of Sphenolithus abies. Above the NN15/16 boundary Reticulofenestrid assemblages appear normal with Pseudoemiliania and R.dozonicoides present, and of normal size. At the boundary there is a distinct small Gephyzocapsa interval. Coccolithaceae: Coccolithus pelagicus is virtually absent throughout the Early Pliocene. It is, however, rare in the Early Pliocene Indian ocean material, so this may not be a local environmental effect. Calcidiscus leptoporus occurs but is extremely rare, this is definitely unusual. Geminilithella zotula and Umbilicosphaeza jafari have patchy preservation dependant patterns. Helicosphaeraceae: Helicosphaeza caztezi is present in virtually every sample, and usually at higher than normal abundances. Specimens are frequently large, with well developed flanges. It seems to have flourished under the prevailing conditions - this could possibly be in part a reflection of near shore conditions, but it also must indicate at least a tolerance of raised salinities. At Site 227 another form occurs, this is similar to //.intermedia - i.e. with an optically discontinuous bar. This is well above the normal stratigraphic occurrence of //.intermedia, it may be an ecologically restricted morphotype. Pontosphaeraceae: Both Scyphosphaeza and Pontosphaeza species are well represented and form a more significant proportion of the assemblages than is usual (several examples in Plate 5). Of the

104 Scyphosphaera species, S.apsteinii and S.globulata are consistently present, and a number of other species also occur. This is particularly significant since they have a low preservation potential, so their abundance is unlikely to be a result of poor preservation. Discoasteraceae; Discoasters show variable behaviour. D.vaziabilis seems to be entirely absent, even from the Upper Pliocene. D.brouweri, and D.surculus occur but are very rare, markedly rarer than might be expected given the latitude of the samples, and the abundant occurrence of sphenoliths and other warm water species. Also when they do occur they are often noticeably worse preserved than other nannofossils from the same samples. Possibly these discoasters did not live in the Red Sea, but occasionally drifted in. They occur more consistently in the Upper Pliocene than the Lower Pliocene. D.pentaradiatus by contrast is anomalously abundant. It is present in virtually all samples up to its last occurrence level, and in a couple of samples disarticulated D.pentaradiatus arms are the most important nannofossil component. Its distribution pattern in detail is, however, strongly affected by preservation, with highest abundances in overgrown samples. Ceratolithaceae: Ceratoliths like discoasters are rare, but less so than might be expected given their general rarity in the NW Indian Ocean. I found C.cristatus in several samples and C.armatus in one. Boudreaux records many occurrences of C.cristatus, C.rugosus, and Amaurolithus tricorniculatus. Possibly ceratoliths were more euryhaline than discoasters. Braarudosphaeraceae: I only found one or two specimens of Braazudosphaera bigelowii, and they were in the Upper Pliocene; Boudreaux's results were similar. This is concordant with an interpretation of raised salinity, B.bigelowii is well known to favour lowered salinities. Thoracosphaeraceae: Thoracospheres are a conspicuous element of the nannofloras, both complete tests and debris, in virtually all samples. They are much more abundant than in any of the Indian Ocean samples I examined and were particularly common in the Early Pliocene material although Boudreaux also records them consistently from the Pleistocene. The main species present are T.heimii and T.albatrosiana (recorded as T.saxea by Boudreaux). Some other, unidentified, species with fine grained walls also occurred. Evidently Thozacosphaeza flourished under the prevailing conditions.

105 8.5.3 Discussion Prom the notes above it is clear that species shoved highly variable responses to the ecological factors, which affected the assemblages seen in these Red Sea cores. As argued above altered salinity is likely to be the most important of these. It also seems clear that these factors acted more strongly in the early Pliocene. Subsequently assemblages are more normal, suggesting that there was a better connection to the Indian Ocean. A likely cause of this is renewed rifting in the Red Sea - Gulf of Aden system as suggested by palaeomagnetic evidence (e.g. GIRDLER & STYLES 1974, 1978). Sea level control is unlikely since, according to the eustatic sea level curve of HAQ et al (in press), sea level was falling at this time. Unfortunately it is not possible to detect a simple hierarchy of samples with varying assemblages related to increasing ecological restriction. The distribution of individual species seems also to be controlled by preservation, and possibly by more than one ecological factor. With more and better samples it might be possible to employ mathematical factor analysis to resolve more clearly the effects of preservation, temporal change, and one or more ecological factors. Nonetheless a crude table of ecological preferences can be constructed from the general distribution patterns, as discussed above.

FLOURISH NORMAL REDUCED EXCLUDED S.abies ceratoliths C.1.macintyrei ?C.pelagicus Thoracosphaera R.p.haqii C. l.leptoporus R.clavigera H.carteri ?G.rotula D. brouweri D.variabilis H.cf.intermedia ?U.ja£ari D.surculus R.clavigera D.pentaradiatus R.p.pseudoumbilica B.bigelowii Scyphosphaera Pontosphaera

EXPLANATION Flourish: Species which occur at higher than normal abundances, or dominate assemblages. Normal: Species which do not show anomalous distribution patterns. Reduced: Species which occur, but at markedly reduced abundances, particularly in the Early Pliocene. Excluded: Species which are entirely absent from the Early Pliocene, although in some cases they do occur in the Late Pliocene assemblages. (N.B. trinomial combinations are abbreviated here as discussed in Chapter 10, eg R.p.hagii = Reticulofenestra pseudoumbilica var. hagii).

106 8.6 OTHER SAMPLES Isolated samples were examined from several other sites. The notes below briefly outline the justification for my biostratigraphic assignments.

LEG 22, Sites 211-218. References: GARTNER (1974), BUKRY (1974b). 214-6-2, 214-6-3: Top NN15; assignment of Gartner and Bukry; co-occurrence of Pseudoemiliania lacunosa and Reticulofenestza pseudoumbilica var. pseudoumbilica.

LEG 24, Sites 231-238. References: ROTH (1974), BUKRY (1974d). 238-11-2, 238-11-5: Top NN15; assignment of Roth and Bukry; co-occurrence of R.p.pseudoumbilica, small Gephyzocapsa, and ?P.lacunosa.

LEG 25, Sites 239-249. References: MULLER (1974a), BUKRY (1974e). 239- 4-4: 1NN11 - R.p.zotazia subzone of THEODORIDIS (1984); occurrence of R.p.zotazia, also Discoastez quinquezamus and Amaurolithus tricorniculatus. 239-5-2: eNNll: Co-occurrence of D.quinquezamus, Minylithina convallis, and Catinastez calyculus; Bukry placed this section in his A.primus zone (=1NN11), but this seems unlikely. 239-8-4: NN9; assignment of Bukry; Occurrence of D.hamatus, also C. coalitus and C.calyculus} Other samples from this interval are very poor. 241-7-1; base NN16; Assignment of Muller (Bukry places it lower); Occurrence of P.lacunosa without R.p.pseudoumbilica. 241-7-5; top NN15; Assignment of Muller (Bukry places it lower); Occurrence of R.p.pseudoumbilica and Cezatolitbus czistatus. 249-4-1, 249-4-5: 1NN11, above R.p.zotazia subzone; Assignment of Muller and Bukry, planktonic forara data (N17); Co-occurrence of D. guingueramus and Amaurolithus spp., absence of R.p.zotazia. 249-12-6, 249-13-3: NN10; Assignment of Muller and Bukry; Co-occurrence of D.calcazis, D.pzepentazadiatus, D.pentazadiatus, M.convallis. 249-16-1: NN7; assignments of Muller and Bukry.

LEG 26, Sites 250-258. Reference THIERSTBIN (1974). 251A-4-2 & 4-5: NN15; assignment of Thierstein. 251A-10-5 to 10-CC: top NN11 to early NN12; last occurrence of D.guingueramus within core; T.rugosus and Amaurolithus throughout core. 251A-13-2 to 16-2: top NN9 to eNNll; Catinaster and D.hamatus in 16-2; Minylithina convallis (very common) and D.pentazadiatus in 15-2 to

107 13-2; D.quinquezamis in 13-5 to 13-2. 215A-26-2: NN4/5; Presence of S.hetezomozphosus. 258A-6-1: base NN16; Co-occurrence of small Gephyzocapsa, P.lacunosa, D.b.tamalis and D.b.asymmetricus, without R.p.pseudoumbilica or S.abies; Thierstein gave earlier age. 258A-8-1, 258A-8-2, 258-4-CC: NN12; Planktonic foram data (zone N18, +/-=NN12); Occurrence of A.tzicozniculatus, absence of D.quinquezamus, D.b.asymmetzicus and Ceratolithus spp. Thierstein interpreted this interval as NN11 age with contamination, my data did not support this suggestion.

108 CHAPTER 9 - APPLIED BIOSTRATIGRAPHY; THE MAKRAN OF PAKISTAN.

9.1 INTRODUCTION, REGIONAL GEOLOGY My principal biostratigraphic research whilst at Imperial College was on material from the Makran of Pakistan, as part of a joint Anglo-Pakistani project studying the Makran Accretionary prism (principal researchers Dr.J.K.Leggett & Dr.J.P.Platt). This work was important for developing my biostratigraphic approach, and to a limited extent testing it, but owing to the poor quality of the material it gave few results of interest for nannofossil research. The main purpose of this chapter is thus to document my biostratigraphic results, and to discuss their geological implications. It is worth doing so at some length since the rather problematic nature of the results means that their interpretation and application need to be done carefully. Also the area is likely to be the subject of continuing active research and discussion of problems may aid the planning of this research.

9.1.1 Tectonic setting. The Makran Region as shown on the regional maps (Figs.28, 29) is a reasonably well defined segment of the southern margin of the Eurasian Plate, and so a relic of the northern margin of the Tethyan Ocean. To the East the Makran is bounded by the Chaman and Ornach-Nal Faults which form a major strike-slip system between the Indian and Eurasian plates. To the west it is separated from the Zagros Mountain Belt by the Oman Line. This is a rather poorly defined zone of North-South striking faults, it offsets the Zagros Mountains from their continuation in Oman. Both the Zagros Mountains and Oman are formed of Cambrian to Palaeogene sediments deposited on the NE margin of the Arabian Plate, with the Zagros deformed by continent - continent collision of Arabia and Eurasia. The Makran by contrast consists predominantly of Tertiary turbidites deposited on the Eurasian Margin. The Oman Line thus forms a part of the main plate boundary. The Makran appears to be an accretionary margin with subduction of oceanic material occurring beneath it (STONELEY 1974, FARHOUDI & KARIG 1977, DYKSTRA & BIRNIE 1979, JACOB & QUITTMEYER 1979). Evidence for active subduction is provided by seisraological data, by the presence of a volcanic arc to the north, and by the structural style - with East-West striking compressional features both onshore and offshore. In addition the computed relative motions of the Indian, Eurasian and Arabian plates make subduction virtually essential for a plate-tectonic model of the region (JACOB & QUITTMEYER 1979). There is, however, no true trench, and the volcanic centres are a long way inland, giving an

109 FIGURE 28 - THE WESTERN INDIAN OCEAN Map showing location of the Makran and studied DSDP Sites in relation to plate boundaries. Mainly based on the Geologic Map of the Indian Ocean of HEEZEN et al (1978). Stippled area - Neogene Oceanic Crust. Dotted lines - edge of continental crust and aseismic submarine ridges. f\- active volcano. 110 "arc-trench gap" of nearly 500km, which is several times broader than is usual.

9.1.2 Tectonic divisions, and history of research, of the Makran of Pakistan. As a result of the East-West strike the geology of the Makran is most readily described in terms of a south to north sequence of tectonic divisions. For present purposes five divisions will suffice, these are discussed below, and shown on Figures 29 & 30. This is also a convenient format for outlining other studies since - even though many workers have commented on the likely relations of the various divisions - most research has been confined to single divisions. The most important exception is the study of the Hunting Survey Corporation, who mapped the entire area - primarily by means of aerial photography. Our research was concentrated in the Coastal Makran, as was that of a previous study by Marathon Oil Company.

1. Continental shelf and slope: The deeper parts of the Gulf of Oman Oclkm.) have been extensively studied during geophysical cruises of the RRS Shackleton (Legs 3/75 and 1/80) and MV Atlantis (Leg 96/13). The results of this work are described and discussed in WHITE (1979, 1982, in press), WHITE & KLITGORD (1976), WHITE & LOUDEN (1982), and WHITE & ROSS (1979). The most significant result of this work is the discovery that the thick sediments of the Gulf of Oman abyssal plain are folded at the base of the Makran continental slope into a series of anticlinal ridges, with ponded slope sediments between them. These are classic accretionary-front structures, also seen in for instance the Nankai Trough off SW Japan (LEGGETT et al 1985). The continental shelf and upper slope are best documented by a set of multi-channel seismic lines shot by Marathon as part of their evaluation of the Makran Region. Some of these are published and briefly discussed in HARMS et al (1984), others were made available to us. They show that in this part of the margin sedimentation predominates over tectonics. The slope and shelf are not formed of accreted slices of basin plain material but rather of in-situ sediments. These are sigmoidal sets of progradational slope-front fill and more planar, but often very thick, shelf sediments. On some lines these sediments can be seen to unconformably overlie the folded basin-plain sediments.

2. Coastal Makran: This belt extends from the coast 50 to 100 kilometres inland, it is formed of thick moderately deformed mid-Miocene to mid-Pliocene basin-plain, slope and shelf sediments. In

111 addition undeformed Upper Pliocene and Quaternary sediments occur in places near the coast, as a result of recent uplift. The area was mapped by HUNTING (1960), and formed the principal study area for both the Marathon Oil group (HARMS et al 1982, 1984, CRAME 1984) and our work (LEGGETT & PLATT 1984, PLATT et al 1985, PLATT & LEGGETT 1986). It is discussed in more detail in later sections.

3. North and Central Makran Ranges: East-Vest striking mountain ranges continue for 200km North of the Coastal Ranges. This zone is the inland continuation of the Coastal Ranges, but is more deformed and lacks the slope and shelf sediments. The only detailed work on these ranges is that of HUNTING (1960). The rocks are poorly dated but probably mainly Oligocene to Early Miocene, based on very limited palaeontological data and regional stratigraphy. In plate-tectonic models this zone is interpreted as the Oligo-Miocene part of the accretionary prism.

4. Hamun-i-Mashkel Depression: This is an actively subsiding area with an alluvial and aeolian fill. In plate-tectonic terras it is interpreted as a fore-arc basin (FARHOUDI & KARIG 1977).

5. Chagai Hills Area: This is the most geologically complex part of the Makran, and the only one in which crystalline rocks are important. In addition to the survey of HUNTING (1960) various authors have worked in the area, this work is synthesised in ARTHURTON et al (1982). Subsequently the intrusives have been studied and redated by BREITZMAN et al (1983). The main elements of the region are; thick volcanics (Sinjrani Group), and associated variable Maastrichtian sediments (Humai Group); extensive Palaeocene fore-arc and flysch sediments; Eocene shallow marine ; and Quaternary Volcanoes. The Cretaceous and Palaeogene rocks are interpreted by ARTHURTON et al (1982) as the remnants of an early volcanic arc - fore-arc - accretionary prism system. The, andesitic, Quaternary Volcanics are widely interpreted as part of a modern volcanic arc. In addition there is extensive granitic to dioritic plutonism, in the Chagai Hills. Field evidence provided a range of possible ages for these intrusions: post-Eocene (VREDENBURG 1901); Cretaceous (HUNTING 1960); and Late Cretaceous to Early Eocene (ARTHURTON et al 1982). Radiometric work by BREITZMAN et al (1983) indicates two distinct pulses of activity: one c.35Ma (Early Oligocene), the other c.20Ma (Early Miocene). These ages are based on a rather small data-set but correlate well with plutonism in SE Iran (cf BERBERIAN & BERBERIAN 1981).

112 Extension into the Iranian Makran: The Iranian Makran is even less well documented than the Makran o£ Pakistan, and due to political problems few studies have extended across the border since the early work of VREDENBURG (1901). The most important references are FALCON (1974) and SHEARMAN (1976)^for the main units, and GANSSBR (1971) for the volcanics. The main elements of the active arc can be correlated directly into Iran although with some offset - these include the volcanic arc, the fore-arc basin (=Jaz Murian Depression), the continental shelf and slope, and the Coastal Makran (=Upper Flysch of Falcon). The older terranes, of the North and Central Ranges and the Chagai Area, are less directly correctable. Rocks comparable to those of both terranes are described by Falcon as occurring together south of the Jaz Murian depression. The explanation is probably that the Lut Block, a microcontinental mass of crystalline basement, has deflected the terranes southward.

9.2 LITHOSTRATIGRAPHY OF THE C0ASTA1 MAKRAN. The sediments of the Coastal Makran appear to have been deposited under conditions similar to those of the modern offshore regime (HARMS et al 1984, our work), with rapid shelf - slope - basin-plain transition. In consequence there are readily distinguishable lithofacies corresponding to these three broad zones, but they are strongly diachronous with complex lateral facies relationships. For this reason it has not been possible to establish a conventional lithostratigraphy, instead three broad facies groups are recognised. These are, the Panjgur Facies (basin-plain turbidites), Parkini Facies (slope siltstones), and Talar Facies (cyclic shelf deposits).

9.2.1 Panjgur Facies (basin-plain turbidites). Within the Coastal Makran this facies outcrops in the cores of hanging-wall anticlines, with thicknesses up to 400m observed (eg Pasni and Branguli sections). Nannofossil assemblages from these sections all give late Middle Miocene or early Late Miocene ages (C.NN7-9). Northward from the coast the facies is increasingly widely exposed. It is the dominant facies of the North and Central Ranges, where it is several kilometres thick, and probably of Late Oligocene to Early Miocene age. The most characteristic sub-facies is thick bedded turbidites. These are decimetre to metre thick beds, amalgamated into packets several metres to several tens of metres thick, separated by thick calcareous mudstone layers. Sole-marks are well developed and within the Coastal Makran consistently indicate a predominant flow toward the

113 Box - area of Figure 3,1. 115

FIGURE 30 - SCHEMATIC SECTION THROUGH THE MAKRAN ACCRETIONARY SYSTEM. Greatly modified version of diagram in JACOB & QUITTMEYER (1979). Vertical scale exaggerated, structure and geometries entirely schematic. Major tectonic divisions numbered 1-5, as in text, and on Fig.29. WSW (ie parallel to the coast, from a source to the Bast, probably the Indus). Between this sub-facies and the Parkini slope siltstones there is usually an intermediate Panjgur sub-facies with thinner (cm-dm) turbidites; these are still usually in packets but are not amalgamated, and form a smaller proportion of the |total sequence. Sole-marks indicate a similar flow direction to that in the thick-bedded subfacies, but ripple marks indicate predominantly southward waning flows. This suggests influence of the slope. This sub-facies was best developed in the Branguli area (locations on Fig.32), where it was mapped separately as the Branguli Facies. Blsewhere it was less readily distinguishable and was mapped as a sub-facies of the Panjgur. In the area to the north of the Kulanch Syncline turbidite beds are particularly rare. Hunting's survey mapped Parkini Facies here; however, deep-marine trace fossils and occasional very thick (several metres) laterally persistent turbidite beds occur. Hacrofossils are absent from the Panjgur Facies but trace fossils are often abundant. These are systematic grazing traces of the Nereites ichnofacies, such as Palaeodictyon and Cosmoraphe, indicative of deep marine conditions. Hunting's do not record any microfauna. HARM'S et al (1984), however, mention finding bathyal to lower bathyal, but not abyssal, benthic foram assemblages. This is in accord with the sample assemblages that I looked at (Table 8). Using the synthesis of PUJOS-LAMY (1973) there are indicators both of moderate bathyal depths (eg B.a££inls) and of deep water (particularly the diverse and abundant agglutinate assemblage). A depth similar to that of the modern basin-plain (c.3km) would be reasonable.

9.2.2 Parkini Facies (slope siltstones). This is the most widely outcropping facies in the Coastal Makran, and sections up to 3.5 km. thick have been measured (CRAME 1984). Nannofossils indicate mainly Late Miocene age (NN10-11) in the south of the area and Middle/Late Miocene age (NN8-10) in the north. The predominant lithology is grey calcareous silty mudstone or muddy siltstone (true mudstones are very rare). Thin turbiditic sands occur near the base of the facies and thin storm sands near the top. Slumped horizons are common, and also low angle syn-sedimentary discordances that may represent slump scars. The facies is interpreted as a slope deposit primarily on the basis of analogy with offshore deposits, and the position between basin-plain and shelf deposits. Macrofossils are confined to rare, impoverished, faunas of small bivalves (CRAME 1984). Trace fossils are also rare, this is surprising and may reflect a low primary organic content in the sediment. Benthic microfossils are abundant in some samples, although not as universally

116 as HUNTING (1960) suggest. The assemblages are dominated by buliminid forams (Table 8), which is concordant with the slope interpretation.

9.2.3 Talar Facies (cyclic shelf sediments). The Talar is the most variable and the most diachronous of the facies, it ranges in thickness continuously, from a couple of hundred metres, to over 6,000 metres. Preserved sediments range in age from Late Miocene to Recent. The main Talar Facies outcrops are in two large asymmetric synclines (the Kulanch and Hinglaj Synclines). These have similar sedimentological features, and similar geometries, with rapid facies variations and transition to Parkini Facies (Fig.31). The successions in the two synclines are, however, different in detail, and they almost certainly represent originally separate depocentres. The differences have lead to a proliferation of formation names, for simplicity all these are considered here as one facies group (the most important names are given on Fig.37). The facies represent environments from shoreline to outer-shelf (our work, HARMS et al 1984, CRAME 1984). In all environments the succession consists predominantly of cycles 20 to 50 metres thick. These coarsen upwards to sandstone, from shale, siltstone or very fine sandstone. They vary continuously from near-shore sand dominated cycles with coarse trough cross-bedded sandstones, to outer-shelf cycles with thin fine-grained sandstones forming only a few percent of the sequence. It is not clear whether the cyclicity is related to tectonic or climatic control but individual cycles, can be traced laterally for tens of kilometres until they pass into slope mudstones, or are truncated by erosion. Thick successions are made up of hundreds of these cycles. Molluscan faunas are abundant and show clear shoreline to outer shelf variation (CRAME 1984), as do trace fossils. Microfossils are common at least in the offshore shales, they are heavily dominated by Ammonia becaiii, but often also contain diverse ostracod assemblages.

9.2.4 Deformation in the Coastal Makran. As shown on the sections (Figs.30, 35 & 36) the sediments of the Coastal Makran have been subjected to strong North-South compression resulting in 20-30% shortening. The Panjgur - Parkini - Talar sequence is, however, entirely conformable, and there is no evidence for syn-sedimentary deformation (HUNTING 1960, AHMED 1969, PLATT et al 1985). These structures thus are not the result of continuous deformation intensifying and rotating accretionary thrusts, as was predicted by FARHOUDI & KARIG (1977), but rather are the product of a single tectonic episode (PLATT et al 1985).

117

FIGURE 32 - TALAR FACIES A. Simplified section across modern shelf showing thick sequence of shelf sediments (stippled). Based on seismic sections and diagrams in HARMS et al (1984), section about 50km across. B. Simplified section across onshore syncline of Talar Facies sediments. The Hinglaj and Kulanch synclines have similar form. Adapted from section in PLATT et al (1985), about 30km across. 119 9.3 BIOSTRATIGRAPHY 9.3.1 Material and methods The principal material available to me was a set of approximately four hundred samples collected by Dr.J.K.Leggett and Dr.J.P.Platt during field seasons in 1983 and 1984, from the area shown in Figure 32. It had been hoped that I would be able to undertake detailed field sampling, but it was not possible to obtain permission for me to enter the field area. Additional samples were, however, available from HDIP (Hydrocarbon Development Institute of Pakistan) fieldwork in other parts of the Coastal Makran during 1982 and 1985, although they did not arrive until Spring 1986. I examined about one hundred of these. Of this material about four-fifths of the samples were barren or virtually barren, results from the others are giVen in tables 1 to 8. The assemblages were usually poorly preserved (weak to strong etching), and of low abundance. Extensive experimentation with different preparation techniques failed to give significantly better results, mainly because detrital clay and fine silt particles occurred at all size grades and so could not readily be separated from the nannofossils. I concluded eventually that simple preparation techniques were the most effective. My standard procedure was overnight disaggregation in water, followed by one minutes settling to remove silt, and dilution to the appropriate concentration to make smear slides. All work was done in the light microscope. The main cause of the disappointing quality of the material was probably the high rate of clastic deposition. This would have resulted in a low initial concentration of nannofossils, whilst diagenesis has reduced this further. The Marathon workers experienced similar problems in both surface and drill samples (Ellis 1984, pers. comm.), contrary to suggestions in HARMS et al (1984). These factors also affect the calcareous microfauna. Additional problems were caused by the low sampling density, by reworking, and by the fact that the total time span involved was rather short (Mid-Miocene to Late Pliocene, clOMa).

9.3.2 Zonation. As a result of the problems mentioned above, it was not easy to apply the standard nannofossil zonation, which is largely based on rather rare species. Instead my biostratigraphy had to be based mainly on secondary markers and on large sale changes in the nannofossil assemblages, and many zones had to be amalgamated. This was necessary to such an extent that it seemed misleading and inappropriate to continue to refer samples to the standard zonation, and so instead I used a simplified scheme of six assemblage zones. The basis of this

120 scheme is discussed below, and shown diagrammatically in Figure 33.

Assemblage Zone I; approximately equivalent to Zones NN3-5. These are the oldest assemblages found. They are primarily characterised by the presence of Cyclicazgollthus flozidanus, as the dominant coccolith. Coccollthus pelaglcus, Discoastez deflandzei, Sphenolithus heteromorphosus, and S.moriformis are also usually common. The occurrence of S.heteromorphosus, or of long ranged Neogene species such as Calcidiscus leptopozus, Geminilithella zotula, or Helicosphaeza carter! indicate a maximum Early Miocene age for all assemblages seen.

Assemblage Zone II; approximately equivalent to Zones NN6 and 7. This assemblage is readily distinguishable from Assemblage I due to the replacement of Cyclicazgolithus flozidanus as the dominant coccolith, by Reticulofenestza pseudoumbilica (large to medium forms). In addition common D.exilis group discoasters replace D.deflandzei. A few samples also contained D.kuglezi, C.miopelagicus, or //.intermedia. C.pelagicus and S.moriformis are common.

Assemblage Zone III; approximately equivalent to Zones NN8, NN9 and lower NN10. Similar to Assemblage II in composition, but with addition of extra discoaster species indicative of post NN7 age; D.bellus, D.calcazis, D.hamatus, D.prepentaradiatus, and C.coalitus. Since many of these species have five rays the presence of even unidentifiable five-rayed discoasters can be an useful suggestion of maximum age. If these species were reasonably common the interval could, of course, be sub-divided into the standard zones, but in practice I rarely found more than half-a-dozen specimens of the entire group in any one sample.

Assemblage Zone IV; approximately equivalent to Zones NN11 and the later part of Zone NN10. As described elsewhere (Chapter 7) there is a clear change in Reticulofenestza assemblages within zone NN10, resulting in a shift in dominant size from 5-8 microns to 3-5 microns. This shift is discernible in the Makran; indeed I first noted it in this material. In addition there is a roughly synchronous change in the discoaster assemblage, with common D.exilis and D.cf.bzouwezi being replaced by rarer D.brouweri (s.s., with down curved ray tips), D.pentazadiatus, D.quinquezamus, D.suzculus, and D.vaziabilis. Of these D.pentazadiatus is particularly rare in the Makran, probably due to dissolution.

Assemblage Zone V; approximately equivalent to Early Pliocene (Zones NN12-15). Similar to Assemblage IV, but without D.guingueramus. Also large Reticulofenestza often occurs, although always at low

121 abundance, and bridges are sometimes present in Coccolithus pelagicus. I did not find any ceratoliths in the Makran material, so bridged C.pelagicus is the only positive indicator of maximum age.

Assemblage Zone VI; approximately equivalent to Late Pliocene (Zones NN16-18). The Early-Late Pliocene boundary is well marked by a number of events, including the last occurrence of sphenoliths, and of R. pseudoumbilica, and the first occurrence of Pseudoemiliania lacunosa and small Gephyzocapsa. This boundary should be very reliable, unless reworking is severe. Subdivision of the interval is impractical in the Makran since discoasters are very rare in the Talar Facies sediments, which form the only outcrops of this age. Pleistocene assemblages were not found, they are usually readily distinguishable by the presence of common Gephyzocapsa.

9.3.3 Reworking A prominent feature of virtually all the nannofossil assemblages is the presence of reworked pre-Middle Miocene species. These are included in the tables, since they give some information on the geological history of the area (see discussion section). They can be divided into three groups: A. "Early Miocene": Essentially the Zone I indicator species (C.floridanus, D.deflandrei and S.heteromorphosus). All these species repeatedly occur in association with much younger nannofossils and so are plainly reworked (eg Tables 1 & 2). Specimens of C.pelagicus and S. raoriformis almost certainly must be reworked with them, but cannot be separated from in-situ specimens. B. Palaeogene: This is an heterogeneous group of species of Eocene and Oligocene age. They are never common, and so reworking of other, less distinctive, Palaeogene species is probably not a great problem. C. Cretaceous: An unambiguous group since all common Cretaceous species become extinct at the Cretaceous - Tertiary boundary. Watznauezia baznesae is the most common reworked species and is present in most samples. Diverse and often quite well preserved assemblages are present in some samples (eg Tables 6 & 7), a number of species in these assemblages are indicative of late or latest Cretaceous age, including Azkhangelskiella cymbiformis, Cezatolithoides aculeus, Micula stauzophoza, and Quadrum tzifidum (a couple of specimens only seen, not on tables). The entire assemblage could have come from Campanian sediments.

122 FIGURE 33 - NANNOFOSSIL ZONATION SCHEME FOR MAKRAN MATERIAL. Chart showing: the relation of the assemblages zones used for the Makran samples to the standard nannofossil zonation of MARTINI & WORSLEY (1970); the ranges of the various species used in making zonal assignments; and the succession of reticulofenestrid assemblages. ABBREVIATIONS: C.lept - Calcidiscus leptoporus; H.cart - Helicosphaeza carteri; G.rotl - Geminilithella zotula; S.belm - Sphenolithus b e le m n o s; S.hetr - S.heteromorphosus; D.defl - Discoaster d e f l a n drei; D.kugl - D.kuglezi; D.ham - D.hamatus; C.coal - C a t i n a s t e z coalitus; D.calc - D . c a l c a z i s ; D.prep - D.pzepentazadiatus; D.var - D.vaziabilis; D.pent - D.pentazadia'tus; D.surc - D.surculus; D.qnq - D.quinquezamus; D.brou - D.bzouwezi; S.abies - Sphenolithus a b i e s; S.mrf - S.moriformis; C.pel - Coccolithus pelagicus.

122A 9.3.4 Results. The results of the biostratigraphic work are given in Tables 1 to 7, with additional notes in the section below. The tables are organised so that the samples are in probable stratigraphical sequence, with the oldest samples on the left, except where separate sections overlap. Species are separated into probably in-situ Neogene and the reworked groups. The Neogene species are further subdivided into families (Coccolithaceae, Prinsiaceae, Helicosphaeraceae, Sphenolithaceae, and Discoasteraceae). N.B. (1) Taxonomic references for, and notes on, the reworked species can be found in PERCH-NIELSEN (1985a,b). (2) This work was done with provisional taxonomic concepts, and the Reticulofenestra group was subdivided at 7 and 4 microns, rather than 5 and 3 microns. Symbols: * - Rare (only one or two specimens found); ** - Common; *** - Abundant (>15% of assemblage, and more than twenty specimens found). ? - uncertain identification; b - bridged variety of C. pelagicus.

1. Branguli area (Table 1, Pigs.34-35). This area has particularly good sections through the lower units of the Coastal Makran and shows well the structural style of the region. It was studied and mapped in detail during the 1983 and 1984 field seasons. The samples come from scattered localities across the area (Fig.34), but have been organised into a single likely sequence in Table 1. This seems reasonable since they come from a small area with simple 1ithostrat igraphy. Most of the samples contained reasonable nannofossil assemblages. In particular better results were obtained from the Panjgur turbidites here than anywhere else. These are still difficult to date owing to the absence of nannofossils with short ranges; but they have, Zone II type assemblages. The absence of large specimens of C.floridanus 0 5 microns), and the common presence of large Reticulofenestra makes it unlikely that they are older than NN6. The minimum age (NN7) is similarly based mainly on deductions from negative evidence, the absence of species such as C.coalitus and D.hamatus, despite the common occurrence of D.exilis. The basal Branguli assemblages (Iocs. B35 and B26) are extremely similar, even the reworked components. This similarity is in accord with the field interpretation of a transitional contact. Samples from the higher Branguli and the Parkini above it have Zone III assemblages, with typical marker species. The Parkini in the south of the area (samples B33 and R51), however, has rather different assemblages without only small to medium Reticulofenestra and with large well formed D.bzouvezif D.suzculus and D.variabilis in place of the D.exilis group. These changes, even without D.guingueramus are

123 diagnostic of Assemblage IV. The southern fault thus causes significant stratigraphical repetition (Figure 35).

2. Pasni Traverse (Table 2, Fig.36). During the 1983 field season a north - south transect was made through the west of the field area, the Pasni Traverse (Figure 36). Samples were collected throughout, but only those from the area around and to the south of the Kulanch syncline were fossiliferous. This part of the section mainly runs through Parkini Facies sediments but is disturbed in the centre by a complex anticlinal ridge of Panjgur turbidites. Host samples came from around this feature, Talo Koh. On Table 2 samples from north and south of Talo Koh are separated. The stratigraphy is similar to that of the Branguli area, although greater structural complexity and worse nannofossils make uncertainty greater. The Panjgur again has Assemblage II nannofossils. There is no intermediate facies but the Parkini has samples with Assemblages II and III and so the base of it is probably correctable with the Branguli Facies of the Branguli area. To the south Parkini Facies samples have Zone IV assemblages, with D.guingueramus. The north end of the section crosses the nose of the Kulanch syncline. The Talar Facies here grade rapidly into the Parkini and are only present on the northern limb. Zone III assemblages occur in both the Talar (PG64) and associated Parkini Facies (PG4, 57).

3. Basol Traverse (Table 3, Fig.31). This traverse was completed in the e

4. Miscellaneous Parkini and Panjgur Facies samples (Table 4, Fig.31). Table 4 includes samples from three localities south of the Kulanch syncline (Kumbi Koh, Ras Juddi, and Kibr Koh), and from two localities further inland. The southern samples provide support for the

124 FIGURE 34 - MAP OF BRANGULI AREA. Redrawn from map of JPP, based on field mapping during 1983 and 1984 field seasons. Location indicated on Figure 31. The succession is repeated by thrusts. Samples localities are numbered, and arrows indicate their projection along strike to sections A-B and C-D (Fig.35). Abbreviations: Pk - Parkini Facies. Bg - Branguli Facies. Pg - Panjgur Facies (stippled).

125 (

FIGURE 35 - SECTIONS THROUGH THE BRANGULI AREA. Lines of section on Figure 31, no vertical exaggeration. Redrawn from sections of J.P.Platt. Sample localities and zonal assignments shown. 1 2 7

Location indicated on Figure 31. No vertical exaggeration. Redrawn from sections o f J K y J.P.Platt, a composite section through the Talo Koh structure is given in PLATT et al (1985). Location and zonal assignments of samples with useable nannofossil assemblages shown. biostratigraphic pattern shown by the Pasni and Branguli areas. Upper Parkini sediments at Kumbi Koh and Ras Juddi have clear Zone IV assemblages (note absence of large Reticulofenestza as well as presence of D.guingueramus). Kibr Koh is the continuation of Talo Koh and the samples show similar assemblages to those of the Talo Koh Panjgur, but since discoasters are very rare it is not possible to confidently assign them to either Zone II or III. Sample 16/3 which was from the North part of the Pasni Traverse is similar. The last two samples were both collected by S.Alara during HDIP fieldwork in 1982 from localities north of the field area shown in Figure 32 (precise locality information is not yet available for the HDIP samples). Both samples have assemblages dominated by Cyclicargolithus flozidanus with only rare Reticulofenestza, and so are plainly from Zone I or earlier (this approximately means Oligocene or Early Miocene). Sample S/B14, which was from a structurally deep level also contains common, and unambiguous Sphenolithus hetezomozphosus and so is definitely assignable to Zone I. Sample S/KB10 is more ambiguous since it does not contain S.heteromorphosus, but neither does it contain any of the species characteristic of the Late Oligocene in the Indian Ocean samples, such as the S.distentus group, Clausicoccus spp., and D.deflandzei is too abundant for it to be any older. Hence the assemblage is probably from NN1 or NN2, (ie earliest Miocene), or conceivably basal NN6.

5. Kulanch Syncline (Table 5). The Kulanch syncline is a large asymmetrical syncline, of the type shown in Figure 31. In the field area (Fig.32) it is mainly formed of inner shelf facies sediments. These were extensively sampled but were virtually all barren, doubtless mainly due to original exclusion of nannofossils from such shallow water environments. A single sample from near the base of the succession (PG64, Table 2), gave a Zone III assemblage. Parkini Facies sediments in the Kumbi Koh area (Table 2) which must be laterally equivalent to the Talar Facies of the north limb give Zone IV assemblages. So too does a single sample from the top of this sequence (3/1, Table 5). None of this is very sound data but it all indicates a similar age for the main sand sedimentation, Zone IV - ie Late Miocene, and synchronous with Parkini sedimentation to the south. This sand sedimentation was followed by outer shelf mud-dominated cycles, termed the Chatti Mudstone. The change to mud-dominated deposition may have been due to sea-level rise or to local sand starvation. The Chatti and Kappar sections (Table 5) are from this unit, to the west of the main field area. They both give rather indeterminate assemblages of probable Early Pliocene age, with the

128 transition to Late Pliocene occurring at the top of the Chatti Section. The relatively common occurrence of Helicosphaeza, and the rarity of discoasters in these samples (Table 5), are typical of near-shore assemblages.

6. Hinglaj Syncline (Table 6). In the Eastern part of the area the main Talar Facies outcrops are in the Hinglaj syncline, with an outlier in the Garuki area (Fig.31). The first samples of the Basol Traverse are from near the base of this succession and indicate a Zone IV age. The assemblages from the Hingol sections (off Fig.3t to the east) are rather unreliable because of extensive reworking in them. They suggest mainly early Pliocene age, and so that the sequence here is laterally equivalent to the Chatti Mudstone of the Kulanch syncline. The Garuki outlier probably is equivalent to the top of the main succession (cf. Fig.29). The assemblages here appear to have come from near the Early / Late Pliocene boundary; good Zone V assemblage in 0/10, Pseudoemiliania lacunosa in 0/3.

7. Mud Volcanoes (Tables 7, 8). Groups of mud volcanoes occur at various places along the Makran coast (Fig.29). Within the main field area there is a group to the north of the Ormara headland (Fig.31). This group occurs on alluvial sediments covering a broad anticline between the Hinglaj syncline and the Garuki outlier. A suite of samples was collected from them in order to try and obtain extra information on the sources of the mud, and on the causes of the diapirism. Most samples contained nannofossils, with variable preservation and abundance, results from the best nine samples are given in Table 7. All the samples have similar nannofloras dominated by small sphenoliths and small-medium Reticulofenestza which suggests Zones IV or V. The presence of D.guingueramus indicates Zone IV, but Zone V material could be included as well. There is no evidence of Middle Miocene or Late Pliocene, and the common Late Cretaceous and Early Miocene are probably reworked elements (compare with the Hinglaj assemblages, Table 6). Microfossils are also present in many samples. The material disaggregates easily but is very pure mud, so large samples are needed to obtain useful sized residues. The microfauna (Table %) is diverse containing most elements identified in normal samples other than indicators of the deepest conditions (various lituolids), or the shallowest water (scaphopods, gastropods). Some mixing has doubtless occurred but overall an upper-slope to mid-shelf sequence is suggested. The microfaunal and nannofloral evidence point to Late Miocene /

129 Early Pliocene age Parkin! Facies sediments as the source of the mud. Early Pliocene Parkin! sediments almost certainly occur directly beneath the mud volcanoes (Fig.32) so a shallow source seems likely for the volcanoes. This is rather different to the deep sourcing suggested for mud volcanoes in other accretionary terranes by BARBER et al (1986). It is, however, supported by an association along the Hakran coast of mud volcanoes with outcrops of Pleistocene sediments. The latter almost certainly occur along the edge of active shelf basins, of the type shown in Figure 31. Overpressuring of shales beneath these basins is the roost likely cause of the mud vulcanism.

9.4 DISCUSSION 9.4.1 Further micropalaeontological work. The most important results of biostratigraphic work in the Makran have been establishment of a broad chronology of events and correlation of the various stratigraphical units. Attempts to use biostratigraphy on a smaller scale, to resolve local stratigraphical and structural problems, have not been very successful. This is in part the result of poor preservation and sporadic occurrence, but the potential of biostratigraphy is severely restricted by the high sedimentation rates (100 - >1000mMa_1:). Future sampling for biostratigraphic work should concentrate on the collection of detailed sequences of samples from structurally simple areas, in order to compare successions in different parts of the Coastal Makran. Notwithstanding the problems encountered, calcareous nannofossils are probably the most useful group to use. Siliceous microfossils are not present, and palynomorphs are very unpromising (Dr.P.R.Grant of Imperial College examined several sets of representative samples). Planktonic foraminifera might be as useful as nannofossils in the Parkini Facies, but they would be difficult to extract from the well indurated older deposits, and the Talar Facies is too shallow for most genera (Table 8). The very limited reconnaissance work I have done suggests that a study of the benthic microfauna could make a useful contribution to overall understanding of the sedimentology of the various facies. If combined with study of modern offshore faunas this would make a very interesting project.

9.4.2 Reworking. As described above and shown on the tables, unambiguously reworked nannofossils are abundant and can be divided into three groups, Early Miocene, Palaeogene, and Late Cretaceous. The Early Miocene assemblages occur in virtually all samples, and

130 certainly from all areas. This strongly suggests that Lover Miocene sediments outcropped extensively in the source area for the sediments. In the Parkini Facies and the 1 Talar Facies of the Kulanch Syncline, palaeocurrents (our data), and heavy mineral assemblages (HARMS et al 1984), both suggest derivation from the north. The Early Miocene turbidites that outcrop in the Makran Ranges only have sparse nannofossil assemblages, but it is likely that Parkini Facies sediments originally also occurred. Erosion of these during the Middle Miocene to Early Pliocene may have provided the Early Miocene nannofossils. The Upper Cretaceous and Palaeogene assemblages show very similar distribution patterns, with the Palaeogene species much rarer and only occurring in samples which also contain good Cretaceous assemblages. In the Panjgur, Branguli and lower Parkini they are quite common, and also occur sporadically in the higher Parkini. In the Talar Facies (Tables 5 and 6) there is a very clear trend of increasing abundance toward the east. They are most common in the Hingol sections, noticeably less so in the Garuki section, rare in the Kappar section and absent in the Chatti section. The simplest explanation is that the nannofossils were derived from a source to the east, as shown in Figure 36. There are extensive outcrops of suitable age in this area; and Palaeogene nannofossils were described from here by HAQ (1971a). A source for this material from outside the Makran is also necessary since the only Cretaceous rocks known in the Makran are the Humai Group of the Chagai Hills, and the coloured melanges of the Iranian Makran (Fig.29), both of which are unlikely sources.

9.4.3 Stratigraphical synthesis An interpretation of the chronostratigraphy of the Coastal Makran based on the data discussed above is given in Figured. The exposed record starts with mid-Miocene basin plain turbidites, although earlier turbidites are exposed further north. In the late Miocene rapid shallowing occurred due to progradation of the slope facies Parkini sediments. In the latest Miocene and Pliocene shelf basins with thick shallow marine fills and rapid lateral facies variations developed on the slope sediments. There is little evidence of extensional tectonics in these basins and they probably developed mainly by isostatic subsidence. Later Pliocene sediments are not present in the field area and the main tectonic compression probably occurred at this time. It affects all previously deposited units. Recent uplift has exposed Pleistocene sediments along the coast (Ormara and Jiwani Formations on Figs.32 & 37). The Coastal Makran thus represents a rather coherent tectonic slice

131 of the accretionary prism - recording development of a Late Miocene to Early Pliocene shelf-slope system, and subsequent deformation of it in the Late Pliocene. To the north are the roots of an older, largely Early Miocene, terrane that formed a major sediment source for the Coastal Makran units. To the south is an undeformed shelf-slope system formed during the Late Pliocene and Pleistocene.

9.4.4 Continuity of processes. Interpretation of the geology of the Makran has been dominated by actualistic models assuming continuity of processes throughout at least the Neogene (eg FARHOUDI & KARIG 1977, HARMS et al 1984, WHITE & LOUDEN 1982). This has been very successful in terms of facies interpretation but is less convincing as a model for the tectonic history since most deformation in the Coastal Makran appears to have occurred during a single Pliocene event (HUNTING 1960, AHMED 1969, PLATT et al 1985). Also there are problems in extrapolating modern subduction rates of 3-5cm per annum back through the Neogene. This would imply subduction of around 500km of oceanic crust since the Middle Miocene and it is not clear where the overlying sedimentary sequence (7km thick in the Gulf of Oman) can be accommodated in the accretionary prism (WHITE & LOUDEN 1982, PLATT et al 1985). Underplating of this material is one possibility (PLATT et al 1985), an alternative is that subduction rates have not been continuous and that the present high rates are a relatively recent phenomenon. This could also explain the discontinuous record of activity in the igneous terrane. Vulcanism there is of predominantly Quaternary, possibly Late Pliocene age (GANSSER 1971, GIROD & CONRAD 1975, ARTHURTON et al 1982). Intrusive activity appears to have ended in the Early Miocene (BREITZMAN 1983), so there is a long interval during which there is no evidence for subduction related magmatism. Taking a broader perspective a similar discontinuous tectonic pattern is also seen around other parts of the Arabian Plate (Fig.28, p.110). The timing of deformation and Tethyan closure in the Zagros is not clearly established, but the igneous record is similar to that of the Makran (BREITZMAN 1983). Discontinuous movement along the Levantine Shear is well documented, with approximately 60km of pre Mid-Miocene movement and 40km of Pliocene to Recent movement (QUENNELL 1959, FREUND et al 1968). GIRDLER & STYLES (1974, 1978) suggested a similar two-phase history for the opening of the Red Sea and the Gulf of Aden (which is supported by the biostratigraphy of the Red Sea Sites). The nature and history of the Owen - Murray ridge is essentially unknown, but WHITE (in press) has described extensional faults on the Murray Ridge which suggested to him that a change in tectonic polarity had

132 occurred. Overall thus there is substantial evidence for important changes in the regional tectonic situation during the Neogene. Hence modern conditions should not be extrapolated uncritically beyond the Pliocene. This should probably be considered in refining existing models of the stratigraphy of the Makran.

133 FIGURE 37 (top)- STRATIGRAPHIC SUMMARY. Schematic representation of the stratigraphy of the coastal Makran based primarily on data discussed here. Only East - West diachronism is shown, so this scheme not applicable outside the coastal area. Dotted vertical lines indicate sections: Ch - Chatti; Kp - Kappar; Ps - Pasni; Bg - Branguli; Gk - Garuki; Bs - Basol; Hg - Hingol. Ormara and Jiwani Formations have been dated by molluscs (HUNTING’S 1960, CRAME 1984), and are laterally continuous with modern shelf basins.

FIGURE 38 (bottom) - SEDIMENT SOURCES DURING THE LATE MIOCENE - EARLY PLIOCENE. Based on distribution of reworked nannofossils, and on palaeocurrent data.

134 BRAN6ULI AREA FACIES r d M L in i — LOCATION RSI B33 B33 R23 R71 RiS B31 84 R27 B26 B35 B3S R42 R57 B34 B34 B34 R55 R8 R43 SAMPLE R2B B16 R6 R17 R70 R12 B13 B1 R21 B8 B20 RIO R30 R56 R9 R7 B17 R SI R4 R24 ASSEMBLAGE IV IV IVIII IIIIII IIIIII III II IIIIIIIIIIIIII IIIIII SPECIES Calcidiscus leptoporus gp. t t t t t Coccolithus pelagicus t t t t t t t t t t t t t t t t t tt tt tt t ttttt tt t t t t t t t t t C.aiopelagicus t t GeainiIithel1 a rotula t t t Reticulofenestra large t i t t t t t t t t tt tt tt ttt tt t t t t t t tt tt tt t tt tt Reticulofenestra aediua i t t t t tt ttt ttt ttt ttt ttt ttt ttt ttt ttt tt tt ttt t t t t t t t t t t t t t t Reticulofenestra saall t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t

Helicosphaera carteri gp. t t t t t t t t t t t t t t t t t t t H. in te r aedia t

S.abies/aori forais t t t t t t t t t t t t t t t t t t t t i t t t t t t t t S.neoabies/coapactus ttt ttt ttt tt ttt ttt tt t t t t t t tt tt ttt ttt tt tt it t t

Catinaster coalitus t t t t Discoaster brouveri t t D.cf.brouveri t ttttt tt tt tt tt tt ttttt tit t D . c a lc a r is t D . e x i l i s t t t t t t t t t t t t t t t t t D.cf.exilis tt tt tt tt tt tt ttttt t t t t t t t t D .haaatus t t D .k u g le ri t D.prepentaradiatus t D.quinqueraaus D .su rc u lu s t D.variabilis t t Unidentifiable 5 rayed discos t t t t Unidentifiable 8 rayed discos t t t t t t t t t t t t t t t t tt tt it it tt t t

Reworked Early Miocene Cyclicargolithus floridanus t t t t t C.floridanus - stall t t t t t t t t t t t D.deflandrei t ttttttt ttt t t S.heteroao*phosus t t t t t t t t t t t t t t t t t

Reworked Palaeoqene Clausicoccus spp. t Coccolithus foraosus t t t t t Dictyococcites scissura t Discoaster barbadiensis gp. t t t t t t Helicosphaera-ovoid t t Neochiastozygus sp. t Sphenolithus radians t

Reworked Cretaceous Arkhangeliskiella cyabiforais t Ceratolithoides aculeus i t t t Cretarhabdus spp. t t t t t t t t t Cribrosphaera ehrenbergii t t t t Eiffellithus spp. t t t Microrhabdulus decorus t t Micula staurophora t t t t t t t Parhabdolithus etbergeri t t t Prediscosphaera cretacea t t t t Uatznaueria barnesae t ttt tttttt ttt ttt i t ttt t tt tt t itt Zygodiscus spp. t t Other Cretaceous spp. t t I t t t t t TABLE 1 (See Section 9.3.4, p .123)

134A n i • * PASNI TRAVERSE FACIES Parkini S.- -Pk. N.~ Tr. Parkini N.-- SAMPLES P616 PG17 P611 6/1 8/13 8/3 8/4 8/15 PG57 PG4 PG64 8/12 P618 PC7 8/9 8/1 6/4 6/5 8/18 8/14 8/17 PC8 8/23 ASSEMBLAGE IV IV IIIIIIIIIIIIIII III IIIIIIIII III in II II II II II / SPECIES Calcidiscus leptoporus gp. * « i i f i i i i Coccolithus pelagicus it f i if if if i i i i if i i i i it it it it if it it if C.aiopelagicus Geainilithella rotula t i Reticulofenestra large t if f if f i i if i i i i i i i i if i i i Reticulofenestra aediua m if f if f i i i f i i i f f if ft i i ti iii it tit ii it Reticulofenestra saall ftiff ii i iii i f it

Helicosphaera carteri gp. t if if i i i i i i i i i i i t i i i i H.interaedia

S.abies/aoriforais t i if i i i f tt i i i i t i • i i i f i i S.neoabies/coapactus i t i i i i if f if f i i f i i i i i i i i i i i if f i i i i i

Catinaster coalitus i Oiscoaster brouveri t i i i i i i D.cf.brouweri ifififiiif i i t • i i t i i i i i D.calcaris i i i i ? i ? i ? D.exilis i t t i t i i i i t D.cf.exilis i i i i i i i i i D.haaatus i i i ? D.kugleri ? D.prepentaradiatus i D.quinqueraaus if if D.surculus if ? D.variabilis i Unidentifiable 5 rayed discos i i i i i i t i i t i Unidentifiable 6 rayed discos i i i f i i if i i t i it

Reworked Early Miocene Cydicargolithus floridanus i t i f i i i i i C.floridanus - saall i i i i i D.deflandrei i if i i i i i ♦ i i S.heteroaophosus if i i i i i i i i

Reworked Palaeoaene Clausicoccus spp. Coccolithus foraosus t Dictyococcites scissura i i Discoaster barbadiensis gp. i i Helicosphaera-ovpM Neochiastozygus sp. i Sphenolithus radians i

Reworked Cretaceous Arkhangel'skiella cyabiforais Ceratolithoides aculeus i Cretarhabdus spp. i Cribrosphaera ehrenbergii i Ei f fellithus spp. Microrhabdulus decorus Nicula staurophora i i i i Parhabdolithus eabergeri Prediscosphaera cretacea Uatznaueria barnesae i i i i i i i i i Zygodiscus spp. Other Cretaceous spp. f i

TABLE 2 (See Section 9.3.4, p.124)

135 BASOL TRAVERSE ZONE ------Outer Zone ------Inner Zone ------SAMPLES 26/1-5 26/6-7 28/5-7 29/1-5 30/13 30/12 33/1 33/4-5 33/12 FACIES Tr/Pk Parkini ------Branguli / Panjgur ------Panjgur ---- ASSEMBLAGE IV - III III/II - III/II I - I SPECIES______Calcidiscus leptoporus gp. t ft Coccolithus pelagicus ft t t t t ttt C.aiopelagicus Geainilithella rotula Reticulofenestra large t ft t ttt t Reticulofenestra iediu« tt t t t tt tt t i Reticulofenestra saal1 tt t *

Helicosphaera carteri gp. t t 9 H.interaedia

S.abies/aoriforais tt t t tt t tt it S.neoabies/coapactus tt t t t t t *

Catinaster coal itus * Discoaster brouveri D.cf.brouweri t D.calcaris D.exilis t D.cf.exilis t D.haaatus D.kugleri D.pentaradiatus t D.quinqueraaus * D.surculus D.variabilis Unidentifiable 5-rayed discos ♦ Unidentifiable 6-rayed discos t ft t

Early Miocene Cyclicargolithus floridanus * * t tt t C. floridanus - saall t D. deflandrei * t t S.heteroaorphosus * it t

Reworked Palaeogene Clausicoccus spp. Coccolithus foraosus Dictyococcites scissura Discoaster barbadiensis gp. Helicosphaera-ovoid Reticulofenestra reticulata t Sphenolithus radians

Reworked Cretaceous Arkhangellskiel1 a cyabiforais Ceratolithoides aculeus Cretarhabdus spp. { t Cribrosphaera ehrenbergii Eiffell!thus spp. Microrhabdulus decorus Micula staurophora tt Parhabdolithus eabergeri Prediscosphaera cretacea Watznaueria barnesae tt i f f t t Other Cretaceous spp. i t

TABLE 3 (See Section 9.3.4, p .12 4 )

136 MISCELLANEOUS AREAMilui Kon rtua juuui Kibr Koh Pasni Tr Central Ranges SAMPLES FACIES — Parkini (top) — — Parkini — Pk/Pg Pg Pg // SAMPLE L4/1 L4/3 L5/1 KbK2 RJ7 RJ1 RJ3 KK1 KK2 16/3 S/B14 S/KB10 ASSEMBLAGE IV IV IV IV IV / IV ?II ?II ?II I <=I SPECIES Calcidiscus leptoporus gp. a a a a a Coccolithus pelagicus H aa aa aa aa a a aaa aaa aa aa aaa C.aiopelagicus a Geiinilithella rotula Reticulofenestra large a aaa aaa aaa Reticulofenestra aediua It a aaa aaa aa a aaa aa aa aa a Reticulofenestra stall m aaa a aa aaa aa aa a aa

Helicosphaera carteri gp. aa aaa a aa ? H.intertedia

S.abies/toriforais a a a aa a aaa aa aa a aaa aaa aaa S.neoabies/coapactus aaa aaa aa aa aaa aaa aaa a a

Catinaster coalitus Discoaster brouweri a a D.cf.brouweri a D.calcaris D.exilis D.cf.exilis a a O.haaatus D.kugleri D.pentaradiatus a D.quinqueraaus a aa a a aa D.surculus a a D.variabilis a a a Unidentifiable 5-rayed discos a a Unidentifiable 6-rayed discos a a a aa a a a

Early Miocene Cyclicargolithus floridanus a a a aaa aaa C.floridanus - stall a a aa D.deflandrei a a a a a aaa aa S.heteroaorphosus a a a aa

Reworked Palaeoaene Clausicoccus spp. Coccolithus foraosus a Dictyococcites scissura Discoaster barbadiensis gp. a a a Helicosphaera -ovoid ' ' ' ' a Reticulofenestra daviesi a Sphenolithus radians

Reworked Cretaceous Arkhangel1 ski el la cyabi forais Ceratolithoides aculeus a Cretarhabdus spp. a a Cribrosphaera ehrenbergii Eiffelfithus spp. Microrhabdulus decorus Micula staurophora a a Parhabdolithus eabergeri Prediscosphaera cretacea Watznaueria barnesae a a a a aaa Other Cretaceous spp. a

TABLE 4 (See Section 9.3.4, p.124)

137 TAII ALnK AD rAPTCC 1 nl/lLu ^nlCtn ADCA i/ndbbi aeiviuu Baswaja KUIANCH SYNCLINE SAMPLE SNJ9 SNC42 SNC23 SNC9 SNC1 S1S4 S132 S135 S145 S141 S12S S124 3/1 ASSEMBLAGE / VI V VVV V VV VV V?IV SPECIES Calcidiscus leptoporus gp. * ft ft ft ft* ft Coccolithus pelagicus i *b ftft »ftb ft* oftb ** ** t » Geiini1ithella rotula ft ft ft * ft Uibilicosphaera jafari *

6ephyrocapsa - sia ll ftft ? Pseudoeiiliania lacunosa ftft ? Reticulofenestra large « ft ftft ft ft ft* ** Reticulofenestra aediua ftft ft» » Oft *** ** >* *** *> **» t(ft *> Reticulofenestra stall ft ftftft ftftft ftftft fto *** ** *** **> ft** ft** ** **

Helicosphaera carteri gp. ftft ft >ft« ftt »«> ** > ( »«* *» > Pontosphaera spp.

S.abies/neobies ft ftftft ««« ftftft * «** ft* ft* *** *** *«« ** S.cf.iorifonis ftftft >ft » *« * * »* ** ******

Discoaster asyiietricus ? D.brouweri » » D.pentaradiatus D.quinqueraaus ? D.surculus D.cf.variabilis Unidentifiable 6-rayed discos ft ft * ******

Reworked Early Miocene Cyc1icargolithus floridanus ft ftft ft ft ? ft* ** ft ft* ** * *** C.floridanus - stall * > D.deflandrei ft * « * ft* *» S.heteroaorphosus » > > *» > *> *

Reworked Palaeogene Chiasaolithus spp. Clausicoccus spp. Coccolithus foraosus Cruciplacolithus spp. D.barbadensis D.saipanensis/lodoensis Dictyococcites scissura Helicosphaera -ovottt Reticulofenestra reticulata Sphenolithus radians S.distentus gp.

Reworked Cretaceous Arkhangel 1 skiella cyabiforais Ceratolithoides aculeus Chiastozygus spp. Cretarhabdus spp. Cribrosphaera ehrenbergii Eiffellithus turriseiffelii Gartner ago obliquui Hicula staurophora Parhabdolithus esbergeri Prediscosphaera cretacea Watznaueria barnesae « *» h Zyqodiscus spp. Other Cretaceous spp. * * TABLE 5 (See Section 9.3.4, p .128) TALAR FACIES AREA — Garuki ______— — Hingol 1 - Hingol 2 - HINGLAJ SYNCLINE SAMPLE 0/3 0/6 0/10 XP/11 XP/10 A/l A/7 B/12 H/8 H/30 H/27 D/16 ASSEMBLAGE VI VI V VV V V IV V V V?IV SPECIES Calcidiscus leptoporus gp. i i i i i i Coccolithus pelagicus * if i i i i i i i i i i i i i i i iib «b Geiinilithella rotula i i Utbilicosphaera jafari i i

Gephyrocapsa - stall * Pseudoeailiania iacunosa i i Reticulofenestra large i i i i i i i i i Reticulofenestra lediui H it iii iii ii ii ii i i i i i i i i Reticulofenestra stall If f i i i i i i i i i i i i i i i i i i i i i i i

Helicosphaera carteri gp. H i i i i i i i i i i i i i i i i Pontosphaera spp. i i i i

S.abies/neobies if i i i i i i i i i i i i i i i i i i i i i i S.cf.iorifortis i i i i i i i i i i i i i i i i i

Discoaster asyitetricus D.brouweri i i i O.pentaradiatus i i D.quinqueratus i i D.surculus i i D.cf.variabilis i i Unidentifiable G-rayed discos i i i i i i i

Reworked Early Miocene Cydicargolithus floridanus i i i i i i i i i i i i i i i i C.floridanus - stall i i i D.deflandrei i i i i i i i i i S.heterotophosus i i i i i i

Reworked Palaeogene Chiastolithus spp. i i Clausicoccus spp. i i Coccolithus fortosus i i i i i i Cruciplacolithus spp. i D.barbadensis i i i i D.saipanensis/lodoensis i i i Dictyococcites scissura i i i i i i Helicosphaera-owW , Reticulofenestra reticulata i i i Sphenolithus radians i S.distentus gp. i

Reworked Cretaceous Arkhangellskiella cytbifortis i i i i Ceratolithoides aculeus i i Chiastozygus spp. i i i i Cretarhabdus spp. if i i i i i i i i i i i i i i Cribrosphaera ehrenbergil i i i i i i Eiffellithus turriseiffelii i i i i i Gartnerago obliquut i Nicula staurophora i i i i i i i i i i i i Parhabdolithus etbergeri i i i Prediscosphaera cretacea i i i i i i i i i i i i Uatznaueria barnesae if i i i i i i i i i i i i i i i i i i i i i i i i i 1 i i Zygodiscus spp. i i i Other Cretaceous spp. i i i i i i i i

TABLE 6 (See Section 9.3.4, p .129)

139 HUD VOLCANO SAMPLES AREA ----- Mount Zia ...... r ast Field SAMPLE M.V.'83 M.V.3 H.V.G H.V.7 M.V.10 M.V.12 H.V.14 M.V.24 M.V.31 SPECIES Calcidiscus leptoporus gp. a ft ftft « ftft ft ft Coccolithus pelagicus H ftftft ftft ftft <1 ftft ftft ftft ftt Geainilithella rotula ft Uibilicosphaera jafari (

Gephyrocapsa - stall Pseudoetiliania lacunosa Reticulofenestra large ft «< n Reticulofenestra tediut if t» i d lit m lift ftftft ftftft tit Reticulofenestra stall ft* ftft ft « ft* « «ft n it

Helicosphaera carteri gp. > » ft n t n »ft ft H.intertedia ft

S.abies/neobies ftft* m i n m m m i n m tit S.cf.torifortis ftft «« ft n ftft «ft ft

Discoaster asytietricus ? D.brouveri ft ft ft ft ft ft D.pentaradiatus ft D.quinqueratus i ft ftft ftft ft « ft ftft ft D.surculus » ft » > ft D.cf.variabilis t ft ft ft Unidentifiable G-rayed discos It ft ft n »» ftft ftt (ft o

Reworked Earlv Miocene Cydicargolithus floridanus ft «ft * ftft ftft ft C.floridanus - stall 1 ft ft (ft ft ft (ft D.deflandrei t » n ft ft ft ft S.heterotorphosus 1 ft ( < ft tft ft ft ft(

Reworked Palaeoqene Chiastolithus spp. Clausicoccus spp. Coccolithus foriosus ft ft ft ft Cruciplacolithus spp. ( Dictyococcites scissura ft Discoaster barbadiensis gp. ft ft ft > D.saipanensis/lodoensis Helicosphaera ovoid Reticulofenestra daviesii ft ft Sphenolithus radians S.distentus gp.

Reworked Cretaceous Arkhangellskiella cytbifonis « ft Ceratolithoides aculeus ft ft ft Chiastozygus spp. ft ft Cretarhabdus spp. ftft ft ft ftft ftft ftft ftft ft ft Cribrosphaera ehrenbergii ftft ft ft ft Eiffellithus spp. ft ft ft ft ft Gartnerago obliquut Nicula staurophora ft ft ft ft ft ft ft (ft Parhabdolithus etbergeri ft > ft ft ft ft < Prediscosphaera cretacea ftft ft ft ft (ft • » « Uatznaueria barnesae ftft ftft ft* ftftft ftftft ftft ftft ft» >ft Zygodiscus fc Vekshinella spp. ft ft ft Other Cretaceous spp. ft ft ft ftft ft >ft >ft

TABLE 7 (See Section 9.3.4, p.129)

140 MICROFOSSILS FACIES Pg Bg/Pk Parkini - - Pk/Tr - - Talar -- — Hud Volcano Saapl es - SAMPLES 6/4 R18 P617 KbK2 KbK3 KbK4 KbK6 3/2 3/1 0/10 MV1 HV3 HV7 HV10 HV13 HV24 HV'83

Hacrofossil fraaaents Fish teeth t t t t Echinoids t t t t Gastropods t Bivalves t t t t t t Scaphopods t t t Bryozoa t t t t t

Ostracods Miscellaneous ttt ttt ttt tt ttt tt tt tt tt Bairdia t

Planktonic foraainifera Globigerina (s.l.) tt tt ttt tt tt tt t tt ttt t tt tt ttt tt Globoquadrina tit tt t t t Globorotalia t t t t t t Orbulina tt tt t t tt tt tt t t t

Rotaliacea, and aisc. Aaaonia ex gr.becarii ttt ttt ttt ttt tt ttt tt tt ttt ttt t tt Cibicides gp. tt it tt t t ttt t tt ttt ttt ttt Cassidulina t t t Elphidiua tt tt tt t tt t t Gyroidina tt t t Planulina t t Pullenia + Helonis t tt t

Buliainacea Bolivina t t t t t tt t t tt tt tt Bolivina sp A t t tt tt tt Buliaina cf.affinis ttt ttt tt ttt t t t t t tt t Buliaina striata t t t Ehrenbergina t t t Stilostoaella cf annulifera tt tt Stilostoaella spp. t t tt tt Uvigerina t tt t t t t tt

Mi 1iolids Pyrgo t tt t t t Quinqueloculina t t tt t t tt

Lituolids Aaaobaculites t t t Aaaodiscus t Cyclaaaina tt Cylindroclavulina t t Reophax tt ttt t t t Trochaaaina tt tt t tt t t

TABLE 8. MICROFOSSILS. Results of reconnaissance survey of various samples, arranged in likely stratigraphic sequence, but oldest on left. This is approximately equivalent to depth sequence. Locations: R18, Branguli area; 6/4 and PG17, Pasni traverse; KbK2-6, Kumbi Koh Area; 3/1,2, Kulanch syncline (N.limb); 0/10, Garuki; Mud Volcanoes - Ormara area. Specimens only assigned to broad taxonomic groups. The lower four groups are all benthic foraminifera.

141 I? ART O

SYSTEMATIC DESCRIETIOM OE

NEOGENE 1ST A 1ST NOEOSS I L S

142 CHAPTER 10 - INTRODUCTION TO THE SYSTEMATIC SECTION

10.1 TAXONOMIC CONVENTIONS ADOPTED The systematic part of the thesis is organised into chapters dealing with each of the principal family level groups of Neogene annofossils. I adopted this format in order to give a coherent and economical treatment of the structure, evolution and biostratigraphic use of each group. Taxa not included in these chapters are briefly discussed at the end of this chapter. The most important taxonomic innovation is the extensive use of trinomial combinations, with many conventionally recognised species being recombined as varieties, as discussed below. In order to balance the effect of this, I have been conservative in other respects. In particular I have, as far as possible, used well established generic combinations. Similarly, the family level classification is based, with minimal modification, on that of PERCH-NIELSEN (1985b). In each chapter the taxa are arranged in alphabetical order of genera and species. I have given synonymies for all species, since I find them useful. They include important name changes and junior synonyms, and for most species also list additional good quality illustrations. I used the following papers very extensively as sources of illustrations and have consistently cited them, MULLER (1974a), PERCH-NIELSEN (1977), STEINMETZ & STRADNER (1984); also but less extensively PERCH-NIELSEN (1972), ROTH (1973), STRADNER (1973), PROTO-DECIMA et al (1978), SAMTLEBEN (1979), BERGEN (1984), THEODORIDIS (1984). I have rarely cited the syntheses of AUBRY (1984) and PERCH-NIELSEN (1985b) since most of the illustrations have previously been published. For well illustrated species I have usually only cited a few references, followed by the code etcetera.

10.2 SYSTEMATIC TREATMENT OF INTRASPECIFIC VARIATION 10.2.1 Rationale As discussed above (Chapter 5), there is clear evidence that intraspecific variation is important in nannofossils, and that many currently used "species" are intergradational variants. I believe that in order to be scientifically rigorous it is necessary to adjust formal taxonomy to reflect this. Doing so should preserve all the descriptive power of the present system, but make explicit the relationships between taxa. This is plainly vital for evolutionary studies, and for work on diversity changes through time, it also should make taxonomy easier to learn and more informative. For instance in the Pliocene the following

143 "species" of Discoastez are currently used, D.asyimetricus, D.brouweri (+3 rayed variant), D.decorus, D.pansus, D.pentaradiatus, D.surculus, D.tamalis, and D.vaziabilis. For practical biostratigraphy by experts this is an adequate and effective system, indeed for biostratigraphy all that matters is that names are uniformly used and understood. However, without expert knowledge it is not clear that there are only four basic species (D.brouweri, D.pentaradiatus, D.surculus, and D.variabilis), with the other "species" being variants of them. Making this clear seems desirable as well as objective. In doing so the following principles are important: A. Likely biological relationships should be reflected as closely as possible. B. Name changes should be minimised. C. The International Code for Botanical Nomenclature (ICBN) should be followed. All these can be achieved by recombining the existing species as varieties of more generalised species.

10.2.2 Formal usage The ICBN covers in detail the usage of infraspecific taxa, notably in Articles 24 to 27, 49, 53, and 56 (Sydney, 1981 edition). The following points seem important. A. Choice of level: An hierarchy of infraspecific ranks is provided (Art.4.1): subspecies, variety, subvariety, form, subform. For nomenclatural taxonomy these are simply a series of "pigeonholes" any of which could be used. They are also, however, like species, words with evolutionary uses and implications. In particular subspecies is generally used for variants produced by reproductive isolation (either geographical or ecological). In most cases there are no grounds for believing that this was the case with the intraspecific nannofossil categories. Variety, which does not have these implications, is thus a more appropriate level. It also has a shorter formal abbreviation - var. vs subsp. B. Citation: The general form of citation of an infraspecific taxon is; Generic name, specific epithet, rank term, infraspecific epithet, (original author of taxon), author of combination (Arts.24.1, 26A Ex.2, 49). Thus if the "species" closely related to D.brouweri are recombined as varieties of D.brouweri the full citations should be: Discoaster brouweri var. asymmetricus (Gartner) n.comb. Discoaster brouweri var. tamalis (Kamptner) n.comb. Discoaster brouweri var. triradiatus (Tan) n.comb. C. Autonyms: A species is the sum of its subordinate taxa, and any infraspecific taxon which includes the type of the species must use the

144 specific epithet as its epithet (=autonym, Arts.25, 26). Thus in the D.brouveri example there is a further combination: Dlscoaster brouveri Tan var. brouweri. This does not count as a new combination, does not need to be made formally, and has the author name cited after the specific epithet. Typicus, originalis etc. are invalid as alternatives to use of the autonym (Art 24.3). Simple citation of D.brouweri (Tan) would still be valid, but would cover all the varieties. D. Priority: Recombining a species as a variety of another species is similar to synonymising the two and hence the older name must be used as the specific epithet (Art.26). This is unlikely to cause many problems for nannofossils since in general the older names (e.g. D.brouweri) have been used for central morphotypes. Article 60 states that "in no cases does a name have priority outside its own rank". Hence after synonymising D.brouweri and D.triradiatus there is no need to use triradiatus as the varietal epithet of the 3-rayed variety of O.brouweH For the sake of nomenclatural conservatism it is, however, usually desirable to retain the original names.

10.2.3 Informal usage The main disadvantage with the use of infraspecific categories is that the resulting trinomials are unwieldy. The ICBN does not provide for abbreviations, other than of the generic name. Nonetheless they are vital for certain uses, such as range charts, and are desirable for more general use. I have adopted the convention, commonly used by for instance ornithologists, of dropping the rank term and abbreviating the specific epithet on the pattern of generic abbreviations. Thus Discoaster brouweri var tamalis can be informally cited as D.b.tamalis.

10.3 TAXA NOT INCLUDED IN THE SYSTEMATIC CHAPTERS. The list below includes all Neogene to Recent taxa mentioned in the text or included in the distribution charts. (Appendix 1), but not otherwise discussed. Palaeogene and Mesozoic taxa are not included, references to these are given in PERCH-NIELSEN (1985a,b). Notes on the families cover their salient features, and distribution, and give major references on their biology and taxonomy.

DINOPHYCEAE THORACOSPHAERACEAE Schiller 1930 Thoracospheres - included in nannofossils on grounds of size. References; FUTTERER (1976, 1978), TANGEN et al (1982). Sporadic in

145 Indian Ocean, common in Red Sea material. Plate 5/4-5,27. Thoracosphaera Kamptner 1927. T.albatrosiana Kamptner 1963. T.heimii (Lohmann 1919) Kamptner 1941.

CLASS HAPTOPHYTA BRAARUDOSPHABRACEAE Deflandre 1947 Pentaliths - distinctive discrete group from the Mesozoic to the Recent, with characteristic lamellar ultrastructure. Generally considered as littoral or low salinity indicators. References; BYBELL & GARTNER (1972), LEFORT (1972), LAMBERT (1986). Very rare in my material. Plate 5/6. Bzaazudosphaeza Deflandre 1947. B. bigelowii (Gran & Braarud 1935) Deflandre 1947.

CALCIOSOLENIACEAE Kamptner 1927. Scapholiths - distinctive discrete group from the Mesozoic to the Recent, with rhomboidal coccoliths. Motile, and include one non-calcifying form, Navisolenia. The genera Anoplosolenia and Scapholithus are poorly distinguished and should probably be suppressed in favour of Calciosolenia. Reference; MANTON & OATES (1985). Sporadic rare occurrences throughout my material. Plate 5/24, Figure 7/B. Anoplosolenia Deflandre, in Grasse 1952. Calciosolenia Gran, in Murray & Hjort 1912. Scapholithus Deflandre, in Deflandre & Fert 1954. Navisolenia Lecal 1965.

CALYPTROSPHAERACEAE Boudreaux & Hay 1969 Holococcoliths - very diverse in modern oceans, fossil record patchy very thin in Neogene. Predominantly (?exclusively) motile. References; GARTNER & BUKRY (1969), KLAVENESS (1973), HEIMDAL & GAARDER (1980), NORRIS (1985). Tetralithoides symeondesii is the only possible holococcolith I observed (occurrences 219-14-3 (NN4), 242-7-4 (NN6), 251-10-6 (NN11)). Figures 3/F, 4/H, 6/C. Calyptzolithina Heimdal 1982. C. multipora (Gaarder, in Heimdal & Gaarder 1980) Norris 1985. Calyptrolithophora Gaarder, in Heimdal & Gaarder 1980. Calyptrosphaera Lohmann 1902. C.sphaeroidea Schiller 1913. Daktylethra Gartner, in Gartner & Bukry 1969. Sphaezocalyptza Deflandre 1952. Tetzalithoides Theodoridis 1984. T.symeondesii Theodoridis 1984.

146 HYMBNOMONADACBAB Senn, in Bngler & Prandt 1900. Cricoliths - predominantly littoral group with benthonic non-motile phase. Very small coccoliths on the benthonic phase. References; PARKE (1971), OUTKA & WILLIAMS (1971), HIBBERD (1980). Not known from the fossil record. Figures 2/A,C, 3/D, 4/C, 5. Apistonema Pascher 1925. Chrysotila Anand 1937. C.lamellosa Anand 1937. Hymenomonas Stein 1878. Ochrosphaera Schussnig 1930. Pleurochrysis Pringsheim 1955. P.carterae. (Braarud & Fagerland 1946) Christensen 1978 [=Hymenomonas carterae (Braarud & Fagerland) Braarud 1954, Cricosphaeza caztezae (Braarud & Fagerland) Braarud 19601.

RHABDOSPHAERACEAE Lemmermann, in Brandt & Apstein 1908 Rhabdoliths - diverse and abundant in modern oceans (especially l/mbellosphaera), less so in the Neogene. Coccoliths have distinctive spiral arrangement of elements in the central area, usually forming a spine. Motile and non-motile forms. Reference; NORRIS 1984. The only important Neogene species is R.clavigeza, which is often common. Plate 5/28, Figures 4/D, 6/D, 7/D. Rhabdosphaeza Haeckel 1894. R. clavigera Murray & Blackman 1898 (syn. R.stylifera Lohmann 1902). Acanthoica Lohmann 1902. A.quattzospina Lohmann 1903. Discosphaeza Haeckel 1894. Umbel1 osphaera Paasche, in Markali & Paasche 1955.

SYRACOSPHAERACEAE Lemmermann, in Brandt & Apstein 1908. Caneoliths and cyrtoliths - very diverse group in modern oceans, with elaborate coccospheres often showing dithecatism. predominantly (?exclusively) motile, and low latitude. References; GAARDER & HEIMDAL (1977), MANTON & OATES (1982), MANTON et al (1984). Occur throughout the Neogene, and more commonly in the Pleistocene. Figures 6/E, 7/A, 15/C. Deutscblandia Lohmann 1912. Halopappus Lohmann 1912. H.adriaticus Schiller 1914. Michaelsazsia Gran, in Murray & Hjort 1912. Syzacosphaeza Kamptner 1941. S. pulchza Lohmann 1902.

147 NON-CALCIFYING GENERA. Organic scales only - important in shelf seas. References; BONEY 1970, MANTON & LEADBEATER 1974, HIBBERD 1980. Chiysochzomulina Lackey 1939. C.polylepis Manton & Parke 1962. C.pringsheimii Parke & Manton 1962. Prymnesium Carter 1937.

ARCTIC SEMI-CALCIFYING GENERA. Anomalous rarely recorded group with calcification occurring within an organic framework (Section 2.4.5). References; MANTON et al (1977), NORRIS (1983). Calciarcus Manton et al 1977. Papposphaeia Tangen 1972. Papposphaeza lepida Tangen 1972. Tuzzisphaeza Manton et al 1976. Wigwamma Manton et al 1977 Wigwamma arctica Manton et al 1977.

148 CHAPTER 11 - COCCOLITHACEAE Neogene placolith coccoliths divide into two major groups, the Prinsiaceae and the Coccolithaceae. The Prinsiaceae have proximal shields with a simple structure, and both shields are birefringent in plan view. In the Coccolithaceae, by contrast, the proximal shield usually has a complex structure, and the distal shield shows little or no birefringence. In addition there are a few genera which show neither pattern. These have usually been included in the Coccolithaceae mainly as a matter of taxonomic convenience, but this assignment may in fact be justifiable for many of the genera as discussed below.

11.1 STRUCTURE The distal shields of coccolithaceaen coccoliths are formed of a single cycle of elements; these elements are imbricated and inclined, but do not show birefringence in plan view. The structure of the proximal shield is variable, and can be used as the basis for dividing the genera included in the coccolithaceae into three groups. A. Coccolithus and related genera, proximal shield bicyclic and birefringent: This structure is shown by Coccolithus, Clausicoccus, Cycloperfolithus, and Umbilicosphaera (Figs.39/A,B & D, p.164), and also by the Palaeogene genera Chiasmolithus and Cruciplacolithus. It can be considered the typical structure. As shown in Figure 39 a cycle of apparently short elements occurs around the margin of the proximal shield. ROMEIN (1979), working with Palaeogene species, interpreted this outer cycle as structurally part of the distal shield, (as indicated on my cross-sections). The alternative interpretation, implicitly suggested by many workers, is that the outer cycle of the proximal shield is an independent structure. I favour the interpretation, (although not the terminology) of Romein, since in electron micrographs of damaged or tilted specimens it is sometimes possible to observe the elements of the outer cycle of the proximal shield continuing above the inner cycle to the central column, (e.g PERCH-NIELSEN 1971, pi.1/2, Ericsonia oval is; ROTH 1973, pi.2/4, U.sibogae). Also this produces a simpler model for the structure, of the type suggested above from consideration of coccolith development (Chapter 4). Structural variation within this group is provided by: a wall cycle, of irregular concentrically arranged birefringent elements, (absent in Umbilicosphaera and Cycloperfolithus); and by various central area structures (e.g. distal plates in Clausicoccus). B. Calcidiscus, proximal shield monocyclic, birefringent: The

149 structure of Calcidiscus was carefully elucidated by GARTNER (1967b), and has since been veil illustrated by SEMs, it is shown in Figure 39/C. The structure is similar to that of the Coccolithus group, except that the distal shield / column cycle does not extend to form an upper cycle to the proximal shield. As a result the proximal shield is monocyclic. C. Oolithotus, Hayaster, proximal shield monocyclic, non- birefringent: These coccoliths (Figs.39/B-F) are essentially formed of a single cycle of elements which do not show birefringence in plan view. They could have been derived from the Coccolithus or Calcidiscus groups, by loss of the birefringent lower cycle. No other origins seem more likely, so it is reasonable to provisionally retain these genera in the Coccolithaceae. Two remaining genera Geminilithella and Cozonocyclus (Figs39/G-H) have poorly understood structures, in particular the optical orientation of their constituent elements is unclear, so it is difficult to speculate on their relationships.

11.2 BIOSTRATIGRAPHY AND SIZE VARIATION Figure 40 (p.166) illustrates the ranges and probable relationships of the dozen or so*Neogene species of Coccolithaceae. The appearances and disappearances of the lineages seem to be gradual events so the principal biostratigraphic potential of the group is probably intraspecific variation within the main species, Coccolithus pelagicus and Calcidiscus leptoporus. Both of these are abundant, robust, and morphologically variable. I did not do any intensive biometric work on these species, but did measure the lengths of a limited number of specimens from most samples in which they were reasonably common (a total of about 4,000 specimens from about 80 samples). Although the data-sets are individually too small to be meaningful, some useful information can been obtained from them; by combining the data from all samples within a given age range. Obviously the averaging may lose some detail, but as a corollary of this, spurious trends are less likely to survive.

11.2.1 Coccolithus pelagicus group. Figure 41/A shows my biometric data on the Coccolithus pelagicus group, synthesised as described above. Each histogram includes all the available data for the relevant interval. The data is weakest for the Early Pliocene (Histograms NN12-14 and NN15), since C.pelagicus is rare in this interval. Throughout the Miocene (Histograms NN2-5 to NN11) the assemblages are dominated by specimens 6-9 microns long, with a range of 4 to 12

150 microns. Significantly larger specimens (>13microns) occur at the Oligo-Miocene boundary interval (Histogram NP25-NN1), and in the mid-Miocene (Histogram NH7-8). The former are representatives of C.eopelagicus, they typically have large centres and grade continuously into the smaller specimens. The latter, mid-Miocene, group represent C.pelagicus var. miopelagicus. Individual samples often have a bimodal size distribution. They are separated from the C.eopelagicus specimens by an interval representing most of the early Miocene, during which large specimens are very rare or completely absent. Also C.p.miopelagicus has a much smaller central area than C.eopelagicus. These observations suggest that C.p.raiopelagicus was derived from C.pelagicus, by increasing rim width, rather than from C.eopelagicus as suggested by BUKRY (1971c). The only other relevant biometric data is that of BACKMAN (1980), he identified a discrete C.p.miopelagicus population 12-13 microns long from sediments of probable NN6 age. This fits well with my data, but obviously more quantitative information is needed. In the Pliocene and Pleistocene very large specimens do not recur, there is, however, an increase in average size associated with the appearance of bridged forms (shaded area on the histograms). The dominant size range shifts from 6-9 to 8-10 microns, and the total range from 4-12 to 6-13 microns. The bridged forms occur at the upper end of this size range. In this material they first occur in the early Pliocene, but elsewhere they have been recorded elsewhere from within NN11 (eg STEINMETZ & STRADNER 1984). They do not become common, however, until the mid Pliocene (ca. NN15). Since they are biostratigraphically useful these forms are described below, as a new variety of Coccolithus pelagicus; C.p.pontus. The small specimens in the NN15 histogram are somewhat anomalous, they come from a single sample (251A-4-2), and may represent either reworking or an aberrant population, possibly associated with the NN15/NN16 boundary effects. The conclusions from this are of use; normal sized (6-9microns) C.pelagicus specimens, which are probably indistinguishable from any parts of the Neogene occur continuously, but Larger varieties are associated with them at three separate intervals. First, C.eopelagicus (?Late Eocene to Early Miocene - NN1), which is up to 20microns long with similar proportions to typical C.pelagicus. Second, C.p.miopelagicus (mid Miocene - 7NN5-NN8), 13-20microns long with a small central area, and so low ellipticity, and rather few rays for its size. Third, C.p.pontus (Pliocene to Recent), 9-12 microns long with a rather wide central area spanned by a bridge. Plainly these different forms can be of use for low-resolution biostratigraphy. The last

151 occurrence of C.p.miopelagicus seems particularly reliable, it .swell established as occurring consistently within NN8. In addition to this morphological variation the C.pelagicus group shows significant variations in abundance through time. Its distribution in the examined material is shown in Figure 42/A. HAQ (1980) described in detail climatically related fluctuations in C.pelagicus distribution through the Miocene and Early Pliocene, C.pelagicus is always most abundant at low-latitudes, and during cold periods became Increasingly abundant at mid-latitudes. Also various authors have noted that C.pelagicus disappears from low-latitudes at the end of the Pliocene, McINTYRE et al (1967), RAFFI & RIO (1981), NISHIDA (1982). I did not detect the Miocene fluctuations described by Hag, probably because my material was predominantly from low-latitudes. C.pelagicus does, however, definitely become much rarer in the Early Pliocene, as he describes (Fig.42A). It is more common in the Late Pliocene, but virtually absent from the Pleistocene samples I looked at. It is not clear whether this Pleistocene shift in C.pelagicus distribution was due to an increased polarisation of climates, so that low-latitudes waters were warmer in the Pleistocene (cf McINTYRE ET AL 1967), or to an evolutionary change in the temperature tolerance of C.pelagicus, as suggested by RAFFI & RIO (1981) and by NISHIDA (1982). Nonetheless the event is sufficiently well established to be used, with discretion, for biostratigraphy.

11.2.2 Calcidiscus leptoporus group. Figure 41/B shows size data for the C. leptoporus group synthesised in the same way as on the C.pelagicus group. All the histograms show similar distribution patterns, except the first (NN2-5), and perhaps the last. The lower size range in the first histogram is probably meaningful, reflecting size increase during the early evolution of the group. It is supported by the fact that the larger species or variety, C.l.macintyrei, has only occasionally been reported below NN6 whereas C.1.leptoporus occurs down to at least NN3. C.leptoporus probably evolved from one of the various poorly known small Oligocene coccoliths such as Cyclococcolithus bollii Roth 1970, and increased gradually in size and abundance during the Early Miocene. Extinction of the larger form, C.l.macintyrei early in the Pleistocene is widely documented in the literature, and is reflected in the shift to lower sizes in the final sample.

11.2.3 Other species. The other species of Coccolithaceae, shown on Figure 40, are not usually used for biostratigraphy. For the most part this is sensible

152 since the occucrence of species such as U.jafari is strongly influenced by preservation, and most events are gradational. Nonetheless some Information of use can be extracted from the group: A. Last occurrence of Coronocyclus nitescens: All available information suggests this occurs in NN7. Also elliptical forms seem to develop in the Middle Miocene. B. First occurrence of Geminilithella rotula: THEODORIDIS (1984) proposed this as a sub-zonal marker event, within NN2. However, G.rotula only occurs sporadically until about NN6 (my data and reported occurrences) so its first occurrence is not a reliable event. C. Umbilicosphaera: Typical U.jafari which is 2-3 microns in diameter is preceded by a larger variety (3-5 microns) in the Early Miocene, in my material. In the Late Pliocene it evolves into U.sibogae, up to 8 microns. D. Clausicoccus pzimalis and Cycloperfolithus carlae: both these species occur persistently through the Miocene, but not higher (Fig.42/B,C). This might in part be an ecological effect paralleling the restriction of Coccolithus pelagicus in the Early Pliocene (Fig.42/A).

11.3 SYSTEMATICS Family COCCOLITHACEAE Poche 1913

Genus Calcidiscus Kamptner 1950 Calcidiscus leptoporus (Murray & Blackman 1898) Loeblich & Tappan 1978 Remarks: The structure of Calcidiscus leptoporus is described and illustrated above. The species shows high morphological variability in both living and fossil material, which has caused taxonomic problems. The main variation is in size (c.4 - 14 microns), number of rays (c.15-45), ellipticity (circular to weakly elliptical), and central area type (closed vs open, and presence/absence of bars); this variation is illustrated in Plate 2. There is a general trend for larger specimens to be more nearly circular, to have more rays, and a wider central area. This trend is similar to that shown by Pseudoemiliania lacunosa (Chapter 4, p.53), but the ratio of rim width to central area breadth is much greater in C.leptoporus, and so morphology is less predictably related to size. Various authors (McINTYRE et al 1967, JANIN 1981, PERCH-NIELSEN 1985b) have proposed subdivisions of the C.leptoporus group. The most important formalisation of this is the separation of large forms as C.macintyrei. Although virtually all authors note that intermediates occur it has been considered useful to separate this form since it disappears at least temporarily in the Early Pliocene. I prefer to

153 treat it as one variety of C.leptoporus, as outlined below. In addition isolated shields and minor variants of C.leptopozus have repeatedly been described as separate species; HAY & BEAUDRY (1973) and JAFAR (1975a) give extensive synonymies.

Calcidiscus leptoporus (Murray & Blackman 1898) Loeblich & Tappan 1978 var. leptopozus Plate 1/28-29, 2/2,11; Figure 39/C. Coccosphaera leptopora HURRAY & BLACKMAN 1898 p.430f p i.15/1-7. Coccolithas leptoporas (Hurray & Blackaan) SCHILLER 1930 p.245, fig .10. Cyclococcolithas leptoporas (Hurray & Blackaan) KAHPTNER 1954 p.23, fig.20; etcetera. Coceolithas radiatas KAHPTNER 1955 p.34, p i.7/92; JAFAR 1975a p.63f p i.9/10-11,18; HAO & BER66REN 1978 p i.1/7-8. Calcidiscas leptoporas (Hurray fc Blackaan) LOEBLICH Ir TAPPAN 1978 p.1391; STRADNER fc ALLRAH 19B1 p i.14/1-2; etcetera. Cyclococcolithas leptoporas ssp. centrovalis STRADNER & FUCHS 1980 p.255, p i.5/1-9, 6/2-6, 7/1-6. Remarks: C.l.leptopozus is rarely perfectly circular, so distinguishing slightly elliptical variants of similar size (C.radiatus, C. 1.centzovalis) is impractical and pointless. Size: 4-8 microns.

Calcidiscus leptopozus var. A Plate 2/1,6 Cyclococcolithus leptoporas (Hurray & Blackaan) Schiller, GARTNER 1967a p i.1/3. Cyclococcolithas Macintyrei Braalette h Bukry, HULLER 1974a pi.3/11, 16/12. Description: Circular, or only slightly elliptical, variety of C.leptopozus, with an open central area spanned by radiating bars on the proximal side. Remarks: A large proportion of the C.leptopozus specimens in my material have bars in the central area, these are non-birefringent and probably formed from the column. In terms of size, ray number, and central area size, I found this variety was generally intermediate between C.l.leptopozus and C.1.macintyrei. There are few other records of this form, so its occurrence in the Indian Ocean may be a local phenomenon. Size: c7-10 microns. Occurrence: Common in Late Miocene to Early Pliocene samples, sporadic elsewhere.

Calcidiscus leptoporus var. macintyrei (Bukry & Bramlette 1969) n.comb Plate 2/7-9, 13-14 Cycloeoccolithas leptoporus (Hurray & Blackaan) Kaaptner, GARTNER 1967a p.1-4, p i.1/1-2,4, 2/1-4. Cyclococcolithas Macintyrei BUKRY & BRAHLETTE 1969; etcetera. Calcidiscas Macintyrei (Bukry & Braalette) LOEBLICH & TAPPAN 1978 p.1392; etcetera.

154 Description: Large variety of C.leptopozus with many rays. Normally almost perfectly circular with open central area. Remarks: In some samples there is a clear distinction between C.l.macintyrei and C.l.leptoporus, in most cases, however, the two merge continuously, and can only be separated by means of an arbitrary criterion, (e.g. size). This division is probably only worth making with Late Pliocene / Early Pleistocene material. Size: 8 to 13 microns (lower limit is approximate minimum size of specimens showing the typical morphology, for biostratigraphy an arbitrary minimum size of ten microns is more suitable).

Calcidiscus leptoporus var. premacintyrei (Theodoridis 1984) n.comb Plate 2/9-10, 14-15 Calcidiscus preaacintyrei THEODORIDIS 1984 p.81, p i.2/1-3. Calcidiscus aaciatyrei (Bukry & Bra*lette) STEINMETZ It STRADNER 1984 p i.5/1. Description: Large, weakly elliptical variety of C.leptoporus, central area open, and distinctly elliptical. Remarks: This variety is very similar to C.l.macintyrei but is limited in stratigraphical range (Fig.40), and readily distinguished by the elliptical central area.

Calcidiscus leptoporus var. pataecus (Gartner 1967) n.comb Plate 2/3 Coccolithus pataecus 6ARTNER 1967b p.4, pi.5/6-8; HAB It BER6GREN 1978 p i.1/1-2. Coccolithus sp. BUKRY 1971b p i.4/1. Coccolithus iuscus BACKHAN 1980 p.47, p i.1/12, 2/6-9. Coccolithus radiatus Kaaptner 1955, BACKMAN 1980 p i.1/10-11, 2/4. Description: Small (<6microns) and distinctly elliptical variety of C.leptoporus. 10 to 25 elements. Remarks: The smallest C.leptoporus specimens vary more strongly in ellipticity than the larger ones, a geometrical consequence of them having a lower rim width: central area ratio. It is thus possible to distinguish an elliptical variety, it is not, however, very useful. The coccoliths have the same structure and optical characters as C.leptoporus, and the same tendency for the shields to separate (GARTNER 1967b, my obs.). Occurs throughout the range of C.leptoporus.

Genus Clausicoccus Prins 1979. Description: Placoliths with Coccolithus type shields and wall, but with a perforated plate covering the central area on the distal side. Remarks: I agree with PRINS (1979) that this group should be separated from Coccolithus/Ericsonia, since they are structurally at least as distinct as Chiasmolithus and Crucipiacolithus. Moreover as ROMEIN

155 (1979), PRINS(1979), and PERCH-NIBLSEN (1985b) have suggested the group almost certainly evolved from Cruciplacolithus, so retaining them in Coccolithus would make the latter genus polyphyletic.

Clausicoccus fenestratus (Deflandre & Pert 1954) Prins 1979 Plate 1/6-15; Figure 39/D Discolithus feaestratus DEFLANDRE fc FERT 1954, p.139, fig .52, p i.11/25. Ellipsolithas sabdistichas Roth & Hay in HAY et at 1967, p.446, p i.6/7. Eriesoaia fete strata (Deflandre fc Fert) Stradner in STRADNER fc EDUARDS 1968 p. 18, p i.10/1-4, 11/1-7; HAQ 1968 p i.1/10-12; ROTH 1970 p i.1/6; HAQ 1971b, pi.3/7-9; MULLER 1979 pi.7/6. Ericsoaia subdisticha (Roth and Hay) Roth in BAUMANN fc ROTH 1969, p.319; ROTH 1970 pi.2/3-4; HAQ 1971b p i.3/3-6. Ericsoaia bireticulata ROTH 1970 p.840, p i.1/4-5. Ericsoaia paneiperforata ROTH 1970 p.842, p i.2/1; HAQ 1971b p i.3/1-2.

C'.priaalis Roth 1970, HAQ 1971c partiv p i.6/1, 16/9-11, b o b p i.16/7-8; HAQ & LIPPS 1971 pl.3/c-d. Ericsoaia qaadriperforata ROTH 1970 p.483, p i.2/2. Ericsoaia obruta PERCH-NIELSEN 1971 p.14, p i.4/4-7, 8/5-6, 61/10-11; EDUARDS & PERCH-NIELSEN 1974 p i.12/4,7, 13/3-4. Ericsoaia tasaaaiae EDUARDS & PERCH-NIELSEN 1974 p.481, p i.20/5-12, 21/1-6. Clausicoccas feaestratas (Deflandre & Fert) PRINS 1979 p.2-3, p i.1/3; UISE 1983 p i.1/8,9. Claasicoccas obrutas PRINS 1979 p.3, p i.1/4. Claasicoccas sabdistichas PRINS 1979 p.3, p i.1/1-2. Coccolithas sp. MARTINI 1981 p.558, p i.5/7-8. Remarks: Late Eocene - Early Miocene Clausicoccus specimens vary considerably in number of pores, and in prominence of the wall. The various morphotypes are, however, difficult to separate, particularly with a light microscope, and have similar ranges. I suspect only one species is present, although division of it at the varietal level is almost certainly justifiable. Published electron-micrographs (synonymy above) show more elements on the proximal side of the central-area covering than on the distal side, so probably it is formed of two layers. In cross-polars this plate shows, as described by EDWARDS & PERCH-NIELSEN 1974 and PRINS 1979 a Cruciplacolithus type extinction pattern. The pattern appears too complicated to be formed by the large distal elements, and so perhaps only the proximal elements are birefringent; as indicated in Figure 39/D. The wall and proximal shield are also birefringent, although the proximal shield is often lost. Size: 4.5 to 7 microns (my data, published sizes, size of illustrated specimens). The Early Eocene species Clausicoccus cribellum (BRAMLETTE & SULLIVAN 1961) PRINS 1979 is similar but distinctly larger (6-12 microns cf. BRAMLETTE & SULLIVAN 1961 and ROMEIN 1979). Occurrence: Eocene to Early Miocene (NN2 or 3), only sporadic in the

156 Early Miocene but a characteristic component of Oligocene assemblages.

Clausicoccus primalis (Roth 1970) n.comb Plate 1/6-15

Coccolithus pri*alis ROTH 1970 p.839f p i.1/3; HAO 1971c parti * p i.16/7-8, bob p i.61, 16/10-11. ?6en. and sp. indet. VAROL 1985 p i.1/12. Description: Small species similar to C.fenestratus. The central area is usually filled by only two plates, occasionally three or four, and pores are rarely clearly visible. Remarks: In my Oligocene material this form occurs with and grades into C. fenestzata but since it has a much la-ter last occurrence, in the I'cvte Miocene, I prefer to distinguish it as a separate species. B.obzuta PERCH-NIELSEN 1971 appears very similar structurally but rather larger. Size: 3-4.5 microns (my data). Occurrence: My material, Oligocene to Late Miocene (NN11), Figure 24/B.

Genus Coccolithus Schwarz 1894 Coccolithus pelagicus (Vallich 1877) Schiller 1930 Remarks: This species forms the Neogene part of a lineage which ranges from the Early Palaeocene to the Recent. The Palaeogene part of the lineage is normally distinguished as Coccolithus (or Ericsonia) eopelagicvs. Their structure is discussed and illustrated above. They show significant variation in rim width and central area size, with related effects on the number of rays, as discussed above this can be used to distinguish a series of separate larger forms, and also one smaller form described below. I do not regard this biometric variation, without significant structural variation as adequate grounds for dividing the group into species, particularly since a continuous range of intermediates is usually present. Separation as varieties is, however, biostratigraphically useful, although sometimes arbitrary.

Coccolithus pelagicus (Vallich 1877) Schiller 1930 var. pelagicus Plate 1/2-4, 24-26, 2/34-35; Figure 39/B, 43/A Coccosphaera pelagica WALLICH 1877 p.348, p i.17/1-2,9-11. CoccoJithus pelagicas (Wallich) SCHILLER 1930 p.246, fig .123-124; McINTYRE It BE 1967 p i.8/a-c; etcetera. Coccolithus pliopelagicus WISE 1973, p.598, pi.8/1-6. Description: Typical form of C.pelagicus. Varying in size from about 5 to 12 microns, without a bridge. Occurs throughout the Neogene, with varying abundance. WISE (1973) separated forms of this size with a small central area as C.pliopelagicus, 1 did not separate this variety,

157 but it might be useful to do so.

Coccolithus pelagicus var. niopelagicus (Bukry 1971) n.comb. Plate 1/1; Figure 43B C.tiopelagicas BUKRY 1971a p.310, p i.2/6-9; WISE 1973 p.598, pi.8/9-11; BACKHAN 1980 p.B, p i.1/3-4. Description: Large variety of C.pelagicus usually with a small central area. Size: 14 - 20 microns, typically 15 - 17 microns.

Coccolithus pelagicus var. nannopelagicus n.var. Plate 1/27; Figure 43/E C.?pelagicus HAQ k LOHHANN 1975 p i.8/4. Holotype: Plate 1/Z7, Site 242, Madagascar Channel. Sample 8-5-99, Early Miocene. Description: Very small variant of C.pelagicus. Morphology otherwise normal. Remarks: C.pelagicus is rarely less than five microns long, so it seems worthwhile to distinguish small variants when they do occur, as well as large ones. I only found these very small forms in Early Miocene samples. They are similar to and occur with Clausicoccus primalis (q.v.) but have a broader wall, and a central opening. Size: 3 - 5 microns.

Coccolithus pelagicus var. pontus n.var Plate 1/5; Figure 43/E Coccosphaera carteri HALLICH 1877 partia p i.17/6-7 non p i.17/3-4 ( =Helicosphaera carteri). Coccosphaera pelagica Hallich DIXON 1900 p i.3/1-6. Coccolithus pelagicas Hallich Schiller BLACK 1968 p i.143/1-2; MANTON k LEEDALE 1969 p i.1/1,6-7,10, 4/20; BUKRY 1971b p i.1/2; PERCH-NIELSEN 1972 p i.1/6, 2/2,6; STRADNER 1973 p i.1/1/6, 2/1-3; NISHIDA k K0NDA 1974; CEPEK k HIND 1979 p i.2/7, 6/9,11-12; NISHIDA 1979a p i.3/1; NISHIDA 1982 p i.1/1-6. Coccolithus carteri (Hallich) Kaiptner 1941, TOTTEN 1976 p.158, p i.1/1-3; STEINMETZ k STRADNER 1984 p i.3/1,3,5-7. Holotype: Plate 1/5, Site 223, NW Indian Ocean, Sample 6-4-70, Early Pliocene. Description: Variety of C.pelagicus with a bridge spanning the short axis of the central area. The bridge is formed from the proximal part of the wall, and like the wall is constructed of numerous small elements. In cross-polars it shows the same optical orientation as the adjacent part of the wall, and so has slightly inclined extinction. The degree of development of the bridge is variable. Remarks: This form of C.pelagicus has long been noted, and early workers

158 such as DIXON (1900) also observed that bridged and unbridged forms were otherwise identical, and could occur on the same coccosphere (although HEIMDAL & GAARDER (1981) question this), or within the same cultural isolate, MANTON & LEEDALE (1969). In consequence it has not been differentiated. Nonetheless it does have a distinct stratigraphical range. In my material it occurs very rarely in the topmost Miocene (eNNll), rarely in the Early Pliocene and commonly from then to the Recent. Various points suggest that this distribution is widespread: TOTTEN (1976), BACKMAN (1980), and STEINMETZ & STRADNER (1982) record similar occurrence patterns from the Pacific*: and Atlantic; all the specimens in the synonymy above came from Pliocene or Pleistocene sediments; various authors working on older material have noted the absence of the bridged form (e.g. MARTINI 1965, JAFAR 1975a). Size: 7-12 microns (in most samples the larger C.pelagicus specimens more commonly have bridges than the smaller ones). Range: 1NN11 - Recent (common NN16-Recent).

Genus Cozonocyclus HAY MOHLER & WADE 1966. Cozonocyclus nitescens (Kamptner 1963) Bramlette & Wilcoxon 1967 Plate 2/18.-23; Figure 39/H (Jnbil icosphaera nitescens KAMPTNER 1963 p.187, p i.1/5. Coronocyclas serratus HAY et al 1966 p.394, p i.11/1-5 Coronocyclas nitescens (Kaiptner) BRAMLETTE It UILC0X0N 1967 p.103, pi. 1/4, 5/7-8; HAQ 1971c p i.18/10-12; EDWARDS k PERCH-NIELSEN 1974 p i.19/2, 20/4; MULLER 1974a pi.3/9-10; MULLER 1974b p.391 pi.3/28-29; PERCH-NIELSEN 1977 p i.41/1-6; STRADNER k ALLRAM 1981 p i.9/1-2. Coronocyclus prion ion STEINMETZ k STRADNER 1984 pi.43/7. Calcidiscas aequiscutun STEINMETZ fc STRADNER 1984 partin p i.24/1-2, non p i.24/3-5 (= 6 . r o t u l a). Description: Ring shaped coccoliths with a serrate outline and distally directed spines, but without discrete shields. Bright in cross-polars. Structure: The structure of this species, as discerned from published SEMs, is shown in Figure 39/H . It is formed of two cycles of elements. The inner, spine bearing, cycle is imbricate and wraps around the outer cycle of vertical elements. If the outer cycle is non-birefringent in plan view, which is consistent with the extinction figure, then this structure may be homologous with that of Coccolithus pelagicus. Size: 4 to 8 microns. Range: Oligocene to Middle Miocene, never common.

Cozonocyclus nitescens var. ellipticus n.var Plate 2/24-25 Coronocyclas nitescens (Kaaptner) Braalette It Wilcoxon, MULLER 1974b p.391 p i.3/28-29; PERCH-NIELSEN 1977 p i.41/3,6. Holotype: Plate 2/25, Site 242, Madagascar Channel, Sample 7-4-105,

159 Middle Miocene (NN6). Description: Elliptical variety of C.nitescens, otherwise identical. Remarks: Although C.nitescens is usually perfectly circular, markedly elliptical specimens occur in some of my samples, and have also been noted by other authors. They seem to be confined to the Middle Miocene and so are worth distinguishing.

Genus Cyclopezfolithus Lehotayova & Priewalder 1981 Cycloperfollthus carlae Lehotayova & Priewalder 1981 Plate 2/4-5 Cyclococcolithus »acint

Genus Geminilithella Backman 1980. Remarks: BACKMAN (1980) introduced this genus for circular coccoliths with planar, rather than concavo-convex, shields. In fact the type species, G.rotula, is gently concavo-convex but I prefer to retain the genus since G.rotula has a monocyclic proximal shield and cannot be placed in any other Neogene genus.

160 Geminilithella rotula (Kamptner 1956) Backman 1980 Plate 1/20-21, 2/30; Figure 39/G Cyclococcolithus rotula KAHPTNER 1956 p.10, ex KAHPTNER 1948 p.8f pi.2/15; HULLER 1974a p.591, p i.3/6-8; HULLER 1974b p.391, p i.1/7-8; JAFAR 1975a p.72 pi.8/2,18-19 (?pl.2/20-23). Cyclococcolithus aequiscutua 6ARTNER 1967b p.4, p i.7/1-4. Cyclococcolithus cricotus GARTNER 1967b partia p i.7/7, aoa pi.7/5-6. (= Pseudoeailiaaia lacuaosa). Calcidiscus aequiscutua (Gartner) LOEBLICH It TAPPAN 1978 p.1391; STE1NHETZ It STRADNER 1984 partia, p i.24/3-6 aoa pi.24/1-2 (=Coroaocyclus oitesceas). Cyclolithella rotula (Kaaptner) HA8 It 8ER66REN 1978 p.1192, p i.1/15-16. Seaiailithella rotula (Kaiptner) BACKHAN 1980 p.52, p i.1/14-15, 8/3; THEODORIDIS p i.3/1-2. Uabilicosphaera lordii VAROL 1982 p.24B, p i.4/5. Description: Small/medium sized circular coccolith, with wide, open, central area, and numerous (c.50) elements. Weakly birefringent in cross-polars, with brighter central column; dark in phase contrast. The structure of this species is not well documented but can be discerned from electron micrographs (eg STEINMETZ & STRADNER 1984). The distal shield is slightly domed and is formed of shallowly imbricated elements, giving a complex suture pattern (Fig.39/G). The proximal shield consists of a single cycle of non-imbricated elements with nearly radial sutures, it is nearly flat except for a collar on its lower surface. A well developed central column separates the shields. It is not clear if the birefringence is a product of the proximal shield and column or of the entire coccolith. With damage elements are often lost from the distal shield. Size: 4 - 7 microns.

Genus Hayaster BUKRY 1973. Hayaster perplexus(Bramlette & Riedel 1954) Bukry 1973 Plate 2/17; Figure 39/F Siscoaster perplexus BRAMLETTE k RIEDEL 1954 p.400, p i.39/9; BLACK St BARNES 1961 p.144, p i.24/1; etcetera. Hayaster perplexus (Brailette St Riedel) BUKRY 1973c p.308; 0KADA St McINTYRE 1977 pi.5/4-5; NISHIDA 1979a p i.4/2; etcetera. Description: Placolith with only 10-14 rays and a very small proximal shield. Dark in cross-polars. Size: 5-10 microns, larger in recent material than in the Early Miocene. Range: Early Miocene - Recent, most common in Early Miocene and in Late Pliocene deposits.

161 Genus Solidopons Theodoridis 1984. Solidopons petzae Theodoridis 1984 Plate 2/31-33 Solidopons petrae THEODORIDIS 19B4 p.B6, p i.6/1-7. Description: Elliptical coccolith with a narrow rim and broad, open, central area spanned by an arched bridge. Optical characters similar to C.pelagicus. Remarks: This species is very rare but highly distinctive. Its range in my material is similar to that recorded by Theodoridis. In the absence of electron micrographs it is difficult to be confident of the structure but in cross-polars the rim strongly resembles that of Coccolithus pelagicus, hence inclusion in the Coccolithaceae seems reasonable. The bridge is strongly birefringent, with inclined extinction. Size: 5 - 7 microns.

Genus Oolithotus Reinhardt in REINHARDT & COHEN 1968. Oolithotus fragilis (Lohroann 1912) Okada & McIntyre 1977 Plate 2/31-33 Coccolithophora fragilis LOHMANN 1912 p.49f fig .11. Oiscolithus antillarun COHEN 1964 p.236, p i.1/3, 2/2. Oolithotas antillarun (Cohen) COHEN It REINHARDT 1968 p.297, pi.21/1,5,7. Oolithotus fragilis (Lohaann) OKADA It He INTYRE 1977 p .ll, p i.4/3-5; NISHIDA 1979a p i.5/3-4. Description: Sub-circular placolith with 15-20 curved elements and offset centre. Distal shield dark in cross-polars, proximal shield weakly birefringent. Remarks: As noted by OKADA & McINTYRE (1977) this species is rather similar to H.perplexus in construction, and in modern distribution. It is also, however, similar to C.leptoporus, from which it is distinguished by the offset centre and by the very weak birefringence of the proximal shield. Size: c.4-8 microns. Range: Pliocene - Recent.

Genus Umbilicosphaera Lohmann 1902 Remarks: This genus has been used in various senses, it seems sensible to restrict it to species with the same structure as the type species Coccolithus type shields without a wall, or grill.

Umbilicosphaera sibogae (Weber van Bosse 1901) Gaarder 1970 Description: Small to medium size circular species of l/mbilicosphaera, with open central area. Varieties: Two varieties of U.sibogae are generally recognised in modern nannofloras, U.s.sibogae and U.s.foliosa, the coccoliths of these differ

162 primarily in distal shield morphology. In U.s.sibogae (Fig.fVH) the distal shield is smaller than the proximal shield and composed of simple clockwise imbricated elements. In U.s.foliosa (Fig.*P3/F) the distal shield is larger and has peculiar anti-clockwise directed overgrowths (?) on the outer part of the rays. In addition the inner parts of the rays are facetted and the central opening is narrow. Intermediates, which are similar to U.jafari, occur but are not common. The coccospheres of the two varieties are also different (Fig.15, p.67), those of U.s.sibogae are larger and often contain more than one cell (eg LOHMANN 1902, INOUYE & PIENAAR 1984). There is also a weakly elliptical species (or variety?) U.hulbertiana GAARDER 1970. Remarks: The bicyclic proximal shield has not been widely noted but is well illustrated by for instance, ROTH (1970), NISHIDA & KONDA (1974), and INOUYE & PIENAAR (1984). Also only the proximal shield shows birefringence in plan view, this is quite evident on careful examination. These features suggest affinity to the Coccolithus group. Range: Pleistocene to Recent. Size: 3-8 microns (typically 4-5 microns in my material).

Umbilicosphaeza sibogae (Veber van Bosse 1901) Gaarder 1970 var. sibogae Plate 1/30; Figure 15/F, 43/G-H Coccosphaera sibogae WEBER VAN BOSSE 1901 p.140, p i.17/1-2. UBbilicosphaera Birabihs LOHMANN 1902 p. 139, p i.5/66; He INTYRE i BE 1967 p.571, pi. 11/b-c; NISHIDA It KONDA 1974 p i.5/7-11; BACKHAN 1980 p i.8/4; etcetera. Ub bilicosphaera sibogae (Weber van Bosse) GAARDER 1970 p.126, pl.9/c-d; 0KADA & He INTYRE 1977 p.13, pi.4/2; NISHIDA 1979a p i.5/2; etcetera.

Umbilicosphaeza sibogae var. foliosa (Kamptner 1963) Okada & McIntyre 1977 Plate 2/26-27; Figure 15/E, 39/A, 43/F Cfdoplacolitbas foliosus KAHPTNER 1963 p. 167, p i.7/38. llBbilicosphaera Birabilis Lohaann 1902, HclNTYRE It BE 1967 p.571, pi. 12/a; NISHIDA l< KONDA 1974 p i.5/7-11; etcetera. llBbilicosphaera sibogae (Weber van Bosse) GAARDER 1970 p.126, pl.9/c-d; ROTH 1973 p i.2/2,4. llBbilicosphaera sibogae var. foliosa (Kaaptner) OKADA & HclNTYRE 1977 p.13, p i.4/1; NISHIDA 1979a p i.5/1; INOUYE It PIENAAR 1984 figs.9-14; etcetera.

Umbilicosphaeza jafari Muller 1974 Plate 2/28-29 llBbilicosphaera jafari HULLER 1974b p.394, p i.1/1-3, 4/43-44; HAQ It BERGGREN 197B p. 1191, pi. 1/37-37; VAR0L 1985 p i.1/6-10. Cyclococcolithas krejcigrafii JAFAR 1975a p.74, p i.8/4-5.

163 Cjclococcolithas stradaeri JAFAR 1975a p.75, p i.9/22-25 (iabilicospbaera sibogae (Weber van Bosse) 6aarder, HAB & BER66REN 1978 pi. 1/41-42. Seaiailitbella jafari (Muller) BACKHAN 1980 p.52, p i.1/16-17. Uabilicospbaera petal iforais VAROL 1982 p.251, p i.4/7; VAROL 1985 p i.1/10-11. Description: Small circular Umbilicosphaeza. Remarks: The shields often separate. Isolated distal shields show no birefringence and are almost invisible in cross-polars; but are dark in phase contrast, and so discernible. Isolated proximal shields show the reverse behaviour; they are practically invisible in phase contrast but give a weak extinction cross in cross-polars. The entire coccolith has the combined properties; and so is visible in both phase contrast and cross-polars, although due to its small size it is not obvious that the birefringence is a product of the proximal shield. Varol's SEMs of the proximal shield (as U.petal iformis) show that it is flat and has the typical coccolithaceaen bicyclic structure. This is also just discernible in the type illustrations of MULLER (1974b), although she confused the proximal and distal sides, and in (unreproducible) SEMs of mine. U.jafari is distinguished from U.sibogae primarily by its smaller size, but it also has a smaller central opening, and the sutures on the distal shield appear to be straight. Because of its smaller size it is also much less strongly birefringent than U.sibogae. Size: 3 to 4 (rarely 5) microns. Range: Early Miocene to Late Pliocene, almost certainly ancestral to U.sibogae. Often abundant in well preserved samples.

Umbilicosphaeza jafari var.A. Plate 1/22-23 Description: Larger variety of U.jafari with open central area. Remarks: This variety occurred in the Early Miocene material I examined from Sites 219, 223, and 242, and similar specimens, transitional to U.sibogae; occur in the Pliocene. Size: 4-5 microns.

164 FIGURE 39 - COCCOLITHACEAE MORPHOLOGY Distal (left) and proximal (right) views, and schematic cross-sections (below) of representative species. Shaded portions in A-D indicate cycles which show birefringence in plan view. All diagrams are based on data from a number of sources rather than single specimens, are interpretative, and not to scale. A - Umbilicosphaeza sibogae; B - Coccolithus pelagicus ; C - Calcidiscus leptoporus ; D - Clausicoccus fenestratus; E - Oolithotus fzagilis; F - Hayaster perplexus; G - Geminilithella zotula; H - Cozonocyclus nitescens. 165 I

FIGURE 40 - STRATIGRAPHIC RANGES OF COCCOLITHACEAE. Based on synthesis of my own and published data. Line thickness variations represent shifts in size distribution, not abundance variation. Sporadic occurrence indicated by broken lines. A Events used in biostratigraphy. 166 /

NH19 S=4 H=132

NH16-18 S=8 N=127

NN15 S=4 H=38

HH12-14 S=5 N=i21

HH11 S=14 H=291

NN18 S=18 H=324

HH9 S=3 N=125

HN7-8 S=9 H=328

NN6 S=5 N=173

HH2-5 S=7 N=249

NP25-NH1 S=3 H=95 A 0 5 10 15 20 C.pelagicus size distribution, all sites.

HH21 S=1 N=34 NH19 (e arly) S=4 H=68 HH17-18 S=8 H=364 HH16 S=5 H=162 HH15 S=18 N=378 HH12-14 S=6 H=161 HH11 S=7 H=181 HH9-10 S=6 H=196 HH6-8 S=8 H=186 HH2-5 S=2 H=38 B 0 5 10 15 20 Calcidiscus size distribution, all sites,

FIGURE 41 - COCCOLITHUS PELAGICUS AMD CALCIDISCUS LEPTOPORUS SIZE VARIATION. Individual histograms show size distribution of all the speciaens aeasured fro# Indian Ocean DSDP samples of the given age interval. Shaded portion of histograms on C.pelagicas diagra# represents specimens with bridges. Intonation on the right of each histograa is: age interval (Standard Nannoplankton Zonation); number of saaples that the data coaes froa (S); and nuaber of speciaens (N). Horizontal scale in aicrons. 167 168

FIGURE 42 - Distribution of (A) Coccolithus pelagicus, (B) Clausicoccus p r i m alis, and (C) C y c l o p e r f olithus cazlae. Each figure shows distribution of the species in the Deep Sea Drilling Project samples examined (see Chapter 8 for details). Vertical axis is age, numbers indicate standard nannofossil zones, scaling after HAQ et al (in press) - zones NN1-5 at reduced scale. Samples are arranged across the diagram by site, in approximate North - South sequence (locations on Fig.28). All examined samples are indicated. Dots indicate samples without the relevant species; rectangles relative abundance. FIGURE 43 - COCCOLITHUS PELAGICUS AND UMBILICOSPHAERA SIBOGAE VARIETIES. A-E Coccolithus pelagicus group, distal views. Scale bar five microns. A. C.eopelagicus’r B. C.pelagicus var. miopelagicus; C. C.pelagicus var. p ontus; D. C.pelagicus var. pelagicus; E. C.pelagicus var. nannopelagicus. F-H. Umbilicosphaeza sibogaef distal and side views (see also Fig.15). F. U.sibogae var. foliosa; G. Intermediate variety; H. U.sibogae var. sibogae. 169 CHAPTER 12 - PRINSIACBAE Systematics

The structure, morphological variation, and stratigraphical distribution of the Prinsiaceae has been discussed at length above (Chapters 4 & 7), so only synonymies and brief notes are given here. In various places I have used the informal term reticulofenestrids, this includes all Neogene Prinsiaceae but excludes certain Palaeogene genera with different shield structure. Taxonomy of Neogene reticulofenestrids is also discussed at length in HAQ & BERGGREN (1978), BACKMAN (1980), and PUJOS (1985a-c, 1987).

FAMILY PRINSIACEAE Hay & Mohler 1967

Genus Cyclicargolithus Bukry 1971 Cyclicargolithus floridanus (Roth & Hay 1967) Bukry 1971 Plate 3/7-8, 20, 24; Figure 12/C. Coccolithus floridanus ROTH & HAY in HAY et al 1867 p.445 p i.6/1-4. Cyclococcolithus neogannation BRAHLETTE 4 WILC0X0N 1967; etcetera. Coccolithus? abisectus MULLER 1970 p.92 p i.9/9-10, 12/1. bictyococcites abisectus (Huller) BUKRY It PERCIVAL 1971 p.127, p i.2/9-11. Cyclicargolithus floridanus (Roth It Hay) BUKRY 1971c p.312; etcetera. Cyclicargolithus bukryi WISE 1973 p.594, p i.9/1-3. Cyclicargolithus abisectus (Muller) Wise 1973 p.594; etcetera. keticulofenestra floridana (Roth It Hay) THE0D0RIDIS 1984 p.85, pi.5/8. Description: Circular or weakly elliptical reticulofenestrid, with a small central opening and broad shield (a grill has not been observed). Remarks: As a consequence of the small central area this species has fewer rays than other reticulofenestrids of similar size; and so the rays have broad ends, often resolvable in the light microscope. The species varies significantly in size, and less markedly in ellipticity and central opening size. I did not study this in detail, but the various species which have been distinguished on this basis (abisectus, bukryi) would probably be better regarded as varieties. Occurrence: C. floridanus dominates Late Oligocene and Early Miocene nannofloras, then is replaced by R.pseudoumbilica, finally disappearing in NN6. Size: c.4-10microns in my material. Smaller forms (<5microns) seemed to be replaced by R.pseudoumbilica before the larger forms.

170 Genus Dlctyococcites Black 1967 Dlctyococcltes davlesil (Haq 1968) Pezch-Mlelsen 1971 Plate 3/9-10; Figure 12/A. Stradaerius daviesii HAQ 1968 p.32, p i.4/4-5. Dictyococcites daviesii (Haq) PERCH-NIELSEH 1971 p.29, p i.20/1-2. keticulofeaestra daviesii (Haq) HAQ 1971b; BACKHAN 1980 p.58, p i.6/6-9. ?keticulofeaestra gartaeri Roth & Hay, STRADNER & ALLRAH 1981 p i.2/1-3; STEINHETZ & STRADNER 1984 p i.28/4,8. Description: Elliptical; reticulofenestrid with central area spanned by about twelve radial bars which fuse to form a central body. Remarks: Oligocene reticulofenestrids are in need of revision, and the synonymy above is tentative. The species, as illustrated in my plates, is however, quite distinctive. It has a rather small, birefringent, central area sharply separated from the rim by a cycle of pores. Occurrence: ft.Bocene to Early Miocene, always rare in my material. Size: 8-11 microns (my material).

Dictyococcites scissura (Hay, Mohler & Wade 1966) Bukry & Percival 1971 Plate 3/1; Figure 12/B Syracospbaera bisecta HAY, MOHLER & HADE 1966 p.3113, plJO/1-5

keticulofeaestra scissura HAY, MOHLER i HADE 1966 p.387, partis p i.5/1,5-6 (bos pi.5/2-4); GARTNER 1971 p.112. D.daaicus BLACK 1967 p.141, fig .2. k.bisecta (Hay, Hohler fc Hade) ROTH 1970 p.874, p i.3/6; ROTH 1973 p i.4/1, 7/4-5, 9/1/2, 10/2. p.75, pi.2/7,11,12. b.bisectus (Hay, Hohler fc Hade), BUKRY k PERCIVAL 1971 p.127, pi.2/12-13; etcetera. Description: Elliptical reticulofenestrid with the central area entirely covered on the distal side by broad laths. The proximal side has a grill. Remarks: The nomenclatural taxonomy of this species has been extremely confused, see GARTNER (1971) for a succinct discussion. The specific epithet scissura has page priority over bisecta, although bisecta is often used. D.scissura is distinguished from C.floridanus by its ellipticity and broad filled central area. Possible intermediates do, however, occur in some samples and the two species seem closely related. Size: c.6-15microns, typically 8-llmicrons (medium sized forms (?<10microns) sometimes distinguished as D.scrippsae). Range: ft.Eocene - NN1.

Genus Emiliania Hay & Mohler 1967 Emiliania huxleyi (Lohmann 1902) Hay & Mohler 1967 Plate 3/23; Figures.6/B, 9/D, 12/G, 15/G-I. Poatosphaera huxleyi LOHHANN 1902 p.130, p i.4/1-6, 6/69.

171 Coccolithas haxlefi (Lohiann) KAHPTNER 1943 p.44; etcetera. £»iiia tia haxlefi (Lohiann) Hay It Kohler, in HAY et al 1967 p.447, p i.10/1-2; etcetera. Remarks: E.huxleyi is the dominant living coccolithophorid, and has been the subject of intensive research: e.g. McINTYRE & BE 1967r BURNS 1977, NISHIDA 1979a (variation in living forms); de VRIND - de JONG et al 1986, WESTBROEK et al 1986 (biological research); PUJOS 1977 (variation in fossil forms).

Genus Gephyzocapsa Kamptner 1943 Plate 3/5-6, 22, 26-27, 30-31; Figure 12/H. Description: Reticulofenestrids with a bridge across the central area, formed from elements of the wall. Remarks: Gephyzocapsa is a predominantly Pleistocene genus and so X have not examined it in detail. Its taxonomy is fascinatingly complex, and almost certainly would benefit from reconsideration in terms of intraspecific variation. Important discussions are given by GARTNER (1977), PIRINI RADIZZANI & VALUERI (1977), BREHERET (1978), SAMTLEBEN (1980), RIO (1982), and PUJOS (1977, 1985a,b). Of these I agree most strongly with Rio. I only separated smaller (<4microns) and larger (4-6microns) specimens. Occurrence: Smaller forms NN15 to recent, often abundant in samples from round the NN15/NN16 boundary. Larger forms approximately NN19 to recent, with gap in occurrence during the Pleistocene "small Gephyzocapsa interval" (refs, above), after which they dominate most assemblages. This interval seems closely analogous to the "small Reticulofenestra interval" in the Upper Miocene (Chapter 7). Various magnetostratigraphic studies give similar ages for it (c.9-1.1 Ma, GARTNER 1977, SAMTLEBEN 1980, RIO 1982).

Genus Pseudoemiliania Gartner 1969 Pseudoemiliania lacunosa Kamptner 1963 Remarks: As described above (Chapter 4; Figure 13) this species displays considerable size related variation. For routine work two varieties can conveniently, be distinguished, (see also Fig.44B, p.177).

Pseudoemiliania lacunosa Kamptner 1963 var. lacunosa Fig.l2/E, 13/A-B. Ellipsoplacolithus lacunosus KAHPTNER 1963 p.172 p i.9/50. PseudoeMiliania lacunas a (Kaiptner) 6ARTNER 1969 p.598, p i.2/9-10; ROTH 1973 p i.3/4; 6ARTNER 1977 p.15, p i.4/3; SAMTLEBEN 1979 p i.3/2,5,13; NISHIDA 1980 p i.1/13.; etcetera, f i iliania anula (Cohen) BUKRY 1971d p.1514; PR0T0-DECIMA et al 1978 p i.1/14-15. Description: Circular to sub-circular variety of P.lacunosa, with more than twelve rim slits.

172 Remarks: The degree of slitting is visible by light microscopy in reasonably well preserved material, and forms the most convenient arbitrary criterion for division. Size: c.4-7 microns.

Pseudoemiliania lacunosa var. ovata (Bukry 1973) n.comb. Plate 3/28; Figure .13/C-E Coccolithus ninutulus GARTNER 19 6 7 b p i . 5/3-4-. Eniliania ovata BUKRY 1973b p.678, pi.2/10-12; BUKRY 1975 p.690; PROTO-DECIHA et al 1978 p i.1/13. Pseudoeniliania lacunosa (Kaiptner) Gartner, ROTH 1973 pi.3/6; SAMTLEBEN 1979 p i.3/2-3,5-6; BACKHAN 1980 pi.6/12-13. Reticulofenestra reticulata var. Nishida, NISHIDA It KONDA 1974 p i.7/34-35. Reticulofenestra pacifica NISHIDA 1979b p.106, p i.1/4-6; NISHIDA 1980 p i.1/9. Description: Elliptical variety of P.lacunosa with only a few slits. Size: c.3.5-5.5 microns.

Genus Reticulofenestra Hay, Mohler & Wade 1966 Reticulofenestra doronicoides (Black & Barnes 1961) Pujos 1987 Coccolithus doronicoides BLACK it BARNES 1961 p. 142, p i.25/3; etcetera. Coccolithus ninutulus GARTNER 1967b p.3, pi.5/3 Sephyrocapsa reticulata NISHIDA 1971 p.150, p i.17/1-3. Crenalithus doronicoides (Black It Barnes) ROTH 1973 p.731, p i.3/3; SAMTLEBEN 1979 p i.3/4, 9-10; etcetera. Reticulofenestra japonica NISHIDA 1979b (non. substit. pro 6.reticulata) p.105, p i.1/1-3; NISHIDA 1980 p i.1/1,5,17. Reticuloienestra ninutula (Gartner) HAB It BERGGREN 1978 p.1190, BACKMAN 19B0 p.59, p i.7/3-5, 11-13. Reticulofenestra doronicoides (Black It Barnes) PUJOS 19B5c (invalid), p.594, pi. 1/7; PUJOS 1987 p.247, p i.1/13-17. Description: Smallish undistinguished reticulofenestrid. Remarks: As discussed above (Chapter 7) there is a major change in reticulofenestrid assemblages at the Early/Late Pliocene (NN15/16) boundary. Undistinguished smallish Reticulofenestra specimens occur throughout the Pliocene; but in the Early Pliocene they grade into R.p.pseudoumbilica, whereas in the Late Pliocene they grade into Pseudoemiliania lacunosa. Also R.p.minuta is largely replaced by small Gephyrocapsa. It seems sensible to make this a taxonomic boundary. Thus in effect R.p.hagii changes name at this level, to R.doronicoides. I have retained the combinations R.doronicoides, and P.lacunosa, in order to minimise nomenclatural innovation. A more rigorous solution would be to describe them as varieties of a single species. Size: 3-5 microns.

173 Reticulofenestra pseudoumbilica (Gartner 1967) Gartner 1969 Remarks: As discussed in Chapter 7 this species varies considerably in size, as a result of which it is convenient to recognise three varieties. Figure 44 (p.177) outlines the size variation pattern, and indicates the size and age of the holotypes of the principal described species. The taxonomy adopted is based on the premise that a unified system should be adopted for the entire range of R.pseudoumbilica(also discussed in Chapter 7). I have not separated forms with closed central areas but this variation is of importance at high latitudes, and a parallel set of varieties might usefully be used for these forms.

Reticulofenestra pseudoumbilica Gartner 1967 var. pseudoumbilica Plate 3/11-12, 21,25, 29; Figure 12/D. Coccolithus pseudouBbilicus 6ARTNER 1967b p.4, p i.6/1-4. Reticulofenestra pseadouBbilica Gartner GARTNER 1969 p.598, pi.2/4; etcetera. Description: Medium to large variety of R.pseudoumbilica. Size: 5-10 microns (occasionally up to about 12 microns).

Reticulofenestra pseudoumbilica var. hagii Backman 1978 n.comb. Plate 3/3. Reticulofeaestra hagii BACKMAN 1978a p.29, p i.1/1-4, 2/10; etcetera. ReticalofeBestra sp. HAQ It 8ERGGREN 1978, p i.1/23-26. Description: Smallish variety of R.pseudoumbilica. Remarks: Although the name R.minutula is often used in this sense the holotype has a very wide central area and reduced ellipticity, the paratypes have slitted shields, and Pseudoemiliania lacunosa occurred in the type material. I suspect it is a member of the P.lacunosa group. Additionally the epithet minutula is extremely confusing. So use of the epithet hagii is preferable. Size: 3-5 microns.

Reticulofenestra pseudoumbilica var. minuta (Roth 1970) n.comb. Plate 3/4. Reticulofenestra ainiita ROTH 1970 p.850, p i.5/3-4; BACKMAN 1980 p.58, p i.7/1-3; etcetera. Priosias i ioutus HAQ 1971b p.78, p i.6/4-5. Description: Very small variety of R.pseudoumbilica. Remarks: The holotype of this variety is Oligocene, but the epithet has predominantly been used in this sense. Unlike R.hagii, this might be a separate species; but it is impossible to consistently separate, other than on arbitrary criteria. Coccolithus taganus FONSECA 1976 is similar, but with a closed central area. Size: c.2-3 microns.

174 Reticulofenestra pseudoumbilica var. rotaria (Theodoridis 1984) n.comb. Plate 3/2. R.rotaria THEODORIDIS 1984 p.B5, pi.5/1-4. Description: Circular, or very weakly elliptical, variety of R.pseudoumbilica with a wide central opening. Remarks: The shape of this species sharply distinguishes it from normal R.pseudoumbilica. Nonetheless it is identical in other respects and intermediates between the two varieties occurred in all the samples in which I found it. Hence I consider it a variant of R.pseudoumbilica rather than a separate species. Size: 5-7microns. Range: Short interval within the 1NN11 Zone (=R.rotaria subzone of THEODORIDIS 1984), my observations agreed well with those of Theodoridis.

"Genus"Pyrocyclus Hay & Towe 1962, Plate 3/13-19. This genus was described from the Eocene, but BACKMAN (1980), and subsequently other authors (see species list below), have ascribed Neogene specimens to it. They described them as coccoliths with a single shield composed of two cycles; a lower cycle of numerous steeply imbricate elements and an upper cycle of fewer tabular elements. In the light microscope these specimens have a Reticulofenestra type extinction cross, but with

X j e om IIj cy-i/J,£ovtr ph'te.s o r col ^ tubi c ^ d tS . 175 Described "species" of Pyzocyclus Pyrocyclas inversus HAY fc TOME 1962 p.33f p i.6/6; BACKHAN 1980 p.55, pi.3/2-4,8-13. Pyrocyclus hernosus ROTH & HAY 1967, in HAY et al, p.488, p i.6/10-12; HAQ 1971c p.117, p i.16/1-4; BACKHAN 1980 pi.3/1,14-17; HULLER fc LEH0TAY0VA 19B1 p i.7/5-6, 10/3. Coccolithas? orangensis BUKRY 1971c p.312, p i.2/10-11, 3/1/3; STEINHETZ & STRADNER 1984 p i.28/5-7. Pyrocyclas orangensis (Bukry) BACKHAN 1980 p.56, p i.3/6-7; Unbilicosphaera sp. HAQ It BERG6REN p i.1/39-40.

176 FIGURE 44 - A. Reticulofenestra holotypes; B. Subdivision of Pseudoemiliania lacunosa group. A. Outline of size variation pattern, fro« Chapter 7, with the naaes of various described species plotted according to the age and size of their holotypes. D. (Dictyococcites) - indicates holotypes with closed central areas. H.tinata and D.besslaadi have Oligocene holotypes. B. Data set as in Section 4.4.2. Horizontal axis length (aicrons), vertical axis nuaber of slits. These are the easiest parameters to observe, rather than the lost significant. 177 CHAPTER 13 - HELICOSPHAERACEAE

The Helicosphaeraceae aie a distinctive monogeneric family the coccoliths of which (termed helicoliths) are frequently very abundant in Neogene sediments. Helicoliths resemble placoliths, but whereas placoliths have two clearly separated shields giving rise to a continuous double margin, helicoliths are more complex. They have a helical structure giving rise to a double margin on one side but a single margin along most of the other side. This asymmetric form allows a regular and close interlocking of the helicoliths on the coccosphere. The single edge of one helicolith tucks into the double edge of the next one (Fig.6/A, p.33). The coccospheres also differ from those of typical placoliths, they are usually elongate rather than spherical, there is only a single layer of coccoliths, and the coccoliths are organised in an orderly spiral with small coccoliths at one end, and larger ones at the other end, around an opening. This is probably a flagellar opening, although the only published observations of flagella are those of LOHMANN (1902). In some specimens at least the wings of the helicoliths surrounding the aperture are expanded. These appear to be analogous to the specialised circum-flagella coccoliths of Syzacosphaeza, and other genera. Fossil Helicosphaeza has been discussed in numerous papers including KAMPTNER (1954, structure), BLACK & BARNES (1961, structure), CLOCHIATTI (1969, structure, nomenclature), HAQ (1973, systematics), JAFAR & MARTINI (1975, nomenclature), ROMEIN (1979, structure, Eocene systematics), THEODORIDIS (1984, structure, systematics), PERCH-NIELSEN (1985b, general review). By contrast information on living Helicosphaera is confined to scattered observations and micrographs, notably in WALLICH (1877), LOHMANN (1902), GAARDER (1970), BORSETTI & CATI (1972), and NISHIDA (1979a).

13.1 STRUCTURE OF HELICOLITHS As indicated above, numerous authors have described the structure Helicosphaeza, so the following notes are a brief outline only, concentrating on the less well understood aspects. As THEODORIDIS (1984) demonstrated, helicoliths are formed of three structural units: the proximal plate; the flange, which spirals around the proximal plate usually ending in a distinct wing; and the blanket which covers the plate and overlaps partially onto the flange. The proximal plate is constructed of elongate radially arranged elements (Plate 4/26). These appear to have grown inwards, since they

178 aie uniformly spaced around the periphery of the plate, and adjacent elements extend inwards to varying distances. The edge of the proximal plate thus probably corresponds to the proto-coccolith ring of reticulofenestrids. Organic basal scales have been noted beneath the proximal plate of modern species by GAARDER (1970), and illustrated by NORRIS (1971). The flange is constructed of elongate elements radiating outward from the periphery of the proximal plate (Fig 46/G, p.199, Plate 4/27). These are rather like the shield elements of reticulofenestrids and other placoliths. However, in placoliths the shield elements are all virtually identical, whereas in helicoliths they vary continuously to produce the spiral flange. There appears to be a one-to-one correspondence between the elements of the flange and proximal plate, and possibly they constitute a single cycle of complex elements, analogous to those of reticulofenestrids. The blanket is constructed of numerous small concentrically arranged elements (Plate 4/24). It is thus similar to the wall of Coccolithus, irrftju ta r and to the A inner cycles of PontosphaeraA. The blanket only develops on the proximal plate and flange, if the proximal plate does not close the central area then the blanket is similarly restricted. In these species the central opening is often spanned by a bridge but this has a different construction, being formed of larger elements than those of the blanket. Optical orientations: The blanket is strongly birefringent in plan view, giving an oblique extinction cross. Bridge elements of the type described above are also birefringent, but with a different optical orientation. The flange is not birefringent. It is harder to determine the optical behaviour, of the proximal plate since it is usually covered by the blanket. However in some specimens, notably of H.c.sellii and H.c.wallichii, the pores in the blanket are larger than those in the proximal plate. In these cases it appears that the proximal plate does not show birefringence (Plate 4/5-6). Affinities: The optical orientation, and the structure, of the proximal plate suggest that Helicosphaera has closer affinities to Pontosphaera than to the Coccolithaceae (compare Figs.39/E,G,I and Fig.39/A-D); as suggested by HAY et al (1967), THEODORIDIS (1984). These groups are, however, quite distinct throughout their range (Palaeocene to Recent).

179 13.2 TAXONOMIC GROUPS Helicoliths vary considerably, particularly in mode of flange development, central area structures, and size. This very real variation has resulted in numerous species being described, as revieved by HAQ (1973), THEODORIDIS (1984), and PERCH-NIELSEN (1985b), with a few more species described since. Living species, however, show significant intraspecific variation (Chapter 5, Figure 15/A, p67), and it seems likely that many of the described species should be regarded as only of varietal status, as suggested in the taxonomy section. This process leaves four mainAspecies groups, or lineages, as outlined below, and shown on Figure 45 (p.188).

1. H.carter! - intermedia group. (Plate 4/1-6, 11-20, 22-28). These are medium to large species (typically 7-11 microns) with flanges that end in distinct wings. They are the most common helicoliths in all my samples, but vary markedly through the Neogene. Forms with optically discontinuous bars (H.intermedia) dominate in the Early Miocene but are replaced progressively by forms with closed proximal plates. The blankets on these are initially rather chaotic (H.c.granulata) but forms with well ordered blankets with two in-line pores (H.c.carteri) become more common through the Late Miocene. In the Pliocene, forms with large and/or oblique pores (H.c.sellii and H.c.vallichii) become common. All co-existing morphotypes seem to intergrade, although not always in single samples and it is my impression that this is a single evolving lineage (as suggested in Fig.45). The principal morphological division is between the forms with a discontinuous bridge and the others. This forms a convenient means of dividing the lineage. It is nonetheless essentially arbitrary, since many H.c.granulata type specimens have partially discontinuous central areas, and no other feature parallels the division.

2. H .recta - obliqua - orientalis group. This is the second-most common group in my material, they are distinguishable by their flange form, which is quite distinct from that of the H.carteri group. There is no distinct wing, instead the flange terminates abruptly at the apex of the ellipse, often with a spike-like expansion of the last elements. The flange also commences relatively abruptly. In addition the helicoliths have more angular shapes, lower birefringence, and pores inclined in the opposite direction to those of the H.carteri group (toward rather than away from the flange termination). The bridge is always formed from the proximal plate, and

180 is covered by small elements in optical continuity with the rest of the blanket. Although they occur together with the H.carter! group there is rarely any problem in separating species of the two groups, even when poorly preserved. They are genuinely separate species. The group also contains at least three distinct species - H.recta, H.obliqua, and H.orientalis. H.compacta is a likely ancestral species, it has a morphology in some respects intermediate between that of H.recta and the earlier H.seminulum group.

3. H.ampliaperta - nediterranea group. This seems to be a distinct Early Miocene group, but it was very rare in my material. The species are distinguished by weakly developed wings, large to medium size and rather open central areas. On the basis of flange form they probably are related to the H.carteri group, but derivation from H.recta is also conceivable. The main species are H.ampliaperta BRAMLETTE & WILCOXON 1967 (open central area) and H.mediteiranea (with a narrow bridge, synonym H.crouchii BUKRY 1981). Others are H.waltrans THEODORIDIS 1984 (oblique bridge), H.scissura MILLER 1981 (slit-like central opening), and H.porosa LEHOTAYOVA 1985 (delicate grill over central area).

4. H.valbersdorfensis - stalls group. This group consists of small (<7microns) species, mainly recorded from the Middle Miocene. Their wings are typically not expanded, but extend around the single margin, in this respect they are most like the Eocene H.seminulum group, but probably developed from the H.carteri, or H.ampliaperta, group. Most species have well developed oblique pores in the central area, visibly oblique in both cross-polars and phase contrast. Described species - H.walbersdorfensis MULLER 1974b, H.californiana BUKRY 1981, H.vedderi BUKRY 1981, H.minuta MULLER 1981, and H.stalls THEODORIDIS 1984. Extremely rare in my material, more common in the Mediterranean (THEODORIDIS 1984, Varol pers. comm. 1987).

13.3 BIOSTRATIGRAPHICAL USE Various Helicospbaera species, particularly from the H.obliqua and H.ampliaperta groups, have been proposed as biostratigraphical markers, MARTINI & WORSLEY (1970), GARTNER (1977), THEODORIDIS (1984), as indicated in Figure 45. However, the abundance, and seemingly the ranges of species vary markedly from section to section and in consequence few' of these are very satisfactory markers. For instance

181 the last occurrence of H.recta was originally proposed as the NP25 - NN1 zonal boundary, but H.recta has repeatedly been found to continue, at low abundances, into the Early Miocene (eg at Site 242 in my material). Similarly various proposed marker species are very rare or absent in the Indian Ocean material, including H.ampliaperta (which is also often absent in the Pacific), H.mediterranea, and H.stalls, whilst others have extended ranges, for instance H.Intermedia, and H.obligua. It appears that ecological factors, and perhaps provinciality, influenced the distribution of these species. Most obviously they are more abundant and diverse in marginal sea assemblages (e.g. those from the Red Sea Sites) than in open ocean assemblages. This effect makes it harder to derive information from the group, but also makes them potentially more interesting.

13.4 SYSTEMATICS N.B. I have only given brief synonymies for the main species since they are widely illustrated; eg HAQ (1971a-c), HAQ (1973), PERCH-NIELSEN (1977), THEODORIDIS (1984, with good synonymies).

Family Helicosphaeraceae BLACK 1971 Genus Helicosphaera KAMPTNER 1954 Synonym Helicopontosphaera Hay & Mohler in HAY et al 1967. Type species Coccosphaeza carteri Wallich 1877. Remarks: Nomenclatural problems were created by the incomplete initial description of this species by WALLICH (1877), this lead HAY et al (1967) to propose a new type species and genus (Helicopontosphaera kamptneri). This is unnecessary, and so invalid, as discussed by CLOCHIATTI (1969) and by JAFAR & MARTINI (1975), with whom I agree. (The root cause of the problem was probably that Wallich only sketched the central areas of the helicoliths in his illustrations of H.carteri coccospheres].

13.4.1 HELICOSPHAERA CARTERI GROUP.

Helicosphaera carteri (Wallich 1877) Kamptner 1954 Plate 4/1-3,5-6,22-28; Figures 6/A, 15/A. Coccosphaera carteri WALLICH 1877 parti* p.34B, p i.17/3-4 bob p i.17/6-7,12,17 (= C.pelagicus var. pontus), Helicosphaera carteri (Wallich) KAMPTNER 1954, p.21, fig s.17-19; BQRSETTI It CATI 1972 p i.52/1-4: NISHIDA 1979a p i.9/1,3-4; etcetera. Helicopontosphaera katptneri Hay It Mohler in HAY et al 1967, p.44B, p i.10/5, 11/5; etcetera. Helicosphaera burkei BLACK 1971 p.618, p i.45.3/23. Helicosphaera acuta THEODORIDIS 1984 p.119, pl.18/9-11.

1B2 Helicosphaera paleocarteri THEODORIDIS 1984 p.131, p i.23/5-9. Description: Helicoliths with flange terminating in a more or less extended wing, rather than ending abruptly, or merging with the beginning of the flange. Proximal plate closed or with slit-like openings, without structurally (and optically) distinct bridge. Remarks: This is the dominant species from the mid-Miocene to the recent and shows considerable variation in size, flange width, wing expansion, and pore development. Pore form in particular has been used to define various species, but these species are otherwise similar, and intergradational, so are better regarded as varieties.

H.caztezi var. carter! (Vallich 1877) Kamptner 1954 Plate 4/3,24. Description: Variety of H.caztezi with two in-line pores in the blanket, which may or may not extend through the proximal plate. Range: Early Miocene to Recent, dominant variety in the Late Miocene and Pliocene.

H.carter! var. granulata (Bukry & Percival 1979) n.comb. Plate 4/2,23. Helicopontosphaera grami lata BUKRY & PERCIVAL 1971 p.132, p i.5/1-2. HAQ 1973 p i.6/1-2, 7/7-8; HEKEL 1973 p i.4/1-4; PERCH-NIELSEM 1977 p i.25/1-6, 26/2,4,6. Description: Variety of H.caztezi with poorly developed blanket in the central area; the elements of the blanket are blocky with gaps between them. The proximal plate seems to be entire. The wing is often very large, even in small specimens. Remarks: THEODORIDIS (1984) has suggested that this is a preservational variant, it does become more common in poorly preserved sediments. I found it, however, in rather well preserved assemblages from Site 242, and I suspect that it is in part at least a primary morphotype. Range: Miocene, the most common H.caztezi variety in the Early and Middle Miocene.

H.caztezi var. hyalina (Gaarder 1970) n.comb. Helicosphaera hyalin a GAARDER 1970 p.113, pl.l/a-g, 2/a-d, 3/a; CATI fc BORSETTI 1972 p i.52/3-4; NISHIDA 1979a p i.9/1. Helicosphaera carteri var. burkei (Black) THEODORIDIS 1984, p i.23/6-7. Description: Variety of Helicosphaera without pores in the blanket or proximal plate. Remarks: Specimens of H.carteri var. carteri can also have imperforate proximal plates, hence it is not possible to distinguish these two varieties from proximal view electron micrographs, either distal views or light micrographs are needed.

183 Range: Middle Miocene to Recent (my obs.), the variety becomes somewhat more common in recent sediments but can be found throughout the range of H.carteri, and can be produced by overgrowth.

H.carter} var. sellii (Bukry & Bramlette 1969) n.comb. Helicopoatosphaera sellii BUKRY & BRAMLETTE 1969 p. 134, p i.2/3-7; etcetera. Helicopootosphaera carteri (Uallich) Kaiptner, STRADNER 1973 partiw p i.15/1-2. Description: Variety of H.carter} with blanket in central area confined to a narrow bridge, between two large pores. Openings in the proximal plate usually smaller than the pores in the blanket, and obliquely orientated (for this reason there are fewer illustrations of the proximal than the distal side). Remarks: The wing is rarely well developed, but neither is it in most co-occurring H.c.carteri specimens. Range: Pliocene - earliest Pleistocene, its last occurrence has been suggested as a subzone boundary (GARTNER 1977), but has not proven very reliable (BACKMAN & SHACKLETON 1983).

H.carteri var.vallichii (Lohmann 1902) Theodoridis 1984 Plate 4/5-6. Coceolithophora uallichii LOHMANN 1902 p.138, p i.5/58-60. Helicosphaera uallichii (Lohiann) OKADA & Me INTYRE 1977 p.14, p i.4/8. Helicosphaera carteri (Wallich) Ka»ptner, NISHIDA 1979 pl.9/4a-c. H.carteri var.uallichii (Lohiann) THEODORIDIS 1984 p.133, p i.23/8-9. Description: Variety of H.carteri with strongly oblique openings in the proximal plate, and more or less oblique pores in the proximal plate. Remarks: Since the pores in the proximal plate and blanket do not closely correspond it is very difficult to consistently differentiate this variety from H.c.sellii and H.c.carteri. Range: Late Miocene (NNll) to Recent, oblique pores are generally characteristic of the Pliocene H.carteri specimens.

Helicosphaera intermedia MARTINI 1965 Description: Helicoliths with H.carteri type flanges, and so distinct wings. Proximal plate does not close central area, which is spanned by a, structurally and optically distinct, oblique bridge. Remarks: These helicoliths vary considerably in the strength and orientation of their bridge, this variation is continuous and so I have treated the two main types as varieties rather than species. H.truempyi BIOLZI & PERCH-NIELSEN 1982 might be considered a third variety (with thin sub-horizontal bridge), I did not, however, encounter it.

184 H.intermedia var. intermedia MARTINI 1965 Plate 4/11-14, 17, 19-20. Helicosphaera intermedia MARTINI 1%5 p.404, pi.35/1-2; etcetera. Helicopontosphaera rhoaba BUKRY 1971c p.320, p i.5/6-9; HAQ 1973 p i.4/2, 7/10; PERCH-NIELSEN 1977 pi.24/3,5-6. Description: Variety of H.intermediawith bridge inclined at an angle of less than about 60p to the short axis of the ellipse, and so leaving distinct openings above and below it. Bridge often S-shaped. Range: Oligocene to Early Pliocene, but rare above mid Miocene.

H.intermedia var. euphratis (Haq 1966) n.comb. Plate 4/18. Helicosphaera euphratis HAQ 1966 p.33, p i.2/1,3; etcetera. Helicosphaera parallela BRAMLETTE & WILC0X0N 1967 p.106, p i.5/9-10. Description: Variety of H.intermediawith bridge steeply inclined , to the short axis of the coccolith(>c.60°), and so occupying all or virtually all of the central opening. Typically rather elongate and without an extended wing. Range: Oligocene and Early Miocene, the dominant Oligocene member of the H.carteri lineage it is progressively replaced by H.i.intermedia and H.c.granulata in the Early Miocene.

Helicosphaera inversa (Gartner 1977) Theodoridis 1984 Plate 4/4. Helicopontosphaera inversa 6ARTNER 1977 p.23, p i.1/4-5. Helicosphaera inversa (Gartner) THE0D0RIDIS 1984 p.120. Remarks: This species is similar to H.c.sellii but has the bridge inclined slightly upwards. The pores in the proximal plate seem to have a similar form. In my material H .inversa occurred in two samples (219-1-1 and 219-1-3), and in both these it was readily distinguishable from co-occurring H.carteri specimens, it may be a genuine species. Range: Late Pleistocene.

13.4.2. HELICOSPHAERA OBLIQUA GROUP.

Helicosphaera obligua Bramlette & Vilcoxon 1967 Plate 4/7-8 Helicosphaera obligua BRAMLETTE & HILC0X0N 1967 p,106, p i.5/13-14; etcetera. Helicopontosphaera perch-nielseniae HAQ 1971c p.U6, p i.10/5-7; etcetera. Helicosphaera elongata THE0D0RIDIS 1984 p.117, p i.17/6-9. Description: Helicoliths with flange which commences and terminates

185 abruptly, at one end of the coccolith - rather than part way along the side as In H.carter!. The final elements of the flange are usually elongated into a distinct spike. There are two oblique pores, inclined toward the flange termination, in the central area, with elements of the proximal plate extending between them. Rather low birefringence. Remarks: H.pezch-nielseniae was originally distinguished from H.obliqua by the smaller size of its pores. Subsequently THEODORIDIS (1984) suggested emended diagnoses based on coccolith outline (H.obliqua smaller and more elongate). I did not find either division useful, but I have not looked at much Oligocene material. Range: Oligocene to Middle Miocene, last occurrence usually in NN4 or 5, at Site 242 extends into NN6.

Helicosphaeza ozientalis Black 1971 Description: Species with small helicoliths with truncated wing, rectangular outline, and low birefringence, pores when present inclined upwards. Remarks: This species has its first occurrence within zone NN6, and so just overlaps in range with H.obliqua (Fig.45), both species occur in one sample of mine (242-7-1). It shares with H.obliqua low birefringence, upward orientated pores, and a sharply truncated flange (in H.caztezi the wing is gradually ended by a sequence of progressively shorter elements, whereas in H.obliqua and H.orientalis the last elements of the flange are the longest). Differentiated from H.obliqua by its smaller size, larger pores, and lack of a spike on the flange termination.

H.ozientalis var. orientalis Black 1971 n.comb. Plate 4/9-10, 21.

Helicosphaera orientalis BLACK 1971 p .6 1 9 , p i . 4 5 .3 / 2 2 ; THEODORIDIS 19B4 p . i 18, p i . 13/4, 17/11, 18/2-5.

Helicopontosphaera orientalis (Black) JAFAR 1975a p.77, p i.9/16-17,19-20.

Helicosphaera pbilippinensis MULLER 1981 p . 429, p i . 1/7-12. Description: Variety of H.ozientalis with two pores in the central area. Pores normally rather large but can be closed by overgrowth. Range: Middle - Late Miocene, NN6-eNNll, only Middle Miocene in my material.

H.ozientalis var. pacifica (Muller & Bronniman 1974) n.comb.

Helicosphaera pacifica MULLER fc BRONNIMAN 1974 p . 6 6 l , p i . 1/1-10; THEODORIDIS 1984 p . U 9 , p i . 18/6-8. Description: Variety of H.orientalis with proximal plate and blanket not reaching the centre, leaving a central opening. This opening is

186 floored by a finely perforate plate, or grill, which is non-birefringent in plan view. Remarks: Apart from the central area this form appears identical to H.o.ozientalis, and H.o.orientalis specimens often have only narrow bridges in the central area. Delicate grills have also been illustrated in other helicoliths and are possibly of low taxonomic significance. Range: Middle - Late Miocene, about NN7-11, in my material rare occurrences in eNNll only.

Helicosphaera recta (Haq 1966) Jafar & Martini 1975

Helicosphaera seaitulua recta HAQ 1966 p .34, p i.2/6, 3/4.

Helicosphaera trancata BRANLETTE i WILCOXON 1967 p .1 0 6 , p i . 6/13-14.

Helicosphaera recta (Haq) JAFAR & MARTINI 1975 p.391; etcptera. Description: Moderately large helicoliths with sub-rectangular outline, sharply truncated flange (with spike), two large openings in the central area separated by bridge parallel to the short axis of the coccolith. Remarks: Separated from H.obliqua by its larger size, more massive construction and larger pores. I did not find any ambiguous specimens.

187 FIGURE 45 - STRATIGRAPHIC DISTRIBUTION OF HELICOSPHAERACBAE. Based on synthesis of my own and published data. Arranged to show likely phylogenetic relationships. Column thickness related to importance in my material, sporadic occurrence/uncertain range indicated by broken lines. A Events used in published zonation scheme (relation to other events shown in Figure 18). N.B. Vertical scale changes at NN5/6 boundary. Yellow (set p.'so).

1 8 8 CHAPTER 14 - PONTOSPHAERACEAE

Introduction The Pontosphaeraceae have bowl shaped coccoliths which abut simply on the coccosphere to form a single layered test (Fig.46/B, p.198). In the Neogene only two genera are important, Pontosphaera and Scyphosphaera. In Pontosphaera all the coccoliths are of one type, although the coccoliths on one coccosphere do show noticeable variation in ellipticity and length. In Scyphosphaera there are two types of coccoliths, low ones similar to those on Pontosphaera and others with an elevated rim, lopadoliths (Fig.46/A). Only Scyphosphaera is discussed in detail here.

14.1 STRUCTURE. A. Pontosphaera. The structure of Pontosphaera has been discussed by various authors including STRADNER & EDWARDS (1968), BURNS (1973), and ROMEIN (1979). As they have shown, typical Pontosphaera coccoliths are formed of two types of elements (Fig.46/D-F). A. Outer elongate elements; these form a thin outer layer to the coccolith, in two parts. First a basal plate of radiating laths. Second an outer wall of numerous steeply imbricate laths (termed the flange by Burns). These elements tend to be easily damaged, particularly around the edge of the base, where they meet. As a result of this damage it is not clear whether they constitute a single cycle or two discrete cycles. ROMEIN (1979) suggests, however, that they are in optical continuity. B. Inner short elements; these form the bulk of the coccoliths. In plan-view they are short and concentrically arranged . They are similar to the wall elements in Coccolithus and to the blanket elements in Helicosphaera. As in those genera these elements are formed of crystals with their optic axes orientated approximately horizontally, and sub-radially relative to the coccolith (the optic axes are thus oblique to the element elongation). Typically these inner elements forma perforate base and an imperforate wall to the coccolith, with the elongate outer elements below and around them. In some species the perforations continue through the base-plate in others they are confined to the inner cycle (Fig.46/1 vs. E-F).

I B. Scyphosphaera. The low coccoliths of Scyphosphaera have identical structures to those of Pontosphaera coccoliths. The elevated coccoliths appear rather different, but their structure is a straightforward development of the

189 Pontosphaera structure, as discussed below. In the light microscope, the wall of the elevated coccoliths can be seen to consist of two lamellae, with different optical orientations (Fig»46/H-I, PI.5/13). The outer lamella is constructed of elements with their optic axes nearly vertical, and consequently behaves as a single unit optically. The inner lamella is constructed of elements with their optic axes horizontal and sub-radial. In well preserved specimens both lamellae are continuous over the entire coccolith, although often the outer lamella is missing from the basal region. Occasionally the two lamellae can separate entirely, I have observed both isolated fragments of outer lamellae, and smooth specimens with only the inner lamella preserved. In the SEM, Scyphosphaera coccoliths typically show an apparent three-fold structural subdivision of the outer lamella, (Fig.46/H, PI.5/21), with: a basal zone of imbricate slender elements; a central reticulate zone; and an apical zone of elongate ribs. Individual elements can, however, be traced through the three zones in electron micrographs, and in the light microscope the outer lamella is definitely continuous over these three areas. The three zones thus mark changes in form of elements in a single cycle, rather than three discrete cycles. The change from the imbricate to the reticulate zone is interesting since it is a distinct structural change, and is not seen on Pontosphaeza coccoliths. It involves a change in element growth direction, from inclined to vertical, and a reduction in the number of elements. The reticulation is not produced by horizontal elements, but rather by intermittent expansion ot the vertical elements, possibly these are growth lines. The change from reticulate to elongate ornament normally coincides with the maximum width of the coccolith, it is probably not very significant. In forms which taper from the base up, (e.g. S.conica and S.intermedia) the reticulate zone is absent or weakly developed. The lower part of the outer lamella is clearly identical to the outer wall, or flange, of Pontosphaera, and so the entire outer lamella must be a development of this wall. Similarly the inner lamella is probably developed from the short elements of the inner wall. This is supported by the optical orientation of the inner lamella, and by a few SEM micrographs which show concentrically orientated elements in the apical region of Scyphosphaera spp. (e.g. STRADNER 1973, pi.20/3-4). The inner lamella usually is thickened around the opening, forming a collar (Fig.46/H). This can be useful for distinguishing the top from the base in poorly preserved specimens. In some species the wall is continued above the collar as a neck, this is formed from both lamellae. All basal views of Scyphosphaera are rather similar, with two or

190 three cycles of small pores. Radial structure is never obvious but is discernible in some of the plates of STRADNER (1973, eg pi.20/5-6). The variation shown in published basal views could be accommodated in a single Pontosphaera species. Summary: Elevated Scyphosphaera coccoliths have a constant structure, different forms are produced by changes in shape rather than in structure. The uniformity of this structure, particularly the imbricate zone to reticulate zone transition, and the constant form of the basal plate, suggest that the different Scyphosphaera species are closely related, and provides support for grouping them in one genus. The structure is an elaborate modification of the typical Pontosphaera structure, but does not involve the production of any extra elements.

14.2 TAXONOHIC SUBDIVISION OP SCYPHOSPHABRA In both genera the species level taxonomy is rather chaotic. In Pontosphaera species are mainly defined on the number and size of pores and on rim width. Rim height and shape in section are also variable, but harder to study in the light microscope. I have not attempted any detailed work with the genus. Fossil Scyphosphaera species are inevitably based entirely on the elevated coccoliths, and due to the structural uniformity of these, shape in profile has been essentially the only specific criterion. The most important descriptive studies are those of DEFLANDRE (1942) and KAMPTNER (1955, 1967), more recent studies by JAFAR (1975a,b) and RADE (1975) have developed this work. Over forty species have been described by these workers from Neogene and Quaternary material (also reviewed by PERCH-NIELSEN 1985b). There are various reasons for thinking that this is too high a number. Most obviously the lopadoliths of the one living species, Scyphosphaera apsteinii, show considerable variation in form, covering at least half-a-dozen fossil "species" (ampla, antilleana, apsteinii, cohenii, recurvata and recta). Further Kamptner, who described most of the species, explicitly stated in various places that he used very fine species concepts in order to differentiate similar forms, which probably did not correspond to biological species. For example "Venn es auch fraglich erscheint, ob alle bisher publizierten Lopadolithen-Formen von Scvphosphaera erblich fixierte Typen vorstellen, so ist es doch aus praktischen Griinden am Platz, dass wir de neue Form hier als Spezies i benennen." (KAMPTNER 1963, remarks on S.aequatorialis). His approach was valid and, with subsequent additions, has resulted in a scheme which covers the range of observed morphology. This scheme is not, however, useful in practice, virtually all the species have similar ranges,

191 assigning specimens to single species is rarely easy, some species are virtually identical, and there is little likelihood that the taxonomic species approximate to biological species. I have given below a rather tentative, reorganised taxonomy based on my own observations, the original descriptions of the species, and published illustrations of Scyphosphaeza specimens. This is also illustrated in Figure 47 (p.199). Essentially I have reorganised the described morphotypes into a few major categories which are more likely to correspond to genuine species. For the S.apsteinii group, which dominates most Scyphosphaeza assemblages I have also given a subdivision into four varieties.

14.3 STRATIGRAPHICAL DISTRIBUTION OF SCYPHOSPHAERA. Scyphosphaeza specimens occur from the Eocene to the Recent, but they are only common or diverse during the Late Miocene and Pliocene (PERCH-NIELSEN 1985b). This generally observed pattern is closely reflected in my material, Scyphosphaeza specimens only occur in samples assigned to zones NN9 to NN19, and are most common in the NN11 to NN15 interval. The few records of earlier Miocene or Oligocene occurrences suggest that the S.apsteinii var. zecuzvata group was predominant (BRAMLETTE & WILCOXON 1967, MARTINI 1981, VAROL 1984). The first diverse assemblages known are from zone NN9. The main morphological radiation seems to have occurred in the mid-Miocene, but the record is still too patchy to be certain. The only species with a clearly restricted range is S.globulata, which is confined to the Late Miocene and Pliocene, where it is often common. Diversity reduction seems to have occurred during the Early Pleistocene, although as with the previous radiation there is no obvious pattern to it. This distribution pattern is summarised in Figure 48 (p.200). It is not very promising for biostratigraphy, in addition the genus is normally rare and easily dissolved. Nonetheless they may have some value in the NNll to NN15 interval when they are most common, since this interval is hard to zone. In the Indian Ocean, and particularly the Red Sea material, Scyphosphaeza was more abundant, and easier to find, than ceratoliths, which provide the principal zonal indices.

192 14.4 SYSTEMATICS FAMILY PONTOSPHABRACBAB Lemmerman In Brandt & Apstein 1908 Genus Pontosphaeza Lohmann 1902 Type species: P.syracusana Lohmann 1902 Plate 5/1,3,26, Fig.46/B,D-F.

Genus Scyphosphaeza Lohmann 1902 Type species: S.apsteinii Lohmann 1902.

Scyphosphaeza apsteinii Lohmann 1902 Description: Species without neck or elongated body. The body is typically convex from the base up.

S.apsteinii Lohmann 1902 var. apsteinii Plate 5/2,12-15,22; Figures 15/B, 46/A,C.

S.apsteinii LOHMANN 1902 p .1 3 2 , p i . 4/26-30; DEFLANDRE 1942 p .6 , f i g s . 10-15; KAMPTNER 1967 p i . 9 /6 4 -6 5 ,6 7 , 10/69-70; B0RSETTI It CATI 1972 p i . 41/3, 42/1-2; STRADNER 1973 p i . 2 1 /4 -6,

22/3,5-6; MULLER 1974a p i.15/7; JAFAR 1975a 1/1,4,7,9; CEPEK k WIND 1979 p i . 4/1, 6/3; E LLIS k LOHMANN 1979 p i.5/4; NISHIDA 1979a p i.11/1.

S.apsteinii Lohaann foraa a p s t e i n i i , 6AARDER 1970 p l . 4 / e - f .

Scyphosphaera sp.3 SAMTLEBEN 1979 p i.4/7,10. Description: S.apsteinii variety with maximum width near the centre. Remarks: The typical form of living S.apsteinii (as illustrated by Deflandre, Gaarder, Borsetti & Cati, and Nishida) is very similar to that of Lohmann's type illustration, with a rather simple convex outline, they vary markedly in size and in height to breadth ratio. Only forms of this morphology are included in the synonymy above with variants included in the other groups below. Range: Mid Miocene (?earlier) to Recent.

S.apsteinii var. dilatata (Gaarder 1970) n.comb.

S . p o r o s a KAMPTNER 1967 p . 151 f i g . 22.

S . c o h e n i i BOUDREAUX k HAY 1969 p.278, p i.7/5-6; MULLER 1974a p i.13/4; PERCH-NIELSEN 1977 p i.29/12.

S.antilleana BOUDREAUX k HAY 1969 p .2 7 8 , p i . 7/3-4.

S.apsteinii fo ra a d i l a t a t a GAARDER 1970, p . 119, p l. 4 / a - d , 5 /a-d .

S.apsteinii Lohaann, PERCH-NIELSEN 1977 p i.29/16. S .c f .p o r o s a K a a p tn e r, STRADNER 1973 p i . 25/6.

Scyphosphaara s p . MULLER 1974a p i . 14/1. Description: S.apsteinii variants which flare outward continuously. Remarks: This is the most distinctive living variety of S.apsteinii, but GAARDER (1970), showed that they had identical normal coccoliths to those of S.apsteinii, and that S.a.dilatata lopadoliths could be

193 produced by S.apsteinii, so it is unlikely that they are a separate biological species. Range: ?Pliocene to Recent.

S.apsteinii var. recurvata (Deflandre 1942) n.comb. Plate 5/18-19

S . r e c a n a t a DEFLANDRE 1942 p . 132, f i g s . 17-20; BRAMLETTE l WILCOXON 1967 p i . 10/6-7; MULLER 11974a p i . 14/7-9; PERCH-NIELSEN 1977 p i . 2 9 / 5 ,1 4 ,1 5 ; SAHTLEBEN 1979 p i . 4/4; MARTINI 1981 p i . 1/1-2, 5/13-16; BERGEN 1984 p i . 10/3.

S.apsteinii v a r . r e c t a DEFLANDRE 1942 p . 131, f i g . 16.

S . r e c t a (D e fla n d re ) KAMPTNER 1955 p .2 3 , p i . 8/115-116; JAFAR 1975a p i . 2/5-6; PERCH-NIELSEN 1977 p i . 29/13; RAFFI & SPROVIERI 1385 p i . 5/17.

S . a n p l a KAMPTNER 1955 p .2 3 , f i g . 118; JAFAR 1975a p i . 2/2, 3/1; JAFAR 1975b p i . 1/1; BERGEN 1984 p i . 10/4.

S.global osa KAMPTNER 1955 p .2 3 , p i . 8/113; MULLER 1974a p i . 14/4-6; JAFAR 1975a p i . 1/8.

S.pirifornis KAMPTNER 1955 p .2 3 , p i . 8/117; JAFAR 1975a p i . 2/1; JAFAR 1975b p i . 2/6.

S . p r o c e r a KAMPTNER 1955 p .2 3 , p i . 8/108,114; JAFAR 1975a p i . 4 /2 -5 . S.iDverjicoDica VAR0L 1984, p.380, p i.3/9,13.

S.apsteinii Lohaann, BRAMLETTE fc WILC0X0N 1967 p i . 1 0 / 1 -2 ,4 ; SAMTLEBEN 1979 p i . 4 / 2 ,5 ; REID 19B0 p i.3/7; HARTINI 1981 p i.5/11-12. Description: Pear-shaped variety with a narrow base relative to maximum width, and with the position of maximum width near the apex. Above this position the walls curve inward to a variable extent. Remarks: This group includes the majority of fossil Scyphosphaera coccoliths. As with S.apsteinii var. apsteinii there is marked variation in overall width and in the detail of shape. This has lead to the definition of the numerous species listed above. All these forms are very similar to living S.apsteinii and I prefer to treat the whole group as one variety of S.apsteinii. REID (1980) has illustrated a modern coccosphere with coccoliths resembling both S.a.recurvata and S.a.apsteinii. Range: Palaeogene to Recent.

S.apsteinii var. A

S.apsteinii Lohaann, PERCH-NIELSEN 1972 p i.20/6-7; STRADNER 1973 p i.21/1-3, 22/4, 23/1-2. Description: Variety of S.apsteinii with maximum width near the base, at or just above the top of the imbricate zone of the wall. Remarks: This form is probably as different from the typical form as the other varieties noted above and is necessary for a consistent scheme, it is in no sense a separate species from S.apsteinii. It seems to be intermediate between S.a.apsteinii and S.globulata. Range: ?Upper Miocene and Pliocene, rare in my material.

194 "S.cylindrical group

S.cylindrica KAHPTNER 1955 p.24, p i.9/119; JAFAR 1975a p i.1/12-13; JAFAR 1975b pl.3/la2.

S . g a l e a i a KAHPTNER 1967 p .1 4 9 , p i . 9/68, f i g . 1 9 ; JAFAR 1975a p l.3 / 1 1 ; JAFAR 1975b p i . 2 /3 -4 .

S . p e n n a KAHPTNER 1955 p .2 4 , p i . 9/122; JAFAR 1975b p i . 2/7. Description: Broad, open, coccoliths with vertical or slightly inflected sides. Remarks: Forms with this morphology are non-descript, rare, hard to place, and probably best ignored. The species described by KAMPTNER are probably variants of the S.pulcherrima group with very weak collars, (compare illustrations in JAFAR 1975a of these two groups of species).

S.globulata Bukry & Percival 1971 Plate 5/23

S.globulata BUKRY & PERCIVAL 1971 p . l3 B , p i . 7 /1 -6 ; HULLER 1974a p i . 16/3-5; PERCH-NIELSEN 1977 p i.29/1-2; etcetera. S . c f . S . a p $ t e i » i i Loh#ann, BUKRY 1971c p i . 3/6; HARTINI 1981 p i . 5/17-20.

S.globulosa Kaaptner 1955, PR0T0-DECIHA et al 1978 p i.3/3-4. Description: Virtually spherical coccoliths, with a very small aperture and broader base. Remarks: JAFAR (1975a) suggested that S.globulata was a junior synonym of S.globulosa, this was based, however, on an assumption (his p.27) that the type illustration was upside-down, and so that the position of maximum width was near the apex, as in the recurvata group. This is not correct. Range: Late Miocene and Early Pliocene, common in my material and widely reported. Less common, possibly absent, in the Late Pliocene. Total range perhaps mid NN11 to NN16.

S.intermedia Deflandre 1942 Plate 5/16

S.interaedia DEFLANDRE 1942, p.134, figs.32-36; STRADNER 1973 p i.22/4; HULLER 1974a p i.15/10-12; PERCH-NIELSEN 1977 p i.29/10-11; ELLIS & LQHHANN 1979 p i.2/6. Description: Species with a rather low body, and long broadly flaring neck, no distinct collar. Remarks: True to its name this species is intermediate in form, between S.lagena and S.pulcherrima. Nonetheless it seems distinctive. Range: Middle Miocene to ?mid Pliocene.

195 S.lagena Kamptner 1955 Plate 5/20

S.catescens KAMPTNER 1955 p .2 4 f p i . 9/120; JAFAR 1975a p i . 2 /7 -8 .

S.ca»tharellas KAMPTNER 1955 p .2 4 f p i . 9/123; RADE 1975 p i . 1/8. S .c o n ic a KAMPTNER 1955 p .2 6 f p i . 9/130,131; MULLER 1974a p i . 15/9; JAFAR 1975a p l.3 / 8 - 1 0 .

S . l a g e t a KAMPTNER 1955 p .2 5 f p i . 9/124,127; JAFAR 1975a p i . 4/1.

S . t a r r i s KAMPTNER 1955 p .2 6 , p i . 9/132; STRADNER 1973 p i . 2 4 /3 -5, 25/4; JAFAR 1975a p i . 3 /2 -3 .

S.aequatorialis KAHPTNER 1963 p . 176, f i g . 27; JAFAR 1975a p i . 3/14-15.

S . a b e l e i RADE 1975 p .1 6 0 , p i . 4 /4 -5 . S.apsteioii Lohaann, MULLER 1974a p i.13/5. S.cylindrica Kaaptner, MULLER 1974a p i.14/3. S .c f.a v p A o r a K a a p tn e r, STRADNER 1973 p i . 25/1-2. S.cf.S.recarvata PERCH-NIELSEN 1972 p i.20/5.

Scyphosphaera s p . STRADNER 1973 p i . 17/3, 38/1-2; SAHTLEBEN 1979 p i . 4/8. Description: Elongate lopadoliths with narrow bases, and maximum width near the base. Ornament usually weak, slight neck in some forms (turris, canescens), but no distinct collar. Remarks: This is a rare but distinctive group. Forms similar to the illustrations of S.lagena and S.conica were most common in my material, and definitely intergraded. Range: Mid-Miocene to Mid-Pleistocene.

S.pulchexzima Deflandre 1942 Plate 5/6-9,21,25; Figure 46/H

S.pulcherrita DEFLANDRE 1942 p . 133, f ig s .2 B - 3 1 ; STRADNER 1973 p i . 22/1-2; MULLER 1974a p i . 13/1-3; SAMTLEBEN 1979 p i . 4 / 3 ,6 ,9 ; PERCH-NIELSEN 1977 p i . 2 9 / 3 ,4 ,6 , ELLIS i LOHMANN 1979 p i . 2/7.

S . a i p h o r a DEFLANDRE 1942 p.132, figs.21-22; PERCH-NIELSEN 1972 p i.20/1-3; MULLER 1974a p i . 14/10-12; PERCH-NIELSEN 1977 p i . 29 /7 ,9 ; BERGEN 1984 p i . 10/2.

S.campanula DEFLANDRE 1942 p . 134, f i g s . 23-27; MULLER 1974a p i . 13/6-9.

S . c f . c a t pa n til a D e fla n d r e 1942, SAMTLEBEN 1979 p i . 4/1.

S.halldalii DEFLANDRE 1954 p . 137, f ig s .2 2 - 2 4 .

S . d a r r a g h i RADE 1975, p .1 6 2 , p i . 4 /7-8. Description: Coccoliths with well developed neck, barrel shaped body,' collar usually present. Usually rounded in horizontal section, and with strong ornament. Remarks: They show marked and seemingly independent variation in breadth, angularity and position of maximum width, so I eventually decided to treat them as one group. The most attractive possible subdivision is between forms with maximum width near the base (campanula and halldalii), and ones with a distinct flaring, and ornamented, body (pulcherrima and amphora). Range: Middle Miocene to mid Pleistocene (?NN7 to NN19).

196 S.tubifera group

S.tab if era KAMPTNER 1955 p .2 6 , p i . 9/133; JAFAR 1975a p i . 4/16-17 (NN9). S .ia rtiB ii JAFAR 1975a p.41, p i.1/2-3 (NN9).

S.quasitabifera VAROL 1984 p . 3 B l , p i . 3 /4 ,8 (NN6-7).

? Scyphosphaera eleqans ( O s te n fe ld ) DEFLANDRE 1942, p . 138, f i g s . 1,9 (R e ce n t). Description: Forms with small bodies and very elongate tubular necks. Remarks: These forms are very rare, apart from the original descriptions the only record I have noted is rare occurrence in the Pliocene of the South Atlantic (HAQ & BERGGREN 1978). More information is really necessary to be certain even that they are species of Scyphosphaeza. There is, however, a possibility at least that these constitute a long-ranging group separate from the main Scyphosphaera plexus. Given these uncertainties the species are simply noted as a group here.

S.ventriosa Martini 1968

S.veotriosa MARTINI 1968 p .1 6 9 , f i g s . 3 -4 .

S.globulata Bukry h P e r c iv a l 1972, STRADNER 1973.

S.apsteioii Lohaann 1902, STRADNER 1973 p i . 23/3-6.

Scyphosphaera sp . STRADNER 1973 p i . 17/2 HEKEL 1973 p i . 4 / 5 -6 .

S.deflandrei MULLER 1974a p.592, p i.13/10-12, 19/7-8. SJaiptneri MULLER 1974a p.592, pi. 15/6, 19/5-6.

S.rottiensis JAFAR 1975a p.42, p i.5/1-2.

S . p a c i f i c a RADE 1975 p . 158, p i . 3 / 3 ,6 -9 .

S.gladstoaensis RADE 1975 p . 158, p i . 3/7-8.

S.qaeenslatdensis RADE 1975 p . 156, p i . 2/9, 3/4-5.

Scyphosphaera s p .A RADE 1975 p .1 6 2 , p i . 4/1-2. Description: Lopadoliths with weak necks, and narrow openings, normally with distinct collars. Remarks: These forms seem to be intermediate between S.apsteinii and S.pulchezzima, perhaps fortunately, they are not common. Range: Middle Miocene to Pliocene.

197 FIGURE 46 - Pontosphaeraceae and Helicosphaeraceae, morphology. Principal sources (PERCH-NIELSEN 1971, NISHIDA 1979).A lr C, Scfphosphaera apsteinii. B It D -F , Poatosphaera a a l t i p o r a . 6, Helicospbaera lopbota, proxiaal. H, Structural coaponents of elevated Scfpbospbaera coccolith, S.palcherriaa. I, Coaparative sections of Poatospbaera , Helicosphaera (long section), and Scfpbospbaera, blanket/inner eleaents shaded, these are birefringent in plan view. 198 FIGURE 47 - Scyphosphaera holotypes. fV Outline diagrams of the holotypes of most cf the important described^species. These are organised so that, as far as possible,. intergradational forms are adjacent. The basic arrangement is horizontally according to breadth, and vertically according to shape: continuously flaring; flaring-recurved; flaring-recurved-secondary flaring. Taxa adopted are underlined. Bovis jn^i«t<. possible 199 FIGURE 48 - Stratigzaphic distribution of Scyphosphaera species, and varieties. Based on synthesis of ay ovn and published data. Sporadic occurrence/uncertain range indicated by broken lines, possible range extensions indicated by broken lines. N.B. Vertical scale changes at NN5/6 boundary.

200 CHAPTER 15 - SPHENOLITHACEAE

15.1. STRUCTURE Despite useful contributions by ROTH et al (1971), TOWB (1979) and ROHEIN (1979), the fine-structure of sphenoliths has not been clearly described. The interpretation given here is essentially my own, based on my observations, and published micrographs. The individual elements of sphenoliths are elongated parallel to their c-axis (clear from light microscopy), and consist of three lath-like segments arranged in a Y-shaped form (Fig.49/A, p.206). This morphology is visible in SEMs of specimens from samples with very good preservation, or slight etching (e.g. Plate 10/21-22,26). With overgrowth the spaces between the segments become infilled and the elements develop a spinose form (eg illustrations in PERCH-NIELSEN 1977). The triple lath form is probably related to the trigonal symmetry of calcite, with laths developing parallel to the x-axes. The elements radiate from a single origin, which gives the sphenoliths a compact form, and a clear extinction-cross, in polarized light. The lower-part of all sphenolith species is composed of a single cycle of 8 to 16 of these elements, conventionally termed the shield. The axes of the shield elements slope down from the median plane. A concave base is formed by laths from adjacent elements meeting. No elements point directly downward. The upper half of sphenoliths is normally formed of two or three cycles of elements, radiating from the centre, at decreasing angles to the vertical. The details of the structure are, however, variable; particularly toward the apex, which gives rise to a large amount of species level variation. The principal structures are: A. S.radians type, Figure 49/A: The final cycle of elements is not steeply inclined (only c.45°), but segments of each element extend upward from it, forming a composite spine. All species are Palaeogene, they include S.editus, S.orphanknollensis, S.pseudoradians, and S.radians (N.B. PERCH-NIELSEN (1985b) gives illustrations and taxonomic references for these species). B. S.abies type, Figure 49/B: Final cycle of elements steeply inclined, resulting in an elevated upper part, but no distinct apical spine structure. Species with this structure include S.abies, and S.moriformis var. dissimilis, (also S.neoabies, and S.verensis, both of which are better regarded as varieties of S.abies). S.anarrhopus, S.capricornutus and S.guadrispinatus have similar structures; but with respectively one, two and four of the final cycle elements elongated.

201 C. S.heteromorphosus type: One element vertically orientated, producing a monolithic spine, which behaves as a single optical unit. At least some SEM illustrations suggest that this spine consists of three lath-like segments, as might be expected. Species with this structure include S.belemnos, S.calyculus, S.delphix, S.elongatus, and S.heteromorphosus. D. S.distentus type: Upper part reduced to a large bipartite spine, possibly formed by fusion of laths (as in S.radians), from an otherwise indistinct cycle. Shield rather reduced, forming a girdle around the base of the spine. Species include S.distentus, S.predistentus, S.ciperoensis, and S.tribulosus; probably also S.obtusus (BUKRY 1971a), and S.intercalaris (MARTINI 1976). This is structurally the most distinct group of species, and appears to be monophyletic. It might be useful to create a separate genus for them. N.B. S.moriformis does not fit clearly into any of these groups since the name is applied to a range of forms without distinct apical enlargement. S.primus appears to have a rather disorganised arrangement of its apical elements. S.grandis (HAQ & BERGGREN 1978) is unusually large and does not have the typical sphenolith extinction pattern, it is probably not a sphenolith but an ascidian spicule

15.2 BIOSTRATIGRAPHICAL USE As indicated on Figure 50 (p.207) sphenoliths are not taxonomically diverse during the Neogene, but do provide several valuable marker events. All the marker species are distinctive and common, and the events are sharp. The only problem is that the more robust species, including S.heteromotphosus are liable to reworking; in the Makran material this species had to be used with extreme caution. The S.delphix - S.capzicornutus group of species (or one diverse species?) have only been recorded from around the Oligo-Miocene boundary, they may however range higher since there are records of S.belemnos below its normal first occurrence (basal NN3), and S.delphix is intermediate in morphology between S.moriformis and S.belemnos. The S.abiesframmoriformis transition is too gradational to be used as a marker event, but the species are sufficiently distinct to be used for low-resolution biostratigraphy. Biometric work on these species might be fruitful, since they vary considerably in size between samples; although biometric studies by TOWE (1979), and NAGYMAROSY (1987) have not been particularly fruitful.

202 15.3 SYSTEMATIC^ Family SPHENOLITHACEAE Deflandre 1952 Genus Sphenolithus Deflandre 1952 Type species S.radians Deflandre 1952.

Sphenolithus abies Deflandre 1954 Plate 10/4-5,22; Figure 49/C

Sphenolithus abies D e fla n d r e in DEFLANDRE fc FERT 1954 p .1 6 4 , p i . 10/1-4; PRDTD-DECIMA e t a l 1978 p i . 1/17; STEINNETZ i STRADNER 1984 p i . 41/9-10; e t c e t e r a .

Sphenolithus neoabies BUKRY & BRAHLETTE 1969 p . 140, p i . 3/9-11; ELLIS & LOHMANN 1979 p i . 4/10.

Sphenolithas verensis BACKHAN 1978a p . l l l , p i . 2 /4-6. Remarks: S.abies evolves from S.moriformis and is very similar to it, but well formed specimens can be distinguished by their greater elevation, cuspate outline, and the presence of an extinction line down the axis of the spine. S.verensis was distinguished only by slight elongations of the shield elements, and S.neoabies by its smaller size. I saw both morphotypes but did not distinguish them, they could be used as varieties. Size: 2-8 microns (size varies markedly between samples). Range: NN9-15, very common.

Sphenolithus belemnos Bramlette & Wilcoxon 1967 Plate 10/11-14,24; Figure 49/E

Sphenolithus belennos BRAMLETTE fc WILCOXON 1967 p-118, p i.2/1-3; ROTH et al 1971 fig .3, p i.4/1-8; PERCH-NIELSEN 1977 p i.34/5-12; THE0D0RIDIS 1984 p i.7/1-4; etcetera. Remarks: This species has a well developed monolithic spine with straight extinction. It can be distinguished from other species of this type by the nature of its base. This has a tall well-developed shield cycle, but only diminutive lateral elements, overall it is distinctly elongate, and merges smoothly into the spine (Fig.49B). Size: Length 4-8 microns, width 2-4 microns. Range: NN3.

Sphenolithus calyculus Bukry 1985 Sphenolithus calyculus BUKRY 1985 p . 600, p i . 1/13-19

Sphenolithus elongatus MARTINI 1981, p.753, p i.2/7-8. Remarks: One of Oligo-Miocene boundary group of delicate sphenoliths (see S.delphix). Characterised by a single long apical spine, and a compact base, in contrast to the flaring base of S.delphix. It is smaller and thinner than S.heteromorphosus. Range:NP25 and NN1.

203 Sphenolithus capricornutus Bukcy & Percival 1971 Plate 10/17

Sphenolithus caprieornutus BUKRY fc PERCIVAL 1971 p ,1 4 0 , p i . 6/4-6; MULLER 1974a p i . 5/1, 19/13-14; MARTINI 1976 p i . 3 / 4 -6 , 13/13-14; THE0D0RIDIS 1984 p i . 8 /4 -7 ; KNUTTEL 1986 p l3 / 3 -4 , 7 -8 . Remarks: One of the Oligo-Miocene boundary group of delicate sphenoliths (see S.delphix). Characterised by two diverging apical spines, and usually a flaring base. Range: NP25 and NN1.

Sphenolithus delphix Bukry 1973 Plate 10/18

Sphenolithus delphix BUKRY 1973b p .6 7 9 , p i . 3/19-22; MARTINI 1976 p i . 3/9-12, 13/13-14; KNUTTEL 1986 p l4 / l - 5 , 8.

Sphenolithus spp 1 & 2 MULLER 1974a, p i.4/3-8, 19/15-16. Remarks: This is the most common of a group of rather delicate spinose sphenoliths. The other described species are S.capricornutus and S.calyculus. All three have only been reported from zones NP25 and NN1, I suspect they may in fact be a single, rather variable, species. In my material they only occurred in Sample 242-9-1-100 (also recorded, and illustrated, by MULLER 1974a). S.delphix is characterised by a widely flaring shield cycle and a single moderately long apical spine. Size: Up to 8 microns long, but only 2-3 microns wide. Range: NP25-NN1.

Sphenolithus hetezomozphosus Deflandre 1953 Plate 10/23; Figure 49/F

Sphenolithus heteronorphosus DEFLANDRE 1953 p . 1785, f i g . 1 ,2 ; PERCH-NIELSEN 1977 p i . 3 4 /3 -4, 35/1-4; THEODQRIDIS 1984 p i . 7/5-12; e t c e t e r a .

Sphenolithus conicus Bukry 1371c; PERCH-NIELSEN 1977 p i.34/1-2, 35/1-4. Remarks: S.hetezomozphosus has a well developed quadrate base, which distinguishes it clearly from S.belemnos. Size: Length 5-15 microns, varies considerably in all populations, possibly in part due to diagenetic overgrowth. Width usually c.3-5 microns. Range: NN4-5 (rather prone to reworking).

204 Sphenolithus morlforrais (Bronniman & Stradner 1960) Bramlette & Wilcoxon 1967 Plate 10/3,7-9,21,25-27; Figure 49/D Hannotarbella tor Ho r n is BRONNIMAN i STRADNER I960 p .3 8 6 , f i g s . 11-16.

Sphenolithus pacificas MARTINI 1965 p .407, p i.36/7-10.

Sphenolithas norifornis (Bronnisan & Stradner) BRAMLETTE & WILCOXON 1967 p.124, p i.5/1-6; e t c e t e r a .

Sphenolithus conicas BUKRY 1971c p .3 2 0 , p i . 5/10-12.

Sphenolithas conpactas BACKMAN 1980 p.59, p i.3/20-21. Remarks: It is very difficult to separate low sphenoliths, and the various morphotypes seem to intergrade, so I have used S.moriformis as a generalised species for this group. Detailed work might show that the subgroups are biostratigraphically useful subgroups, but they would probably be better considered as varieties than species. S.compactus is a small form, its relationship to S.moriformis is similar to that of S.neoabies to S.abies. S.conicus was distinguished by having a triangular profile and an apical element which went into extinction as one unit, like a truncated S.heteromorphosus spine. I found that Early Miocene S.moriformis specimens often showed this morphology. S.pacificus is a well established junior synonym. Size: 3-8 microns (size varies considerably between samples, but I did not note any consistent trends). Range: Eocene to Mid Miocene (c.NN9), very common.

Sphenolithus moriformis var. dissimilis (Bukry & Percival 1971) n.comb. Plate 10/1-2,6; Figure 49/B Sphenolithas dissinilis BUKRY fc PERCIVAL 1971 p . 140, p i . 6/7-9; PERCH-NIELSEN 1977 p i . 3 3 / 4 ,8 .

Sphenolithas c f . S . c o n i c a s Bukry, EDWARDS & PERCH-NIELSEN 1974 p i.19/1,6-10.

Sphenolithas abies EDWARDS & PERCH-NIELSEN 1974 p i . 20/4. Description: Variety of S.moriformis with the apical cycle of elements extended to form a circlet of spines. Remarks: The structure of this variety is very similar to that of S.abies, but the apical elements are more strongly developed and more separated than in S.abies. It grades into more normal S.moriformis types. Size: 6-8 microns. Range: Early Miocene, rare in Site 242, Core 8, samples, other records of similar age.

205 A B

FIGURE 49 - SPHENOLITH STRUCTURE AND MORPHOLOGY.

A-B. Diagrams Illustrating construction of sphenoliths, from radiating elements composed of three laths. With overgrowth these elements become rod-like. A. S.radians; B. S.moriformis var. dissimilis.

C-F. Sketches of the principal Neogene species, as seen in light microscopy. Stippled - elements in extinction, when sphenolith is parallel to polars. C. S.abies; D. S.moriformis (conica type); E. S.belemnos; F. S.heteromorphosus.

206 FIGURE 50 - STRATIGRAPHIC DISTRIBUTION OF SPHENOLITHACEAE. Based on synthesis of ay own and published data. Arranged to show likely phylogenetic relationships. Sporadic occurrence/uncertain range indicated by broken lines. A Events used in zonation scheaes. CHAPTER 16 - DISCOASTBRACBAB

Introduction Discoasters are the most important component of Tertiary nannofloras for biostratigraphy and have been intensively studied, three major syntheses covering Neogene discoasters have recently been published, AUBRY (1984), THEODORIDIS (1984), and PERCH-NIELSEN (1985b). I present here a somewhat simplified reassessment, building on this work, through the application of my taxonomic concepts and a consideration of morphological homologies. In the systematics section I have discussed the genus Catinaster in extra detail, since it provides an interesting example of structural conservatism during radical morphological change.

16.1 STRUCTURE Although discoasters were abundant and diverse during most of the Palaeogene only a single species, D.deflandzei, survived in the Late Oligocene and all Neogene discoasters evolved from it (PRINS 1971). As a consequence they form an homogeneous group, and they are separated from typical Palaeogene species by a range of characters (N.B. a few members of the "Neogene" group do occur the Palaeogene). The most important of these characters are: A. Overall shape: Palaeogene discoasters typically have their rays in contact for most of their length giving a "rosette" shape, in contrast to "star" shaped Neogene discoasters. B. Ray shape: The rays of Palaeogene discoasters are often - curved and asymmetrical, whereas Neogene discoasters nearly always have straight bilaterally symmetrical rays. C. Number of rays: Palaeogene species usually have more rays (8-30) than Neogene ones (5 or 6). D. Ray attachment surface: The rays of Palaeogene species usually join along inclined and curved surfaces, whereas the attachment surfaces of Neogene species are planar and vertical (THEODORIDIS 1984). It is reasonable to treat the two groups as separate genera. Unfortunately there are complications in doing this, due to the fact that TAN (1927, 1931), who first described discoasters, coined a number of generic names without regard to nomenclatural legalities. THEODORIDIS (1983, 1984) argued that the name Discoaster was invalid and that instead the names Helio-discoaster and Eu-discoaster should be used for, respectively, the Palaeogene and Neogene groups. The validity of his argument has not been accepted, and probably requires a referral to the ICBN. Most workers have continued to use the name Discoaster in the traditional sense. As a convenient compromise, I have retained

208 Discoaster but also use eu-discoaster and helio-discoaster as informal terms, with the same sense as Theodoridis. Eu-discoaster morphology: The ancestral eu-discoaster species, D.deflandrei (Figure 53/D, p.230) has a rather simple form; the two sides are similar, and lack elaborate central area structures. Subsequent species show increased complexity, with development of a range of structures. There is, however, a pattern to these structures. In particular different structures occur consistently on the two surfaces. It is convenient to differentiate these two surfaces as proximal and distal. It is, of course, not known how or in what orientation discoasters where born on the nannoplankton cell. However, it seems reasonable to take the analogy of coccoliths and assume that for concavo-convex species the concave side was inmost. On this basis the concave side can be termed proximal, and the convex side distal, as recommended by FARINACCI (1971). The various structures developed are illustrated in Figure 51 (p.229/5), using D.surculus as an example, since it shows the greatest range of features. On the distal side there is a distinct central area formed by the rays widening and uniting. Within this central area the rays may be slightly depressed and/or separated by low sutural ridges. In the middle of the central area there is often a stellate knob the arms of which point toward the ray sutures. Away from the central area the rays extend nearly horizontally, with a rather flat distal surface, on which distal ridges may occur. These distal ridges are always confined to the rays, never running into the centre of the discoaster. Bifurcations occur at the tips of rays in many species, these are primarily formed from the distal surface of the ray. On the proximal side there is no clear central area / ray division, instead ridges run continuously from the ray into a central knob. This proximal knob thus has a radial stellate form, in contrast to the inter-radial distal knob (compare views in Fig.51, & Fig.53). The proximal ridges may run continuously along the rays and then build downwards, giving the discoaster a concavo-convex form. The contrast between proximal ridge and distal surface gives the rays an asymmetrical section (centre diagram Fig. 51). Since rays are formed of single crystals this division of the rays into separate structural elements is essentially artificial. Nonetheless they are recognisable and used together such features as convexity, central knob orientation, and central area development allow consistent differentiation of proximal and distal surfaces. Some species show much stronger development of the features of one side or the other. For instance D.brouweri has a well developed proximal side:

209 it has little or no central area, distal knob, or bifurcations; but the proximal blades, and proximal ridges are well developed, and there is i often a proximal knob. D.deflandrei by contrast has a virtually featureless proximal surface but well developed distal features (particularly central area and bifurcations). It is tempting to draw an analogy between the morphological consistency shown by discoaster rays and that shown by the elements of reticulofenestrid coccoliths (Chapter 4). In both cases morphological differentiation of species is created primarily by differential development of homologous structures rather than by creation of new structures. The analogy extends to ray number (which is variable in both cases), and to crystallographic orientation (which is constant). There is no direct analog for a proto-coccolith ring since eu-discoasters, like sphenoliths, seem to develop from a single centre. As a result ray spacing is not variable, and discoasters are never elliptical. It is possible that this analogy is meaningful and that discoasters were produced in the same manner as heterococcoliths. Crystallography: A partial alternative explanation for the constancy of form was provided by BLACK (1972); see Figure 52 (p. 229/D. From an analysis of crystal faces developed during overgrowth he showed that the rays of discoasters have radially symmetrical crystallographic orientations. He also showed that, as a result of the low symmetry of calcite, the two faces are crystallographically distinct. The effect of this is most consistent in the central area; on one side two crystal faces are readily developed resulting in a radial ridge. On the other side only one face is preferentially developed, leading to radial flats. BLACK (1972) termed these respectively the E- and F- surfaces. However they correspond to the proximal and distal surfaces, and the central area structures to radial knobs and depressions. It is not clear how many other features of discoaster, or coccolith, ray form can be related to this class of control, but it may be an important contributor to morphological regularity. Summary: Neogene eu-discoasters form an homogeneous group clearly separable from the Palaeogene helio-discoasters. The considerable morphological diversity that they show can be related to variable development of homologous structures, this forms an invaluable basis for understanding their taxonomy and evolution. The style of variation is analogous to that of heterococcoliths and suggests that they may have been formed by a similar developmental process. In addition crystallography appears to control some aspects of morphology.

210 16.2 SUBDIVISION, STRATIGRAPHICAL DISTRIBUTION AND EVOLUTION. Various authors have proposed evolutionary schemes for Neogene discoasters with the apparent assumption that they are genuine discrete species through most of their range, resulting in punctuated equilibrium type diagrams (e.g. PRINS 1971, THEODORIDIS 1984). In Figure 54 (p.231). I give a rather different scheme, based on the working hypothesis that much of the variation is intraspecific. The most prominent feature of this scheme is the, somewhat unconventional, treatment of the D.deflandzei - exilis - variabilis group as a single lineage. This interpretation is based in particular on my observations that: D.deflandrei and D.exilis intergrade continuously in assemblages from late NN4 to early NN6; and that D.variabilis and D.exilis intergrade during the NN9-10 interval. It is noteworthy that although the discoasters of this lineage dominate Miocene nannofloras they contribute little to conventional biostratigraphy. The lineage shows a general trend of increasing complexity, and consequent proximal / distal differentiation (Figure 53/B-D). Most other described species with bifurcate ray tips show closely similar features to the co-occurring main lineage species. Many of these are intergradational throughout their range with the main species and I have either not used them (e.g. D.aulakos) or recombined them as varieties (e.g. D.variabilis var. decorus, D.exilis var. petaliformis, Fig.53/G). Others appear to be more distinct and I have left them as discrete, but closely related species (e.g. D.surculus, D.bollii, Fig.53/A,E). The most important single evolutionary development seems to be the branching of D.brouweri from the main lineage, which occurs during NN9-10. This is a complex phase of evolution and many described species are associated with it (e.g. D.calcar is, D.bamatus), for the most part I have left these as separate species. Detailed study of this interval would be particularly interesting. During the Late Miocene and Pliocene D.brouweri constitutes a second major lineage with associated species (e.g. D.altus) and varieties (e.g. D.brouweri var. asymmetricus). Finally D.pentaradiatus, D.quinqueramus, and Catinaster have more cryptic origins and possibly constitute examples of the punctuated equilibrium style of evolution. They are consistently distinct from co-occurring species, and so of considerable biostratigraphic value.

211 16.3 SYSTEMATICS Family DISCOASTBRACEAB TAN 1927 Genus Discoaster TAN 1927 The species of the main lineage are dealt with first, in three stratigraphical groups, followed by: the other discoaster species; preservational forms; and Catinaster.

16.3.1 EARLY MIOCENE - D.deflandrei GROUP.

D.deflandrei, Bramlette & Riedel 1954 Plate 6/7-11,15,19,24; Figure 53/D

D.deflandru BRAMLETTE & RIEDEL 1954 p i . 59/6; MULLER 1974a p i . 7/3; PERCH-NIELSEN 1977 p i.15/18-20; etcetera.

D . a u l a k o s 6ARTNER 1967b p .2 , p i . 4 /4 -5 .

D . d i l a t u s Hay, in HAY et al 1967 p.450, p i.4/3-4.

D.divaricatus Hay, in HAY et al 1967 p.451, p i.3/7-9.

H . D e p h a d o s Hay, in HAY et al 1967 p.452, p i.2/4-5.

d.saundersii Hay, in HAY et al 1967 p.453, p i.3/2-6.

d.trhidadezsis Hay, in HAY et al 1967 p.453, p i.2/10-12.

D.»oo r e i BUKRY 1971 p.46, p i.2/11-12, 3/1-2; PRQTO-DECIMA et al 1978 p a r t i l p i . 7/5 dob p i . 7/6-8

(=D.prepentaradiatas); E LLIS & LOHMANN 1979 p i . 3/9. Description: D.deflandrei characteristically has a well developed central area, and rays ending in strong short wide bifurcations. There is often a distal knob, and sometimes weak distal ridges on the rays. The proximal side is essentially featureless, although the sutures may be incised. Remarks: The species varies in number of rays (5-7), size of central area, and degree of development of the ray bifurcations (PI.6/7-11). This lead Hay to create the large number of species listed above (all from the Early Miocene of Trinidad). These forms intergrade and separating them has not proven useful. A possible exception are forms with small central areas (D.aulakos, D.saundezsii). These become more common toward the end of the Early Miocene, and could be separated as an intermediate variety between D.deflandrei and D.exilis. In practice I found this was not useful. Range: Eocene to NN7. In the Middle Miocene D.exilis replaces it as the dominant form. D.deflandzei persists, however, at low abundances; the specimens of this interval show lower variation, with large centred forms dominant.

212 D.deflandrel var. druggii (Bramlette & Vilcoxon 1967) n.comb. D.exte»sas BRAMLETTE It UILCDXDN 1967 p.110, pl.B/2-8. D. draggii BRAMLETTE It WILCOXON 1967 p.220, noi. substit. pro D.extexsas; MARTINI 1971 pi.3/21; KNUTTEL 1986 pll/5-B, 2/1-2,5 , 3/1. E. draggii (Bra»lette It Wilcoxon) THE0D0RIDIS 19B3 p.17; THEODORIDIS 1984 p.157, pi.12/1, 32/6-7. Description: D.druggii is essentially a variety of D.deflandrei but is more consistently developed than most others. It is characterised by its large size, and also tapering rays ending in a terminal notch and lateral nodes, but no true bifurcations. A weak distal knob may be present but otherwise both sides are featureless. Range: NN2-NN3. In the absence of more reliable markers its first occurrence is used in sub-division of the Early Miocene.

D.kugleri Martini & Bramlette 1963 Plate 6/12,16,20, D. kugleri, MARTINI It BRAMLETTE 1963 p.B52, pi.104/1-3; BRAMLETTE It WILCOXON 1967 p.110, pi.7/1-2; MULLER 1974a pi.7/2; STEINMETZ & STRADNER 1984 pi.23/6-7. E. kugleri (Martini It Braalette) THEODORIDIS 19B3 p.17; THEODORIDIS 1984 p.170, pi.12/3-4,8, pi.35/15-18; MUZA et al 1987, pi.1/1-7. Description: D.kuglezi has a large central area which is flat on the distal side. The free rays are thickened by distal ridges, and have notched ends. In my material, as in the type area (Trinidad), the proximal surface is also flat, or only has very weak ridges. However, STEINMETZ & STRADNER (1984), THEODORIDIS (1984) and others have illustrated specimens from other areas with distinct proximal ridges. Remarks: This is a Middle Miocene species but it appears to evolve directly from D.deflandrei (not D.exilis), intermediate specimens occur. Range: NN7 (zonal marker).

16.3.2 MIDDLE MIOCENE - D.exilis GROUP. Remarks: This group dominates Middle and early Late Miocene discoaster assemblages. It merges into the D.deflandrei, D.variabilis, and D.brouweri groups; so understanding the characteristics of the group is important (see also Fig.53/C,E,G). They have low proximal ridges running along the rays. These unite at the centre, forming a stellate proximal knob, but wedge out before reaching the end of the rays. As a result the ray tips are thinner than the rest of the ray - which gives them a distinctive appearance, even in the light microscope. Also the discoasters are typically complanate, and have limited central areas. In all species most specimens have their rays terminated by acutely angled bifurcations, but otherwise identical specimens may have broadly

213 bifurcating or blunt ray ends. This can result in superficially very different looking morphotypes. This variation seems to be intraspecific and/or diagenetic, it is best ignored. Otherwise it is necessary either to create a very large number of varieties, or to extend the range and concept of typical Late Miocene species such as D.variaWlis and D.brouveri to cover these specimens. Both these approaches seriously confuse and obscure discoaster taxonomy.

D.bollii Martini & Bramlette 1963 Figure 53/E d.bollii MARTINI It BRAMLETTE 1963 p .B 5 1 , p i . 105/1-4,7} BRAMLETTE It WILC0X0N 1967 p .1 0 9 , p i . 8/11; MULLER 1974a p i.7/5-6. E .b o llii (Martini It Braalette) THE0D0RIDIS 1983 p .17; THE0D0RIDIS 1984 p.165, p i.33/8-11. Remarks: D.bollii has a large central area with knobs on both sides. The distal knob is considerably larger, than the proximal knob. Possibly better considered as a variety of D.exilis. Range: Characteristic of NN9, but does not have usefully sharp first or last occurrences.

D.exilis Martini & Bramlette 1963 D.exilis Martini & Bramlette 1963 var. exilis Plate 6/3-6,14,18,22-23, 7/6-7; Figure 53/C. D.challenger! BRAHLETTE k RIEDEL 1954 p .4 0 1 , p i . 39/10; MARTINI k BRAMLETTE 1963 p .8 5 1, p i . 103/11-12; HAY e t a l 1967 p i . 4/9-10; BRAMLETTE It WILC0X0N 1967 p . 109, p i . 8/1; H0JJATZADEH 1978 p i . 1/6. d.exilis MARTINI It BRAMLETTE 1963 p.852, p i.104/1-3; BRAMLETTE It HILC0X0N 1967 p .110, p i.7/3; HAY e t a l 1967 p i . 4 /7 -8 ; MULLER 1974a p i . 7/7; H0JJATZADEH 1978 p i . 1/10-11. D.extensas Hay, in HAY et al 1967 p.451, p i.3/10,12, 4/1-2. d.variabilis M a r t in i It B r a a le t t e , HAY et a l 1967 p i . 3/12; H0JJATZADEH 1978 p i . 2 /5-6. ti.fortosus MARTINI It HORSLEY 1971 p . 1500, p i . 2/1-8. d.signus BUKRY 1971a p.3, p i.3/3-4; C non E.sigms (Bukry) THE0D0RIDIS 19841. d.ptignosa H0JJATZADEH 1978 p.8, p i.2/9-10. D. gozoensis H0JJATZADEH 1978 p .1 0 , p i . 2/9-10. d.prisnatica H0JJATZADEH 1978 p .ll, p i.3/6-7. E. exilis ( M a r tin i k Braalette) THE0D0RIDIS 1983 p .17; THEODORIDIS 1984, p .163, p i.12/2, 33/3-5. Description: Typical form of D.exilis group (see discussion above). Distal and proximal knobs small. Remarks: This is the dominant Middle Miocene discoaster, and the long synonymy reflects various attempts at sub-dividing it. D.challengeri and D.exilis were both described from the Middle Miocene of Trinidad. The type specimens differ in central area size (virtually absent in D.challengeri) and bifurcation form, (wider in D.challengeri). These characters are, however, variable and I would include both type

214 specimens in the same species. D .challenger i has priority over D.exilis, but the holotype of the latter vas much better illustrated, and more typical of the species. In consequence D.exilis has become the established name for this species. Range: Evolves from D.deflandrei in NN5, and is replaced by D.variabilis in NN10. Both events are gradational and unsuitable as marker horizons.

D.exilis var. petal iformis (Moshkovitz & Ehrlich 1980) n.comb Figure 53/G

d.petal ifortis HOSHKOVITZ & EHRLICH 1980 p .1 7 , p i . 6 /1 -7 .

E . s i g n us (Bukry) Theodoridis, e«end. THE0D0RIDIS 1984 p. 163, p i.12/5-7, p i.33/14-17, p i.34/1-6. E.ef.E.signus (Bukry), THEODORIDIS 1984 p. 164, p i.34/7-14.

D. t u b e r i FILEWICZ 1985 p .3 5 7 , p i . 1/1-9. Description: Variety of D.exilis with large mushroom shaped knob on the distal side, and smaller simple knob on the proximal side. Central area normally small (specimens with larger central area illustrated by THEODORIDIS (1984), as E.cf.E.signus). Remarks: The forms illustrated by Filewicz, Theodoridis, and Moshkovitz & Ehrlich, are all the same age (NN4-5) and extremely similar. They are clearly the same taxon, and given their distinctive form and restricted range it is useful to distinguish them. D.signus is an inappropriate name for them, since the specimens described by BUKRY (1971a) lack central knobs. Apart from the knob they are normal D.exilis specimens and all authors noted that forms with large knobs graded into others with small knobs. The orientation of the distal knob appears aberrant, with its rays pointing in radial rather than inter-radial directions (Fig.53/G). However, it has normal sutural ridges and the peculiar orientation of knob may be associated with its flaring form.

D.exilis var. subsurculus (Gartner 1967) n.comb

d.subsurculus GARTNER 1967 p .3 , p i . 5/1-2.

E. subsurculus (G a rtn e r) THEODORIDIS 1983 p . 18; THEODORIDIS 1984 p . 165, p i . 33/12-13. Remarks: Variety of D.exilis with a protrusion or node between the arms of the bifurcation. It resembles D.pseudovariabilis and D.surculus, but the protrusion is in the plane of the rays, and it has a D.exilis type central area. It has not been widely recorded, and I found few distinct specimens in my material.

D.musicus Stradner 1959

l . M U s i c a s STRADNER 1959 p .1 0 8 8 ; S tra d n e r in STRADNER & PAPP 1961 p a r t i * , p.85, p i.17/7-10, 18/5,

»o» p i.17/4-5 (= U . k u g l e r i and C.coalites). D . s a n t ig u e le s s is BUKRY 1981 p . 462, p i . 2/7-10, 3/1-14.

215 E.msicus, (Stradner) THEODORID IS 1983 p.18; THEODORIDIS 1984 p.164, pi.34/15-18, pi.35/5. Remarks: D.musicus has, like D.bollli, a well developed central area and distal knob. However, it lacks a proximal knob; the distal knob is lower than in D.bollii; and sutural rays run from it to the edge of the central area. Possibly better considered as a variety of D.exilis. Range: Confined to NN5 in the Mediterranean (THEODORIDIS 1984) but ranges into NN6 in the East Pacific (BUKRY 1981), distribution probably patchy.

16.3.3 LATE MIOCENE - D.variabilis GROUP.

D.surculus Martini & Braralette 1963 Plate 7/16,20-21,24-25; Figure 53/A. D.surculas MARTINI It BRAMLETTE 1963 p.854, pi.104/10-12; MULLER 1974a pi.6/12; PERCH-NIELSEN 1977 pi.15/1,2,5,6,10; etcetera. D.pseadovariabilis Martini It Horsley 1971, MULLER 1974a pi.7/9-10; PR0T0-DECIMA et al 1978 pi.7/9, 8/2. Description: Distal surface similar to that of D.variabilis, except that it often has strongly developed distal ridges (e.g. Plate 7/21). The proximal ridges, however, do not bifurcate as in D.variabilis, but continue undivided, beyond the bifurcations of the distal surface. This gives the ray tips a distinctive trifurcate appearance. The tips of the proximal ridges are curved downward, as in D.brouweri. Remarks: D.surculus, like D.variabilis, varies considerably in size, and in ray tip morphology. It has not, however, been subdivided. The two species are closely related, they first appear at much the same time and disappear together at the top of NN17 (my observations plus data from other authors). Furthermore some specimens, particularly of the D.variabilis decorus type, are intermediate between the two main species. It is thus conceivable that they are not genuinely separate. Range: Common from eNNll to top NN17. Its last occurrence is used as a datum, the first occurrence is not since it is rather gradational, and because D.surculus and D.pseudovariabilis are hard to discriminate.

D.variabilis Martini & Bramlette 1963 D.variabilis MARTINI k BRAMLETTE 1963 p.104, figs 1-9; MULLER 1974a pi.8/11,12; PR0T0-DECIMA et al 1978 pi.7/10-14; etcetera. D.challengeri Braalette It Riedel, PERCH-NIELSEN 1977 pi.15/11,13; STEINMETZ It STRADNER 1984 pi.17/2; STRADNER 1973 pi.34/1-2. Description: D.variabilis differs from D.exilis primarily in having more strongly developed proximal ridges (Fig.53/B,C). These run to the ends

216 of the rays and merge into the bifurcations, as a result the ray ends and bifurcations are much thicker than those of D.exilis, and have a higher relief in the light microscope. In addition the rays usually taper less than those of D.exilis, and have widely separated bifurcations. On the distal side there is normally a distinct central, area, with scalloped depressions on the beginning of each ray, and often a central boss, mid-Miocene discoasters rarely have well developed distal areas of this type. Remarks: D.variabilis evolved from D.exilis during the early Late Miocene (NN9-10), but the first occurrence is gradational, and of no use for high resolution biostratigraphy. PRINS (1971) and THEODORIDIS (1984), derive D.variabilis directly from the D.deflandrei group, and record it throughout the Middle Miocene. In ray experience this is only possible if the D.variabilis species concept is extended to include specimens with weakly developed central areas and D.exilis type proximal ridges; in other words by dividing the D.exilis species group. This is reflected in the differences between the synonymy above and that given by THEODORIDIS (1984) for B.variabilis. D.variabilis varies considerably (hence the name) in form, in particular the bifurcations vary considerably in length, from less than the width of the ray to two or three times it, and can be asymmetric. Also the proximal ridges below the bifurcations vary considerably in degree and style of development (PI.7/13,17,21). Many of the variants have been described as species or subspecies, I believe they are best considered as varieties. Range: NN9 or 10, to NN16.

D.variabilis Martini & Bramlette 1963 var. variabilis Plate 6/1,13; Figure 53/B. Remarks: Typical form with moderate sized bifurcations.

D.variabilis var. decorus (Bukry 1971) n.comb Plate 6/2 d.variabilis decorus BUKRY 1971a p.48, pi.3/5-6.; PROTO-DECIMA et al 1978 pi.7/15-16, 9/10. d.decorus (Bukry) BUKRY 1973b p.677, pi.2/8-9, 4/11; STEINHETZ k STRADNER 1984 pi.22/3. d.exilis Martini k Brailette, STEINMETZ k STRADNER 1984 pi.17/1, 20/3. E.decom(Bukry), THEODORIDIS 1983 p. 17; THEODORIDIS 1984 p.l&l ,pi.32/11. Remarks: Characterised by small acutely angled bifurcations, and large size. Good specimens only found in the Pliocene, where they are widespread.

217 D.varlabilis var. loeblichii (Bukry 1971) n.comb. Plate 7/10-11 S.loeblichii BUKRY 1971c p.315, p i.4/3-5; PR0T0-DECIMA et al 1978 pl.9/7-8; STEINMETZ It STRADNER 1984 p i.19/9. E.loeblichii (Bukry) THE0D0RIDIS 1983 p.17; THEODORIDIS 1984 p.161. Remarks: D.loeblichii is characterised by asymmetric ray bifurcations. The sense of asymmetry is consistent on any one specimen. The morphology is otherwise conservative, with moderate central area and weakly developed distal ridges. Mainly recorded from NN10, I also found rare specimens throughout NN11.

D.variabilis var. pansus (Bukry & Percival 1971) n.comb. Plate 6/17,21. V.variabilis paisas BUKRY k PERCIVAL 1971 p.129, p i.3/8-9; PERCH-NIELSEN 1977 p i.15/17. D. 'variabilis' Martini k Brailette, PERCH-NIELSEN 1972, p i.4/1-4,6. d.icaras, STRADNER 1973 p.1138, fig .1, p i.41/7,10-11 p i.42/1-6,9,10. STEINMETZ k STRADNER 1984 p i.22/4,23/10. 0.paisas (Bukry k Percival) BUKRY 1973b p.678, p i.4/25; BUKRY 1976 p i.1/18. E. paisas (Bukry k Percival) THEODORIDIS 1983 p.18; THEODORIDIS 1984 p.161 p i.32/9-10. Remarks: Variety with large broad bifurcations. The proximal ridges are often strongly developed beneath and beyond the bifurcations, appearing as "flaps” or "webbing”. I did not find it practical to distinguish forms with this feature (icarus), from those without (pansus). Occurs occasionally throughout the range of D.variabilis.

D.variabilis var. pseudovariabilis (Martini & Vorsley 1971) n.comb. D. pseadovariabilis MARTINI k HORSLEY 1971 p.1500, p i.3/2-8; STEINHETZ k STRADNER 1984 p i.22/1-2; MARTINI 1981, p i.3/10-11. non MULLER 1974a pi.7/9-10 (= d.sarculus)-, non PR0T0-DECIMA et al 1978 p i.7/9, 8/2 (=D.surcaIas); . E. pseudovariabilis (Martini k Horsley) THEODORIDIS 1983 p.18; THEODORIDIS 1984 p.159, pi.32/12-13. Remarks: This form has bifurcations with well developed proximal ridges which extend prominently. In other respects it is intermediate between D.exilis and D.variabilis, having a poorly developed central area, narrow rays and weak bifurcations. It occurs sporadically in the NN9-10 interval, and is not a very useful variety.

218 16.3.4 OTHER NEOGENE DISCOASTERS

D.altos Muller 1974, emend. Plate 8/17-25 D.sp.aff. V . f o n o s a s Martini & Horsley, HEKEL 1973, pi.2/5-7. D.altus MULLER 1974a, p.592r p i.9/1-3. M a l t a s MARTINI 1976 p.420, p i.12/17-18. D.tristellifer BUKRY 1976 p.499, p i.1/1-17} BUKRY 19B1 p.462, pi.4/1-6; PUJOS 1985c p i.2/28-32. Catinaster altus (Muller) PERCH-NIELSEN 1984 p.42j PERCH-NIELSEN 1985b p.479. x Emended description: D.altus is^six rayed discoaster with blunt ended complanate rays. It has a very large stellate boss on the proximal side, and a small one on the distal side. Remarks: Discoasters with this morphology occur frequently in samples I examined from DSDP Site 242 Core 3. BUKRY (1976) and MULLER (1974a) described, respectively, D.tristellifer and D.altus from this material, (Section 242-3-1), without reference to each other's work. Bukry's description, based on light microscopy, is similar to that given above and is definitely applicable to the specimens I found. Muller's description of D.altus is rather different; she states that it has a very large proximal knob, but that the distal side is flat, and that the rays are short. Her illustrations are, however, of the species I found. The discrepancies are probably due to her using the SEM, the distal knob is then not visible, and the rays appear shorter relative to central knob size (see Plate 8). Plainly only one species is involved, and the name D.altus has priority. The emended description corrects the original one. Range: In the material I examined D.altus only occurs in the NN15 age sediments of Site 242. BUKRY (1976, 1981) has also recorded it from the Early Pliocene of the low latitude Pacific - Sites 83, and 317. The paucity of records combined with the very distinctive morphology suggest either a sporadic, or a very restricted occurrence. All the records are from NN15, low latitudes and the Indo-Pacific, so possibly its occurrence is restricted to this time-space volume. If it occurred in the Caribbean, Mediterranean or Atlantic it would almost certainly have been observed. Relations: D.brouveri variants with largish proximal knobs occur in the Pliocene, (sometimes distinguished as D.intercalaris). I suspect that D.altus evolved from these, but that it was a genuine species rather than a variety of D.brouweri.

D.bellus Bukry & Percival 1971 Plate 7/4 D.bellus BUKRY It PERCIVAL 1971 p.128, p i.3/1-2; BUKRY 1973b p i.4/6.

219 E.bellas (Bukry & Percival) THEODORIDIS 1983 p.17; THEODORIDIS 1984 p.174, p i.12/14, 37/1-3. Remarks: D.bellus is a name available for non-birefringent symmetrical pentaradial discoasters with simple ray ends, and small central areas. Such forms occur commonly in association with D.hamatus, but also persist at low abundances in NN10. Outside this interval rare specimens with this type of morphology occur, but are better assigned to other species.

D.braarudii Bukry 1971 D.braaredii BUKRY 1971a p.45, p i.2/10; ELLIS fc LOHHANN 1979 p i.3/1 D.broaueri Tan, HAY et al 1967 p i.5/1-4. Remarks: Bukry proposed this name for D.brouveri variants without bent ray tips (i.e. proximal blades). As he stated, such specimens occur throughout the range of D.brouweri; but they are seldom common and are mainly preservational artefacts. He also, however, included Middle Miocene forms in his synonymy, and species concept. I found such forms, with very weak bifurcations or blunt ray tips, to be rather common in NN9/10. They seem to represent an intermediate between D.exilis and D.brouveri. Consistently differentiating them was impracticable when counting, and instead I included them in a broad D.exilis species concept.

D.brouveri Tan 1927 emend Bramlette & Riedel 1954 D.broaueri TAN 1927 p.415, fig.2/8. D.broaueri Tan e«end BRAHLETTE & RIEDEL 1954 p.402, fig.3, p i.39/12; MULLER 1974a p i.6/1-3; PERCH-NIELSEN 1977 p i.15/9; STEINMETZ & STRADNER 1984 p i.12/1-4; etcetera. D. broaueri ratellas GARTNER 1967b p.2, p i.1/1-2. E. broaueri (Tan) THEODORIDIS 1983 p.17; THEODORIDIS 1984 p.178, p i.36/11-12. Description: D.brouveri has non-bifurcate rays the ends of which are extended by proximal blades giving it a distinctive concavo-convex form. The distal surface is weakly developed; the central area is usually small, and often smooth although weak depressions and/or a low knob may be present. Distal ridges are not developed or, of course, bifurcations. The proximal side by contrast is well developed, a stellate knob is normally present, much of the rays can be regarded as proximal ridges, and the ray tip blades are proximal features. Bramlette & Riedel's drawing (their fig.3b) is rather misleading in showing the rays as bent in a continuous curve. Usually the rays are only slightly deflected from horizontal, but have prominent blades built down from their tips (Plate 7/26). These blades may extend beyond the end of the rays proper.

220 Remarks: The species as described above is very common throughout the Late Miocene and Pliocene, until its extinction, at the end of NN18. Like D.vaziabilis it evolved from D.exilis during the NN9-NN10 interval. This process involved loss of bifurcations, and development of the proximal blades. THEODORIDIS (1984) recorded forms with rotated distal knobs as being dominant in the Miocene, but sporadic in the Pliocene, he named these B.brouveri subsp. stzeptus. This may be valid, but I found separating the form impracticable and time-consuming. Associated species: D.altus (=D.tristellifer) is a related Pliocene species. D.bzaazudii, D.calcazis, D.giganteus, and D.neozectus are forms developed during the evolution of D.brouveri from D.exilis. D.b.asymmetzicus, D.b.tamalis and D.b.triradiatus are varieties defined on ray number.

D.brouveri Tan 1927 var. brouveri Plate 7/9,11,26. Description: Typical six rayed variety of D.brouveri.

D.brouveri var. asymmetzicus (Gartner 1969) n.comb. Plate 7/12.

D.asynetrictts GARTNER 1969 p.587, p i.1/1-3; PERCH-NIELSEN 1977 p i.15/3; PROTO-DECIHA et al 1978 p i.6/2, 9/2; STEINMETZ & STRADNER 1984 15/1-6. Description: Variety of D.brouveri with five asymmetrically disposed rays. The rays may also vary in length. Remarks: Most six rayed species produce asymmetric five rayed specimens, and D.b.asymmetzicus can be found throughout the range of D.brouveri. However, they are notably more abundant, and more consistently present, in mid Pliocene sediments. It seems to be an intraspecific variety with a restricted stratigraphical acme. Both the beginning and the end of the acme can be used for biostratigraphy (BUKRY 1973a, BACKMAN 1986).

D.brouveri var. tamalls (Kamptner 1967) n.comb. Plate 7/13 d.taaalis KAMPTNER 1967 p.166 fig .29; STEINMETZ & STRADNER 1984 p i.13/1-4; etcetera. Description: Variety of D.brouveri with four, symmetrically disposed rays. Remarks: Unlike D.asymmetzicus, this form is very rare outside the mid Pliocene. It seems to have genuine first and last occurrences at, or very near to, the levels of the beginning and end of the D.asymmetzicus acme. Within this range its abundance is closely related to that of D.asymmetzicus (BERGEN 1984, BACKMAN 1986, my observations). The two

221 varieties are also of similar size in any one sample.

D.brouveri var. triradiatus (Tan 1927) n.comb Plate 7/13

d.triradiatas TAN 1927 p . 417, unfigured. d.broaueri Tan, HULLER 1974a p i.8/4; SAHTLEBEN 1979 p i.5/4; etcetera. b.brouteri trithallas TAKAYAMA 1973, p.54. Description: Three rayed variety D.brouweri. Remarks: Tan's description was extremely cursory but the concept is simple. These specimens occur throughout the range of D.brouveri, somewhat more consistently than D.asymmetricus. They do not increase in abundance in the mid-Pliocene, but do at the end of the Pliocene, just prior to the extinction of D.brouweri (BACKMAN et al 1983, BACKMAN 1986).

D.calcaris Gartner 1967 Plate 7/19, 23. d.calcaris GARTNER 1967b p.2, p i.2/1-3; MULLER 1974a pl.7/8; JAFAR 1975a p.47, p i.5/13-14; PROTQ-DECIHA et al 1978 p i.6/4-5; ELLIS l LOHHANN 1979 p i.3/3. 9.ieohatatas BUKRY 1 BRAMLETTE 1969 p.133, p i.1/4-5; JAFAR 1975a p.50, p i.6/1-2; ELLIS & LOHMANN 1979 p i.3/10; THEOD0RIDIS 1984 p.181; BERGEN 1984 p i.2/1. d.batatas Martini & Braalette, BLACK 1968 p i.153/7; PROTO-DECIMA et al 1978 p i.6/6; BERGEN 19B4 p i.1/3. E.calcaris (Gartner) THE0D0RIDIS 1983 p.17; THE0D0RIDIS 1984 p.171, p i.12/12, 35/13-14. Description: D.calcar is is a 6-rayed species with asymmetric ray tips. There are usually weak bifurcations on the distal side, one of which is underlain by a proximal blade. The central area is small. A proximal knob is usually present. In many specimens the asymmetry of the ray tips is increased by the blade sloping away from the ray. Diagenetic overgrowth occurs preferentially on the blade, further increasing the asymmetry. The overgrown forms are sometimes distinguished as D.neohamatus. Remarks: D.calcaris occurs in the NN9 to eNNll interval and seems to be associated with the evolution of D.brouweri from D.exilis, it can be hard to differentiate it from either of these species. D.hamatus is a 5-rayed species with the same structure, but it has a shorter and better defined range, hence differentiation of the two is important.

D.giganteus (Theodoridis 1984) n.comb. E.gigantens Theodoridis 1984 p. 172, p i.36/4-10. Remarks: This species was described as being very large with downturned ray tips like those of D.brouweri, but with thick bluntly truncated ray ends. It is one of the complex of species associated with the

222 development of the D.brouweri lineage in NN9-10.

D.hamatus Martini & Bramlette 1963 Plate 7/1-3, 17,27 d.haiatas MARTINI k BRAMLETTE 1963 p.852, p i.105/8,10-11; BRAMLETTE & WILC0X0N 1967, p i.7/9-11; PERCH-NIELSEN 1972 ?partit p i.9/3, «oi pi. 11/2,4,6 M.pettaradiatas and D.qaitqaeratas); etcetera. E.batatas (Martini k Braelette) THE0D0RIDIS 1983; THEODORIDIS 1984 p.174, p i.12/13, 37/4-7. Description: D.hamatus has five (or less) symmetrically disposed rays with asymmetric proximal blades on their tips. The central area is small with a featureless distal surface and small proximal knob. Remarks: This species seems to have an identical structure to that of D.calcaris (q.v.). There are, however, few electron micrographs of it. Most of the specimens illustrated by PERCH-NIELSEN (1972, pi.11) do not resemble typical D.hamatus, and they came from a DSDP sample (111A-6-3-77) of uncertain age. I found that only large specimens Oc.lO microns) had well developed offset blades, associated smaller specimens lacked clear blades and had to be placed in D.bellus for biostratigraphic rigour. Range: Confined to Zone NN9, which it defines.

D.neorectus Bukry 1971 D. teorectus BUKRY 1971c p.316, p i.4/6-7; PROTO-DECIHA et al 1978 p i.9/9. E. neorectus (Bukry) THEODORIDIS 1983 p.18; THEODORIDIS 1984 p.179, p i.36/14,17. Remarks Bukry described this species for large (>20microns) D.brouweri variants without ray tip bending. He recorded it as restricted to his D.neorectus subzone, (= the upper part of NN10). Other authors have recorded it in this interval, so it seems to be a useful form. I did not observe it in my material.

D.pentaradiatus TAN 1927 Plate 8/12-15,26; Figure 53/G. D.pentaradiatus Tan, MARTINI k BRAMLETTE 1963 p i.105/5; MULLER 1974a p i.7/11-12; PERCH-NIELSEN 1977 p i.15/7-8; PROTO-DECIMA et al 1978 pi.8/4; CEPEK k HIND 1979 p i.2/3, 7/7; etcetera. d.tridetas KAMPTNER 1967, p.166, fig .30; BUKRY 1981 p.462, p i.3/15-17. D.stradneri CATI k 80RSETTI 1970, p.629, pi.81/2-4. D. anconitatus CATI k B0RSETTI 1972, p.373 (no*, substit. pro. D.stradneri Cati It Borsetti). E. tisconceptas THEODORIDIS 1984 p.168, fig.A-D, p i.37/19-20. D.niscotceptas (Theodoridis) MUZA et al 1987, p.614. Description: D.pentaradiatus is very distinctive; it is symmetrically pentaradial (with rare 3 and 4 rayed variants) and birefringent. The rays are inclined, and have acutely angle bifurcations. The central area is small, but distinct. It often has scalloped depressions and/or

223 sutural ridges on the distal side, also a stellate proximal knob. Optical orientation: Attached rays show straight, length fast, extinction. Detached rays tend to fall on one side and show straight, but length slow, extinction. These observations indicate that the optic axis is perpendicular to the ray length, as usual in discoasters, and so that the birefringence is a product of the ray inclination. The degree of birefringence is unique, and is a reliable means of identifying the species in etched samples - where ray ends are rarely preserved. Remarks: The species is highly variable in size, and in degree of development of the bifurcations. Specimens lacking ray ends have been described as separate species by KAMPTNER (1967), and CATI & BORSETTI (1970, 1972). THEODORIDIS (1984), demonstrated conclusively that use of the name D.pentazadiatus for this species was based on a misconception of Tan's type specimen, and of Bramlette & Riedel's emended diagnosis. Nonetheless I have retained the conventional nomenclature, for the sake of simplicity, and because D.tridenus and D.anconitanus have priority over D.misconceptus. Range: Mid-NNIO to NN17, often very abundant.

D.prepentaradiatus Bukry & Percival 1971 Plate 7/5 D. pentaradiatus Tan e»end BRAMLETTE It RIEDEL 1954 p.401, fig .2, p i.39/11. Q.prepentaradiatus BUKRY & PERCIVAL 1971 p.129, p i.3/6-7. d.toorei Bukry, PROTO-DECIMA et al 1978 partii p i.7/6-8 ooo p i.7/5 (=5. exilis). E. peataradiatus (Tan) THEODORIDIS 1983 p.18; THEODORIDIS 1984 p.166, p i.12/10-11, 35/1-4. Description: Complanate non-birefringent pentaradial discoasters with obtusely angled bifurcations. The central area is not large but is distinct, there are low knobs on both the distal and proximal side. The rays are symmetrically disposed. Remarks: This species is only superficially similar to D.pentaradiatus: It is complanate (and so does not show any birefringence); has broad bifurcations; and is generally more robust. I did not observe any intermediates or ambiguous specimens. I do not believe the two are closely related, even though the last occurrence of D.prepentaradiatus is close to the first occurrence of D.pentaradiatus. My suspicion is that D.prepentaradiatus evolved from D.bollii; whilst D.pentaradiatus was independently derived from D.exilis. Size: c.lOmicrons. Range: NN9-eNN10

224 D.quinqueramus Gartner 1969 Plate 8/16,27-29; Figure 53/F D. quinquerantis GARTNER 1969 p.598, p i.4/6-7; PERCH-NIELSEN 1972 p i.9/1-2, pi. 10/1-4; MULLER 1974a p.592, pi.8/5-10; PERCH-NIELSEN 1977 p i.15/12,15; etcetera. l.quintatus BUKRY t BRAMLETTE 1969 p.133, p i.1/6-8. E. qainqueranas (Gartner) THE0D0RIDIS 1983 p.18; THEQDORIDIS 1984 p.175, p i.12/9, 37/8-18. Description: Discoaster with five curved, symmetrically disposed rays, and prominent central area structures. The proximal knob is large witK slightly asymmetric structure; weak proximal ridges radiate from it. The distal knob is much weaker, or absent, but there are prominent sutural ridges (Fig.53/F). The rays have rounded ends, and often low distal ridges (eg MULLER 1974, PERCH-NIELSEN 1972). They show distinct proximal curvature and inclination, particularly toward their tips, and the discoaster is often weakly birefringent. Remarks: This species is distinct throughout its range and its affinities are not clear. Range: Confined to NN11 (zone defining species). Size: Highly variable, c.5-15microns.

D.quinqueramus Gartner 1969 var. quinqueramus Description: Typical variety of D.quinqueramus, with central area width less than free ray length.

D.quinqueramus var. berggrenii (BUKRY 1971) n.comb. i.berggrenii BUKRY 1971a p.45, p i.2/4-6; PERCH-NIELSEN 1972 p i.10/5-7; PROTO-DECIHA et al 1978 p i.7/1-3; STEINMETZ l STRADNER p i.18/4; BERGEN 1984 p i.1/7-8. d.sp.cf.berggrenii BERGEN 1984 p i.1/6. Description: Variety of D .guinqueramus with central area width greater than ray length. Remarks: As observed by MULLER (1974a) and THEODORIDIS (1984) D.q.berggrenii grades into D.g.guingueramus, and is not a separate species. It is, however, only common in the lower part of the range of D.guingueramus; and discrimination of it is easy. As an arbitrary criterion I found restricting the name to specimens with central area width equal to or greater than the free ray length useful. I found typical specimens of D.guingueramuswere more abundant than specimens of D.q.berggrenii even in basal NN11 samples; hence I do not think that D.q.berggrenii is an ancestral form, as suggested by BUKRY (1971a).

225 16.3,5 Dlscoastez aster, AND OTHER PRESERVATIONAL SPECIES. Described species V a s t e r 8RAMLETTE * RIEDEL 1954 p.400, p i.39/7. H.HOQdringi BRAMLETTE k RIEDEL 1954 p.400f pi.39/8. V.argutas Hay, in HAY et al 1967 p.450 ,pi.5/8. d.laatus Hay, in HAY et al 1967 p.451, p i.5/7. b.adatanteas BRAHLETTE k UILC0X0N 1967 p.108, p i.7/6; HAS 1971c p.118, p i.24/2, 25/8-11. MULLER 1974a pi.9/4; HAMILTON k HOJJATZADEH 1982 p i.6.3/16-18. d.abtasus GARTNER 1967 p.2, pi. 3/1-6. D.stellalas GARTNER 1967 p.3, p i.4/1-3; HAMILTON k HOJJATZADEH 1982, p i.6.3/13-15. D.rufus ROTH 1970 p.867, p i.12/3; HAQ 1971c p.119, p i.22/5-6, 23/1-2, 25/6-7. D.fortosas MARTINI k HORSLEY 1971 p.1500, p i.1-8. D.chatbrayeasis HOJJATZADEH 1978 p i1, pi.3/8-9. D.zaivitiaevpeii HOJJATZADEH 1978 plO, pi.3/4-5. D.intercalaris BUKRY 1971c p.315, p i.3/12, 4/1-2; STEINMETZ k STRADNER 1984 p i.14/3-4. D.toralas ELLIS et al 1972 p.53, pi.1&/2-6; STEINMETZ k STRADNER 1984, p i.18/1-2. V.capricorDensis RADE 1977 p.279, p i.3/28. D. gillii RADE 1977 p.279, p i.3/23. E. htercalaris (BUKRY 1971) THEODORIDIS 1984 p. 179, p i.36/15-16. Discussion: Throughout the Neogene discoasters can be found with blunt ray terminations, but otherwise similar morphology to co-occurring specimens with bifurcate ends (and lacking the characteristic rays of D.brouweri). Such specimens are probably produced both by diagerietic alteration and by original intraspecific variation, related to variable development. There are at least four different ways of dealing with such specimens, they can be; A. Assigned to the dominant discoaster species of the assemblage, this is normally the best policy, particularly if diagenesis has not destroyed the central area features. B. Placed in a "dustbin" category such as Discoaster sp. or M.U.D. (miscellaneous unidentified discoasters). If a specific name was wanted D.aster would be roost appropriate since it has priority, and is suitably vague sounding. C. Placed in D.brouweri, this is tempting for mid Miocene specimens with longish rays but should be avoided, since it devalues the D.brouweri species concept. D. Assigned to species of their own. As the long "synonymy" above shows this has been a popular approach, fortunately most of these species have never been cited since their original descriptions. Important exceptions are D.adamanteus and D.intercalaris. D.adamanteus is applied to thick specimens with short blunt rays. This morphotype develops particularly readily from D.deflandrei and so is most frequently found in Early Miocene sediments. It can, however, develop

226 from virtually any six rayed discoaster, and so is stratigraphically unreliable. D.intercalaris is used for Late Miocene and Pliocene forms with short blunt arms and the distinctive D.variabilis type of central area, D.tozalis is identical. I found examples of such specimens sporadically throughout the range of D.variabilis.

16.3.6 CATINASTER Genus Catinaster Martini & Bramlette 1963 Type species: C.calyculus Martini & Bramlette 1963 Remarks: Catinasters are rather curious sub-cylindrical or cup-shaped nannofossils. They have six-fold radial symmetry and the same crystallographic orientation as discoasters. There are two main species, C.coalitus which has septae-like radial structures confined to the cup, and C.calyculus in which radial structures extend beyond the cup. Although all authors agree that they are a modified form of discoaster, there are two separate interpretations of their morphology. MARTINI & WORSLEY (1971, p .1477) suggested that they evolved from a small discoaster "by becoming highly concavo-convex and reducing its interray areas until they are closed". This process is illustrated schematically in Figure 55/a-b-c (p.232). THEODORIDIS (1984, p,17 3), however, described catinasters as "asternliths ___ with very pronounced sutural ridges on the distal face of the central area" . According to Martini & Worsley’s interpretation these ridges are reduced rays so the two points of view are incompatible (compare Figs.55/b and 55/d). Theodoridis*s interpretation was based on the striking similarity of the distal face of C.coalitus to the distal central area of D.musicus [compare STRADNER & PAPP (1961) pi.17/4 & pi.17/7,8; or THEODORIDIS (1984) pi.35/5 & 35/7). On this basis C.coalitus would have evolved from a discoaster such as D.musicus simply by loss of the rays, as shown in Figure 55/a-d-e. I favour the interpretation of MARTINI & WORSLEY (1971) for the following reasons. A. C.coalitus specimens with breaks in the rim on the distal side are common, (Fig.l/b, cf. MARTINI (1981) pi.3/1). These appear to be residual gaps between bifurcations. On the sutural ridge theory they should not occur. B. A proximal view of C.coalitus given by MULLER (1974a, pi.10/4) shows a stellate central structure (my Fig.55/1). Comparison of this figure with others (MULLER 1974a, pi.10/3,5) indicates that this structure is orientated parallel to the ridges on the opposite face, this in turn suggests that the ridges on the distal face are rays. C. In C.calyculus the ridges are extended as free rays (Fig.55/f-i).

227 D. Some SEM views show apparent weak inter-ray sutures in the centre of C.coalitus (eg MARTINI (1981) pi.3/1), shown in Figure 55/b,c. All these points favour interpretation of the distal features as rays rather than as sutural ridges, and so support the evolutionary model of Martini & Worsley.

C.calyculus Martini & Bramlette 1963 Plate 8/74-9; Figure 55/h-i,m. C.calyculus MARTINI i BRAMLETTE 1963 p.850, p i.103/1-6; BRAMLETTE & WILC0X0N 1967 pl.B/13; MULLER 1974a p i.10/9-12; JAFAR 1975a p i.6/11-12; MARTINI 1981 pi.3/5-9, 5/3-6; PERCH-NIELSEN 1985 fig.33/7. E.calyculus (Martini it Braalette) THE0D0RIDIS 1984 p.174, p i.35/8. Description: Catinasters with curved free rays extending beyond the main body. Remarks: The rays lengthen through the range of the species (MARTINI 1981). The specimens in Plate 8 have short rays, and so are intermediate between C.calyculus and C.coalitus. Range: NN8 - NN9, ?NN10.

C.coalitus Martini 6 Bramlette 1963 Plate 8/1-3,10-11, Figs. 55/a-b,f-g,l C.coalitns MARTINI it BRAMLETTE 1963 p. 851, p i.103/7-10; BRAMLETTE it WILC0XQN 1967 p i.8/9-10; MULLER 1974a p i.10/1-5; MARTINI 1981 pi.3/1-4; STEINHETZ & STRADNER 1985 pi.23/4-5. E. coalitus (Martini & Brailette) THE0D0RIDIS 1984 p.173, p i.12/15, 35/6-7,9. Description: Catinaster with rays which bifurcate to form a distal rim, but which do not extend significantly beyond this rim. The rim continues to the proximal pole producing a basket like body. Range: NN8-9, this is the first Catinaster species to appear and is the most common.

? Catinaster mexicanus Bukry 1971 Figure 55/j C.iexieaaus BUKRY 1971a p.50, p i.3/7-9; MULLER 1974a p.591, p i.10/6-8; BUKRY 1981 p i.1/1-3; MARTINI 1981 p.557. ?E.ie xicam (Bukry) THE0D0RIDIS 1984 p.173, pi.35/10-12 [possibly C.coalitus]. Remarks: This species resembles C.coalitus in gross morphology, but appears to have stubby bifurcate rays with sutural ridges on the distal side. The structure is apparently different to that of C.coalitus and C.calyculus (see{ Fig.55). Hence I suspect that this species is either an unrelated homoeomorph, which did evolve by ray reduction, or is a preservational fragment of a discoaster, rather than a genuine species. On either basis the form is probably not referable to Catinaster. Range: Scattered records only, ?NN9 - ?eNNll. Also in NN15 age

228 sediments from site 241, MULLER (1974a) recorded, and illustrated, abundant specimens from 241-7-CC. I did not resample this material, but I did find rare specimens in 241-7-5. Since there are no other indications of middle Miocene material in this sample it appears to be a genuine occurrence, not a result of reworking. PUJOS (1985b) also recorded mid Pliocene C.mexicanus specimens, and suggested that they might be relicts of D.tristellifer (=D.altus).

\

229 FIGURE 51 (top) - MORPHOLOGICAL ELEMENTS OF DISCOASTERS. D.sarcalas - distal view (left), proxiial view (right), ray section (centre), and side view (bottoi).

FIGURE 52 (bottom) - CRYSTALLOGRAPHIC ORIENTATION OF DISCOASTER RAYS. Diagraiiatic representation of the crystallography of discoasters, as elucidated by BLACK (1972), frofl a consideration of crystal fori. The double Y syibols represent the crystallographic axes of calcite (Holosyiietric Trigonal crystal class) - in projection above (solid lines) and below (dotted lines) the paper. Note that: (1) The rays are separate crystals, in radially syuetric orientations. The trigonal synetry leans that each of the rays of 3-rayed foris, and every other ray of 6-rayed fo n s are in crystallographic continuity. This is best shown by diagenetic overgrowths. (2) The upper and lower surfaces of the discoasters are not identical. This is best shown by central structures, (knobs, sutural ridges and depressions). Based on various sources.

2 2 9 A FIGURE 53 - MORPHOLOGY OF VARIOUS DISCOASTER SPECIES. Distal (left), proximal (right), and side (bottom) views, also ray section in A-D (centre). Diagrams based on synthesis of data rather than individual specimens, side views schematic. A. D.surculus; B. D.variabilis; C. D.exilis var. exilis; D. D.deflandrei; E. D.bollii; F. D.guingueramus; G. D.exilis var. petaliformis, flaring distal knob shows atypical orientation, dotted lines indicate sutural ridges below knob; H. D.pentaradiatus.

230 FIGURE 54 - STRATIGRAPHIC DISTRIBUTION OF DISCOASTERACEAE. Based on synthesis of ay own and published data;. Arranged to show likely phylogenetic relationships in the aain groups. Coluan thickness related to iaportance in ay Material, sporadic occurrence/uncertain range indicated by broken lines. Various species oaaitted in the NN9-10 interval.A Events used in published zonation scheaes, saall arrows less reliable. N.B. Vertical scale changes at NN5/6 boundary.

ABBREVIATIONS: aspa - asjaietricus; braa - braaradi; brg - berggrevii; calc - calcaris; calf - calfculus. coal - coalitas; dec - decoras} tasic - aasicas; petal - petal if onis; p'peit - prepeataradiates; taa - taaalis; tri - triradiatas. 231 232

FIGURE 55 - CATINASTER ORIGINS, DEVELOPMENT, AND MORPHOLOGY, a-t. - ORIGINS: Distal views of C.coalitas (sitple forts b,d; tore cotplex, later, forts c,e), and a D.exiii; group discoaster. Sequence a-b-c - itplied hotologies of evolutionary sequence proposed by MARTINI & HORSLEY (1971). Sequence a-d-e - alternative hotologies suggested by THEODORIDIS (1984). Morphological details such as the rit breaks and central area sutures shown in figures b and c suggest that sequence a-b-c is tost likely. f-iDEVELOPMENTAL SEQUENCEI Distal views illustrating evolution frot Catinaster coalitas to C.calycalas. f, Early C.coalitfl**, g,late C.coalitus} h, early C.caljculus} i, late C.caiycflla*. j-t. COMPARATIVE H0RPH0L06Y: Distal (top row), cross-section (centre row), and proxital (bottot row) views of: j, C? Mexican as; k, D.exilis; 1, C.coalitas; t, C.calycalas. Note: (1) Cup shaped fort of C.coalitas and C.ca/ycaia; produced by deepening and inclination of the bifurcations. (2) Different structure of C? nexicanas which does appear to have distal sutural ridges. CHAPTER 17 - CERATOLITHACEAE AND TRIQUETORHABDULACEAE.

Introduction The Ceratolithaceae, Triquetorhabdulaceae, and a couple of other genera are considered together here, since they have similar optical properties, and ultrastructure; and since they form a continuous succession of reliable, if somewhat rare, biostratigraphical marker species (Fig.58, p.247). Fossil ceratoliths have been extensively discussed and illustrated (KAMPTNER 1954, GARTNER & BUKRY 1975, PERCH-NIELSEN 1977, 1985b, BERGEN 1984), and some observations made on living forms (NORRIS 1965, 1971, BORSETTI & CATI 1976). For the Triquetorhabdulaceae the original descriptions are the main sources of information, with some additional discussion in LIPPS (1969), BIOLZI et al (1981) and PERCH-NIELSEN (1985b).

17.1 STRUCTURE 17.1.1. Triquetorhabdulaceae (Figure 56, p.245, & Plate 9). The Triquetorhabdulaceae are elongate nannoliths formed of three laths arranged back-to-back, in optical continuity. The two most common species are T.carinatus (Oligocene and Early Miocene) and T.rugosus (Middle and Late Miocene). In T.carinatus the three laths are similar and are arranged at 120° to each other; the nannolith is thus radially symmetrical in cross-section. The optic axis of the nannolith is parallel to the length, consequently they show strong birefringence and straight, length-fast, extinction, in any common orientation. The laths are normally featureless, very well preserved specimens, however, sometimes show a fine transverse striation, resembling the rodded ultrastructure of T.rugosus (eg PERCH-NIELSEN 1977 pi.36/6). T.rugosus is formed of three dissimilar laths. The two basal laths are broad and meet at a shallow angle. The third, median, lath is low, and curved in plan view (Fig.56/F). All three laths often show a rodded ultrastructure, perpendicular to the length. This is always clearest on the broader of the basal laths. In addition subsidiary ridges sometimes form on the basal laths. The optic axis of the laths is perpendicular to their length and in the plane of the median ridge; it is therefore vertical when the basal plates are horizontal, and the median ridge vertical. This is the usual rest position of the nannolith, which is consequently dark in cross-polars. An alternative, only slightly less common, position is for the median ridge to be pointing downward, the basal plate is then tilted from the horizontal; in this orientation the

233 nannolith shows weak birefringence, with straight, but length-slow extinction. Thus these two species have distinctly different structures and optical orientations: T.carinatus has three identical laths with the optic axis parallel to their length; T.rugosus has three dissimilar laths with the optic axis perpendicular to their length. Of the minor species T.auritus, T.challengeri, and T.milovii all show the same structure as T.carinatus and were plainly derived from it. T.milowii is similar in shape to T.rugosus, and their ranges overlap; but T.milowii always has the T.carinatus type optical orientation. Orthorhabdus serratus superficially appears to be a third distinct type since it shows birefringence, but with the opposite orientation to T.carinatus (i.e. length-slow). However, this species can be considered a variant on the T.rugosus pattern, with a larger median lath (compare Figs.56/E & F). Since the median lath is larger than the basal laths the nannolith usually rests on its side, and so shows strong birefringence. The two species are readily separable on other grounds: 0.serratus has strongly differentiated ends; is bilaterally symmetrical; and does not obviously have the rodded ultrastructure of T.rugosus. Nonetheless their similarity is striking, particularly when compared to T.carinatus. This suggests that T.rugosus is more likely to be related to 0.serratus than to T.carinatus. For convenience of reference I have retained the conventional combinations, but it would be sensible to adjust the taxonomy to reflect the structure. This could be done in one of three ways: (1) 0.serratus could be placed in Triquetorhabdulus; (2) A new genus could be created for T.rugosus; (3) T.rugosus could be transferred to Orthorhabdus. I favour the last option.

17.1.2 Ceratolithaceae (Figure 57, p.246) Ceratoliths are horse-shoe shaped nannoliths. They have been observed on living cells by NORRIS (1965) who found that the ceratolith enveloped the protoplast, with only one ceratolith occurring per cell. In some cases the protoplast and ceratolith were observed surrounded by an outer sphere of thin hoop shaped "coccoliths" (Fig.57/A). The structure of these coccoliths was further described by NORRIS (1971), but it is not clear if they are calcified, or not. The mode of occurrence of ceratoliths and their large size suggest that they may form at least in part outside the protoplast, but within the extended wall zone of the coccosphere. Ceratoliths behave as a single unit optically, and unlike discoasters, develop sutureless overgrowths. So they probably are genuinely formed from single calcite crystals, nonetheless they have complex structures; and some conventions are needed to describe them.

234 Since the ceratolith envelopes the cell there is no true biological way-up, and the terms proximal and distal are inappropriate, instead an arbitrary orientation needs to be adopted. I have termed the more heavily ornamented side the upper surface, and the closed end the front, left and right arms can then be distinguished (Fig.57/B). In C.cristatus the right arm is usually shorter than the left arm, and more strongly curved, forming most of the arch. This is particularly clear on the lower side. Both arms have keels on both surfaces. Those on the lower surface are more or less smooth, whereas those on the upper surface have a rodded structure. This rodding is most strongly developed on the right arm, and the rods are more nearly vertical. The asymmetry of structure makes the two surfaces quite easy to distinguish, in the light or electron microscope. The optic axis of the crystal which forms the ceratolith is horizontal, and perpendicular to the arms, consequently they show high birefringence and straight, length-slow, extinction. The various described species of Ceratolithus, except C.atlanticus, all show the same structure, variation affects the degree of development of the different structures rather than their nature. They all show the same optical orientation. The earlier ceratoliths, of the genus Amaurolithus, by contrast have their optic axes vertical, and so show little or no birefringence in plan view. They also have different structure, only the right arm has a rodded keel, and this is. usually less strongly developed than in Ceratolithus. Also it is the left rather than the right arm which is strongly curved, and apparently continuous with the arch. The upper keel on the left arm continues round the arch, and in some cases is expanded to form an apical flange.

17.2 POSSIBLE PHYLOGENETIC RELATIONSHIPS AND BIOSTRATIGRAPHIC USE. (Figure 58, p.247). As discussed above four distinct groups can be recognised within these two families, T.carinatus, T.rugosus, Amaurolithus and Ceratolithus, Each of these has a characteristic structure, and optic axis orientation, and within each group there is little problem in discerning likely phylogenetic relationships. However, development of any of the groups from any of the others would require radical structural change, and crystallographic reorientation. Nonetheless, they constitute a rather coherent group: The entire group forms a temporally continuous series (Fig.58). All the nannoliths are constructed of single calcite units, and examples of the distinctive rodded ultrastructure are seen in each group. Each sub-group contains some elongated varieties which it is difficult to

235 envisage forming intracellularly. Finally, certain species show such striking homoeoiiorphy that it is difficult to believe they are unrelated, for instance: T.challengeri and T.r.striatus; A.t.delicatus and C.cristatus; T.rugosus and C.cristatus arms. A possible explanation is that the entire group is a natural taxon, and that in all cases calcification occurred in a partly extracellular manner as in C.cristatus. In this mode reorientation of crystallography may have occurred more readily than in normal coccolithogenesis. Virtually all the species of the group have reliable, well determined, ranges and their biostratigraphic use is only limited by the fact that they are often very rare. There is possibly further potential from subdivision of the T.rugosus lineage, which seems to show a development from simple parallel sided forms in the Middle Miocene to more elaborate asymmetric forms in the Late Miocene.

17.3 SYSTEMATICS 17.3.1 Family Ceratolithaceae NORRIS 1965.

Genus Amauzolithus Gartner & Bukry 1975 Type species A .tricozniculatus Gartner 1967. Remarks: Half-a-dozen species of Amauzolithus have been described, with sharply differing type specimens. In practice many specimens can be assigned to these species, indeed the resemblance to the holotypes is often striking, and in these cases the age of the specimens is always close to that of the typical range of the species. However, there are also, as BUKRY & BRAMLETTE (1968) noted "gradations between these varieties". I suspect that their original concept of a single species with distinctive subspecies might be the most biologically valid way of approaching this group. This, however, would have the result that the single species A.tricorniculatus would cover a very wide range of morphotypes (Fig.58/H-L). The compromise which I have adopted is to separate the distinctive early forms as a separate species, A.primus. Other described species I regard as varieties, this has the distinct advantage for biostratigraphy that varieties such as A.p.primus and A.t.tricorniculatus can be arbitrarily restricted to the distinctive forms.

Amaurolithus primus Bukry & Percival 1971 Description: Robust amauroliths with well developed arch and short arms.

236 Amaurolithus primus Bukry & Percival 1971 var. primus Plate 9/10; Figure 57/L Cera toll thus prims BUKRY It PERCIVAL 1971, p.126, pi.1/12-14; HEKEL 1973 pi.1/15-16 HULLER 1974a pi. 11/3; EDUARDS l PERCH-NIELSEN 1974 pi.21/49-50. /}»aarolithus pritus (Bukry St Percival) GARTNER It BUKRY 1975 p.457, pi.7/g-i. PERCH-NTELSEN 1977 pl.1/3,5-6; PROTO-DECIHA et al 1978 pi.5/10; THEODQRIDIS 1984 pi.1/1-2. Description: Simple form with short arms, broad arch, and weakly developed keels. Remarks: Only forms which show the characteristic shape and expanded arch are included in the synonymy, all these specimens come from late NN11. Range: Reported range l.NNll-14, but with restricted definition probably confined to late NN11.

Amaurolithus primus var. amplificus (Bukry & Percival 1971) n.comb. Figure 57/H. Ceratolitbas aaplifieas BUKRY It PERCIVAL 1971 p.125, pi.1/5-7. Ceratolithas dentatas BUKRY 1973b p.676, pi.2/1-3. Amaral ithas atplificas (Bukry It Percival), GARTNER It BUKRY 1975 p.454, pl.6/g-l; PERCH-NIELSEN 1977 pi.2/3,5-6,8,11,13, pi.10/4-5; PROTO-DECIHA et al 1978 pi.5/8; THE0D0RIDIS 1984 pi.1/4-8; BERGEN 1984 pi.2/8. Description: Variety with short angular arms, left arm markedly shorter than the right, rods on right arm usually straight. Range: eNNll-early NN12.

Amaurolithus tricorniculatus (Gartner 1967) Gartner & Bukry 1975 Ceratolitbas tricoraicalatas GARTNER 1967b p.5, pi.10/4-6. Ceratolithas triconicalatus Gartner eaend. BUKRY It BRAHLETTE 1968, p. 152, pi.2/1-4.

Amaurolithus tricorniculatus Gartner 1967, (Gartner & Bukry 1975) var. tricorniculatus Figure 57/1 Ceratolitbas tricoraicalatas 6ARTNER 1967b p.5, pi.10/4-6; HULLER 1974a pi.11/11-12, 19/1. Huarolitbas tricoraicalatas (Gartner) GARTNER It BUKRY 1975 p.457, pl.8/c-h; PERCH-NIELSEN 1977 pi.6/1,4,7,11, 8/1,4,7-13, 9/1,4; ELLIS It L0HHANN 1979 pi.1/2; BERGEN 1984 pi.2/3,6. Description: Variety with right arm extended into apical spur. Arms typically delicate without well developed keels. Extra spurs may develop in the apical region. Remarks: Specimens intermediate to A.t.delicatus are common, and have been illustrated repeatedly (e.g. STRADNER 1973). Range: NN12-NN14.

237 Amauroil thus tricorniculatus var. blzzarus (Bukry 1973) n.corab. Figure 57/H Centolithus tricoraiculatus 6artner, BUKRY It BRAHLETTE 1968 pi.2/2. Ceratolithus bizzarus BUKRY 1973b p.676, pi.1/6-10. flaaarol ithus bizzarus (Bukry) GARTNER It BUKRY 1975 p.456, pl.B/a-b; BUKRY 1978 pi.2/1. Description: Extreme form of A.t.tricorniculatus - with long rod developed on the right arm, occasionally also a rod is developed from the upper keel on the arch. Range: Early Pliocene; few records but mostly from NN12.

Amaurolithus tricorniculatus var. delicatus (Gartner & Bukry 1975) n.comb. Figure 57/C Ceratolithus tricoraiculatus Gartner, MULLER 1974a partia pi.11/7-9. Haaurolithus delicatus GARTNER It BUKRY 1975, p.456, pl.7/a-f; STEINMETZ & STRADNER 1984 pi.10/1,3, 11/1; BERGEN 1984 pi.2/4-5; etcetera. Description: Generalised form, with smoothly curved longish arms. Remarks: A.delicatus is the name usually given to simple horseshoe shaped amauroliths, as such it is the most commonly observed form, and grades into most others. Later specimens very often show some birefringence, with straight length-slow extinction, they thus are hard to separate from Ceratolithus cristatus specimens. It is not clear whether this is due to development of nodes on one arm, and consequent tilting of specimens, or to rotation of the optic axis toward the horizontal. Range: eNNll to top NN14.

Amaurolithus tricorniculatus var. ninae (Perch-Nlelsen 1977) n.comb. Figure 57/C Ceratolithus aaplificus Bukry It Percival 1968, HEKEL 1973 pi.1/12-13. Ceratolithus tricoraiculatus Gartner 1967, STRADNER 1973 pi.37/5; EDWARDS It PERCH-NIELSEN 1974 pi.21/24,26,36-37,40-42. Aaaurolithus aiaae PERCH-NIELSEN 1977 p.745, pi.2/8-9, 4/3,6-14, 9/12-14, 49/5. Description: Generalised variety with well developed arch. Remarks: This is one of the two rather generalised varieties of A.tricorniculatus, it is distinguished from the other one, A.delicatus by broad development of the apical area. Range: approx 1NN11-NN12.

238 Genus Ceratollthus Kamptner 1950 Type species C.cristatus Kamptner 1950.

Cezatolithus azmatus Muller 1974 Plate 9/10; Figure 57/L. Cento] ithas arntus MULLER 1974a p.591, pi. 11 /4-6, 19/3-4; PERCH-NIELSEN 1977 pi.6/6,9-13, 7/1,2,4,6-7,9-10,12,14; GARTNER k BUKRY 1975 pi.5/f-i; ELLIS fc LOHMANN 1979 pi.1/4; BER6EN 1984 pi.2/10-11. Cento!ithas acatas GARTNER fc BUKRY 1974 p.116, pi.1/1-4; GARTNER & BUKRY 1975 pl.6/a-f; PR0T0-DECIMA et al 1978; BUKRY 1978 pi.2/3-4; pi.pi.3/5,8;10-11; BERGEN 1984 pi.3/1-6. Description: Ceratoliths with left arm extended apically, producing a triangular apical area. Right arm longer than left. Remarks: C.armatus and C.acutus were both described in 1974 from Indian Ocean sediments of Early Pliocene age (zone NN12, Sites 242 and 214 respectively). The holotypes, and paratypes, of the two species are very similar. The C.armatus specimens have slightly longer arms and more distinct ornament, but this is almost certainly a result of the much better preservation at Site 242. They are clearly synonyms, and not worth distinguishing, even at varietal level. According to the publication dates C.armatus has two months priority (May 1974 vs. July 1974). Range: late NN12.

Ceratolithus atlanticus Perch-Nielsen 1977 Plate 9/5; Figure 57/G Cento!ithas atlasticas PERCH-NIELSEN 1977 p.745, pi.3/1-14, 5/1-7,10, 49/2-4; BERGEN 1984 pi.4/4-5,7-8. Description: This species has a similar gross form and optical orientation to other Ceratolithus species, but both arms are extended into apical horns, the left arm is long and straight with an angular lower keel. The right arm is shorter, and typically has a couple of spines instead of keels. Remarks: The only two previous records are from sediments of zone NN12, I found several unambiguous specimens in sample 242-6-CC. This sample contains an excellent NN9 age nannoflora, with no evidence whatever of contamination. The species is very rare in the sample but occurred in all preparations. This occurrence is hard either to dismiss or to place in the context of the normal distribution of ceratoliths.

239 Ceratollthus cristatus Kamptner 1950 Plate 9/21,26; Figure 57/B Ceratolithas cristatas KAMPTNER 1950 p.154; KAMPTNER 1954 p.43 fig.44-45; NORRIS 1965 pi.11/1-4, 12/1-4; NORRIS 1971 p.906, fig.8; BORSETTI k CATI 1976 pi.17/1-16; SAMTLEBEN 1979 pi.5/7-10; NISHIDA 1979a pi.13/2; BERGEN 1984 pi.3/7-9, 4/3; etcetera. Ceratolithas ragosas BUKRY It BRAMLETTE 1968 p.152, pi.1/5-9; GARTNER k BUKRY 1975 pl.5/a-e; PERCH-NIELSEN 1977 pi.1/7-8,10-12, pi.9/7,10; PROTO-DECIMA et al 1978 pi.4/1-4,6-7,9; SAMTLEBEN 1979 pi.5/8-9; BERGEN 1984 pi.2/14-15; etcetera. Ceratolithas sitplex BUKRY 1978 p.310, pi.1/17-26. Remarks: The typical C.cristatus morphology is described above. C.rugosus was defined as a more heavily calcified version of C.cristatus, but was thought to be a separate species, not a preservational variant, on the grounds that it occurred consistently before C.cristatus even in well preserved material. Subsequent research has failed to support this distribution, typical C.cristatus can be found throughout the range of C.rugosus; there is little justification for separating the species. C.simplex was described as a small smooth version of C.cristatus, I found it was a common variety throughout the range of C.cristatus, and so probably is not worth distinguishing. Range: NN12-Recent.

Ceratolithus cristatus var. telesmus (Norris 1965) n.comb. Figure 57/D Ceratolithas telesias NORRIS 1965 p.21, pi.11/5-7, 13/1-3; GARTNER k BUKRY 1975 pl.4/d-h. Ceratolithas cristatas Kaiptner, BUKRY k BRAMLETTE 1968; MARTINI 1971 pi.2/3; NISHIDA 1979a pl.l3/2a. Ceratolithas cristatus aorpha telesius (Norris) BORSETTI k CATI 1976, p.224, pi.17/4. Description: Variety with long arms which curve together, almost coming into contact. Keels typically low. Remarks: Gradation into C.cristatus is plainly visible (eg BORSETTI & CATI 1976), but the form seems to be restricted to the Pleistocene. Range: NN19-Recent.

Ceratolithus cristatus var. separatus (Bukry 1978) n.comb. Figure 57/D Ceratolithas separatus BUKRY 1978 p.310, pi.1/1-16; BERGEN 1984 pi.4/1-2, 11/1-3; BUKRY 1985 pi.1/5. Description: Variety with both arms extended slightly beyond the arch, arms typically rather short and robust. Remarks: This is a distinctive, if not radical, modification of C,cristatus. It seems to have a restricted range. Range: ?Late Pliocene - Early Pliocene (sporadic specimens in my

240 material).

17.3.2 Family Triquetorhabdulaceae LIPPS 1969 Genus Orthorhabdus Bramlette & Wilcoxon 1967 Type species O.serratus Bramlette & Wilcoxon 1967.

Orthorhabdus serratus Bramlette & Wilcoxon 1967 Plate 9/11,17,22; Figure 56/E Iriquetorhabdulus sp. MARTINI 1965 p.408, pi.36/4-6. Orthorhabdus serratus BRAMLETTE It WILCOXON 1967 p.114, pi.9/5-10; LIPPS 1969 p.1030; MULLER 1974a pi.12/12. Triquetorhabdulus >artioii GARTNER 1967b p.6, pi.10/1-3; THE0D0RIDIS 1984 pi.10/11-16. Rhabdothorax serratus (Braalette k Wilcoxon), ROTH 1970 p.807. Remarks: This species is described above. Rhabdothorax is a genus of living spinose Thoracosphaerids (GAARDER & HEIMDAL 1973). As discussed above O.serratus shows close affinities to Triquetorhabdulus, so the combination Rhabdothorax serratus is unduly speculative, and unlikely to be justifiable. Range: NN2-6, in my material only occurs in samples from Site 242.

Genus Triquetorhabdulus Martini 1965 Type species T.carinatus Martini 1965

Triquetorhabdulus challenger1 Perch-Nielsen 1977 Plate 9/13,19,24; Figure 56/D Triquetorhabdulus challenger! PERCH-NIELSEN 1977, p.749, pi.36/3,7-8,10-11, 49/1; BI0LZI et al 1981 p.90; THE0D0RIDIS 1984 pi.10/9-10. Description: Short broad species with one to four subsidiary ridges parallel to the median lath. Optic axis parallel to length. Size: 8-15 microns long, typically about half as wide as long. Range: NN1-2 (records of Perch-Nielsen and Theodoridis, and occurs in samples of this age from Site 242).

Triquetorhabdulus carinatus Martini 1965 Plates 9/12,27-28, 10/15; Figure 56/A-B. Triquetorhabdulus carinatus MARTINI 1965 p.408, pi.36/1-3; BRAHLETTE k WILCOXON 1967 p.728, pi.9/14-16; LIPPS 1969 pi.126/1-4; MULLER 1974a pi.12/1; PERCH-NIELSEN 1977 pi.36/6; etcetera. Remarks: This species is described above. As noted by MARTINI (1965) specimens vary considerably in dimensions with two main types (Figs.56/A-B): (A) long thin ones with parallel sides (entire specimens 25-50 x 1.5-2.5 microns); (B) shorter and broader specimens with

241 distinct taper (15-25 x 2-4 microns). There does not, however, seem to be any consistent pattern to the distribution of these forms so I have not separated them. Range: NP25-NN2, common in my samples of this age.

Triguetorhabdulus milowli Bukry 1971 Plate 9/18,23; Figure 56/C. Iriqaetorhabdalas »ilouii BUKRY 1971c, p.325, pi.7/9-12; PERCH-NIELSEN 1977 pi.36/4,9; ELLIS fc LOHMANN 1979 pi.2/12; THEODORIDIS 1984 pi.10/3-4,7-8; MARTINI 1988 pi.2/6. Jriqaetorhabdalas aaritas STRADNER & ALLRAM 1981 p.595, fig.3, pi.7/1-8. Description: Short broad species of Triguetorhabdulus. Remarks: T.auritus was described as similar to T.milowii but with a more wedge shaped outline, and occasionally ridges at the ends. These seem to me minor points so I have, tentatively, included it in the synonymy. Size: 8-15 x 5-8 microns. Range: NP25-NN6, but mainly NN1-3.

Triguetorhabdulus rugosus Bramlette & Vilcoxon 1967 Plate9/1,7,29; Figure 56/F Iriqaetorhabdulas rugosus BRAMLETTE St UILC0X0N 1967 p.123, pl.9/17-18; MULLER 1974a pi.19/12; JAFAR 1975a, partii pi.7/10; PERCH-NIELSEN 1977 pi.36/1,2,5; PROTQ-DECIMA et al 1978; ELLIS & LOHMANN 1979 pi.2/12; THEODORIDIS 1984 pi.11/1-3. Ceratolithus iarnsuorthii GARTNER 1967b p.5, pi.9/1-4. Ceratolithus ragosus (Braslette fc Uilcoxon), LIPPS 1969 p.1030. Iriqaetorhabdalas sp. MULLER 1974a pi.12/6,9. Triquetorhabdulus farnsuorthii (Gartner) PERCH-NIELSEN 1984 p.42; PERCH-NIELSEN 1985b p.526, pi.76/3. Remarks: The structure of this species is described above. Specimens vary noticeably in shape and in degree of development of secondary ridges, the most extreme of these have been described as separate species, I think they should be regarded as intraspecific varieties, as listed below. It might also be worth subdividing the more normal forms, noted by PERCH-NIELSEN (1985b). Range: NN6-NN12.

Triguetorhabdulus rugosus var. extensus (Theodoridis 1984) n.comb. Figure 56/1 Iriqaetorhabdalas ex tens as THEODORIDIS 1984 p.83, pi.11/4-6. Description: Variety with one lath unusually broad. Remarks: This form is very similar to some specimens of T.striatus in shape, and has weak lateral ridges (Theodoridis 1984 pi.11/1), it may not be worth differentiating the two varieties. Range: NN12 (only recorded occurrence, I found a few specimens

242 approaching this morphology in NN11).

Triguetorhabdulus rugosus var. striatus (Muller 1974) n.comb. Plate 9/2-4,8—9; Figure 56/G-H 1riqaetorbabdalas striatus MULLER 1974a p.593, pi.12/4-5, 19/1,19. Iriquetorbabdulus ruqosus Braalette k Uilcoxon 1967, JAFAR partia pi.7/9. Description: Variety of T.rugosus with distinct subsidiary ridges parallel to the median ridge. Remarks: Some specimens are dramatically different from typical T.rugosus but they vary considerably, and weak ridges occur on some typical T.rugosus specimens (e.g. GARTNER 1967b pi.9/1). Range: NN9-1.NN11, I found specimens in samples 242-6-3-142 (NN10) 242-5-2-061 (1.NN11), and 239-5-2-142 (e.NNll).

Triquetorhabdulus finifer Theodoridis 1984 Figure 56/K Iriquetorhabdulus iiaifer THEODORIDIS 1984, p.89, pi.11/7-10. Remarks: This species (or variety of T.rugosus?) has a horseshoe shape, with a short broad median lath and elongated curved basal laths. Range: NN12 (only recorded occurrence).

17.3.3 Possibly related ortholiths. Genus Angulolithina Bukry 1973. Type species A.area Bukry 1973.

Angulolithina area Bukry 1973 Plate 9/6 hgalolithiaa area BUKRY 1973b, pi.1/1-5; PR0T0-DECIHA et al 1978 pi.5/14. Description: Horseshoe shaped nannofossil resembling ceratoliths, but with the optic axis parallel to the length of the horseshoe. The arms usually flare continuously with little curvature, and no distinct keels. Remarks: A.area is probably unrelated to the ceratoliths, and may even be a fragment of a larger . Range: Isolated occurrences from NN6 to NN16 in my material, scattered records in the literature of similar age.

Genus Minyllthina Bukry 1973 Type species M.convallis Bukry 1973

243 Hinylithina convallis Bukry 1973 Plate 9/14-15,20,25; Figure 56/J Kioflithina coovaUis BUKRY 1973b p.679f pi.3/12-18} PROTO-DECIHA et al 1978 pi.1/21; THEODORIDIS 1983 p.84, pi.4/1-7; KNUTTEL 1988 pi.8/7. Description: Nannoliths in the form of a rhomboidal plate with a raised rim. Acts as a single crystallographic unit with the optic axis nearly perpendicular to the plate. Remarks: The affinities of this species are entirely unknown, I have included it in this group since it is formed of a single crystallographic unit. As shown by THEODORIDIS (1984) the side view of the species is also distinctive (Plate 9/15,20). I do not, however, agree with his statement that the nannolith is composed of two elements; since it behaves optically as a single unit (his electron micrograph is, of a damaged specimen, with part of the rim missing). Size: 4-8 microns long, usually 4-6 microns. Range: mid NN9-eNNll, often abundant.

244 Sketches based ay ovn observations, and published aicrographs, iaportant sources HULLER (1974), PERCH-NIELSEN (1977), THEODORIDIS (1984). Diaaond syabols indicate orientation of optic axis; circle-optic axis perpendicular to paper. Cross-sections below plan views, (except 6-1, K). A-B, I . a r h i t a s ; C., I.tilovii; D. 1.chilletgerij E. Orthorhabdas serntas (plan l side views); F, T.ragosas var. ragosas; 6-H, l.r.striatasj I, I.r.exteasus; J, Hiiflithia convilUs; K, Whiter.

245 FIGURE 57 - CERATOLITHACEAE, MORPHOLOGY. A. Living c e l l , redrawn from NORRIS (1969), showing protoplast with enveloping c e r a tolith , within coccosphere of delicate hoop like coccoliths. B. Ceratollthus cristatus var. cristatus, upper, sid e and lower views showing keels asymmetry and rodded structure. C. Equivalent views of Amaurollthus trlcornlculatus var. d elicatu s, note d ifferen t asymmetry. D-L. Other taxa, upper surfaces; D. C.c.telesraus; E. C.c.seperatus; F. C.arraatus; G. C .atlanticu s; H. A.t.Mzzarus; I. A.t.tricorniculatus; J. A.t.ninae; K. A.primus var. amplificus; L. A.p.primus. 246 Ha NN21 NN20

COH H NN19 ►JH a. NN18 NN17

NN16

NN1S K m NN13/14) a ▼it£T NN12 i .l

NN11

NX 10

NN9 10 HN8

NN7

DC « 12 Z r H C_>M TJ'O O -H H H z NN6 X

14 HNS

NN4 18 NN3 l-e1 20 >1 NN2 1=3 ft 22 i i. NN1 24 I i

26 CSXI ta NP25 o<

FIGURE 58 - STRATIGRAPHIC DISTRIBUTION OF TRIQUETORHABDULACEAE AND CERATOLITHACEAE. Synthesis o£ my own and published data. Arranged to show likely phylogenetic relationships in the main groups. Sporadic occurrence/uncertain range indicated by broken lines. AlEvents used in published zonation schemes, small arrows less reliable events. N.B. Vertical scale changes at NN5/6 boundary. ABBREVIATIONS: amplif - amplificus; trie - tricorniculatus; bizz - bizzarus

247 CHAPTER 18 - SUMMARY AMD RECOMMENDATIONS.

18.1 SUMMARY OF PRINCIPAL CONCLUSIONS AND RESULTS 1. A model for coccolith function has been proposed. The primary function of coccoliths, and organic scales, is suggested to be protection, with secondary modifications of coccolith, and coccosphere, morphology for: enhanced protection; flotation; and production of a water buffer layer (Chapter 3). 2. A three phase model for heterococcolith rim formation has been developed: base-plate formation followed by; proto-coccolith ring nucleation; and element growth (Chapter 4). 3. Nucleation processes and basic element form are shown to be stable features, at least at generic level. By contrast base-plate development (and so proto-coccolith ring shape and size) is liable to strong variation at the species / intraspecific level. Degree of element development is also variable at this level, and mode of element development at species / generic level (Chapter 4 & 5). 4. A computer program has been written based on this model of morphological development. It has been used both to explore its consequences, and to produce illustrations of a range of Neogene coccoliths (Chapter 4, Appendix 2). 5. It is argued that biological species concepts can, and should, be applied to nannofossil taxonomy (Chapter 5). 6. It is shown that intraspecific variation is likely to be more important in nannofossils than is implied by conventional taxonomy; and so it is recommended that intraspecific taxonomic categories, and particularly varieties, are used (Chapters 5 & 10) 7. Consistent patterns are documented in size variation in assemblages of Neogene reticulofenestrid coccoliths from the Western Indian Ocean. In particular a "small Reticulofenestra interval" is identified in zone NN10, and assemblages from above and below this interval are differentiated (Chapter 7). 8. Neogene Nannofossil assemblages are described from Deep Sea Drilling Project Sites throughout the Western Indian Ocean; and the effects of raised salinity on Early Pliocene assemblages from the Red Sea are discussed (Chapter 8). 9. The biostratigraphy of the Coastal Makran region of Pakistan is re-assessed; evidence is presented that substantial areas are younger than was previously suggested; reworked nannofossils and assemblages from mud volcanoes are used to help reconstruct the geological history of the area (Chapter 9). 10. New structural interpretations are given for the Coccolithaceae

248 and Sphenolithaceae; and modified interpretations, with increased documentation of homologies, are given fox most other families. 11. Intraspecific variation is discussed in most Neogene nannofossil families, and a re-organised taxonomy is developed. A large number of new trinomial combinations are proposed. 12. New or revised schemes are given for the development of most families (particularly the discoasters) during the Neogene; and for the relationship of taxa within these families. 13. A preliminary description of size related variation in the Coccolithus pelagicus lineage is given, with demonstration that three distinct large varieties can be recognised, and used for biostratigraphy. Two new varieties, Coccolithus pelagicus var. pontus and C.p.nannopelagicus are described. 14. New occurrences are recorded of a number of poorly known species, including Ceratolithus atlanticus, Helicosphaera inversa, Reticulofenestra pseudoumbilica var. zotazia, Solidopons petrae, Tetralithoides symeondesii, Triquetorhabdulus challengezi; and particularly Cycloperfolithus carlae and Clausicoccus primalis, which were both found throughout the Miocene. 15. Discoaster tristellifer and Ceratolithus acutus are shown to be junior synonyms of D.altus and C.armatus, respectively. Also Pyrocyclus is shown to be a preservational artefact, and this interpretation is suggested for Catinaster mexicanus and several Discoastez species.

18.2 SUGGESTIONS FOR FUTURE RESEARCH. The majority of nannofossil research projects have, like mine, been rather broad based biostratigraphical and taxonomic studies. As the field has developed the returns from this kind of work have diminished. On the other hand there are a large number of interesting possibilities for more focused studies as briefly discussed below.

1. General nannofossil research. A. Lineage studies: Most groups appear to show gradualistic evolution, similar to that documented here in Reticulofenestra, and there are many opportunities for detailed studies of individual lineages. Promising Neogene examples include: the Helicosphaera carteri to intermedia group; the Discoastez deflandrei - exilis - variabilis lineage; Triguetorhabdulus rugosus; and Ceratolithus cristatus. B. Biometric studies: The study of Pseudoemiliania lacunosa presented here (Section 4.4.2) is only provisional and a detailed study by electron microscopy on well preserved material would almost certainly be worthwhile. Other suitable groups include: Calcidiscus leptopozus; Coccolithus pelagicus; and Umbilicosphaera slbogae.

249 C. Studies of evolution and assemblage changes during key intervals: The majority of change in Meogene nannofossil assemblages is concentrated in a few intervals, intensive study of any of these intervals could form an interesting project. The intervals are: NN5-6 (replacement of Cyclicargolithus floridanus by R.pseudoumbilica, and parallel changes in most other groups); NN9-eNNll (small Reticulofenestza interval, and main Discoastez evolutionary phase); NNl5-early NN16 (extinction of R.pseudourabilica and Sphenollthus abies, evolution of P.lacunosa and Gephyzocapsa, re-appearance of common C. pelagicus); NNl7-early NN19 (extinction of discoasters and almost complete change in nannofloras).

2. Palaeoecological possibilities. A. Study of salinity restricted and nearshore assemblages: The assemblages from the Red Sea are fascinating, and similar assemblages have been found in the Mediterranean, and in the Paratethys (e.g. JERKOVIC 1970, STRADNER & FUCHS 1980). There is a real possibility that a systematic study comparing such assemblages from different areas, and with open marine assemblages, could produce useful palaeoecological results. B. Study of assemblages from laminated black shales: As shown by a number of studies (e.g. NOEL 1973, COVINGTON 1985) such deposits often contain exceptionally preserved nannofloras also they often have ecologically restricted nannofloras (e.g. BUKRY 1974a). The Neogene is not a period in which such deposits are common but it would probably be worth actively attempting to find examples. C. Studies of geographical distribution of single families: As noted by PERCH-NIELSEN (1985b) the Helicosphaeraceae vary greatly in abundance from one area to another and even between samples. In addition the species compositions of coeval assemblages from different areas vary greatly, thus the Helicosphaeza walbezsdozfensis - stalls group is common in the Mediterranean Miocene (THEODORIDIS 1984, O.Varol & D.Catrullo pers. comms. 1987), but extremely rare in the Indian Ocean (my obs.). This variation is probably ecologically caused and a wide ranging study of the distribution of this family would be worthwhile.

3. COMPLEMENTARY RESEARCH ON LIVING NANNOPLANKTON. Although certain species have been extremely well studied, and general taxonomy is becoming increasingly well established, there is an obvious need for basic biological observations on a wider range of species. In addition systematic study is needed of intraspecific variation in living species, in particular the controls on, and significance of coccolith size variation need to be examined.

250 REFERENCES

N.B. Periodical abbreviations follow the British Huseua (Natural History) catalogue, except; IRDSDP - Initial Reports of the Deep Sea Drilling Project. A nuiber of early taxonoaic references are not included, these can all be found in Part I of the Annotated Index and Bibliography of the Calcareous Nannoplankton, LOEBLICH & TAPPAN (1966).

AHMED S.S. 1969: Tertiary geology of part of South Hakran, Baluchistan, West Pakistan. Bull. Aa. Ass. Petrol. 6eol., 53, 1480-1499. ANAND P.L. 1937: A taxonoaic study of the algae of the British -cliffs. J. Bot. Lond., 75, 1-51. ARTHURTON R.S., FARAH A. It AHMED W. 1982: The Late Cretaceous - Cenozoic history of western Baluchistan Pakistan - the northern aargin of the Makran subduction coaplex. In, LE66ETT J.K. (ed.): ■ Trench-Forearc 6eology* Spec. Publ. geol. Soc. Lond., 10, p.373-385. AUBRY 1984: Handbook of Cenozoic Calcareous nannoplankton. Book 1: Ortholithae (Discoasters). Micropalaeontology Press (New York), p.1-265. BACKMAN J. 1978a: Late Miocene and Early Pliocene nannofossil biochronology and biogeography in the Vera Basin, S.E. Spain. Stockh. Contr. Geol. 32, 93-114. BACKHAN J. 1978b: Pliocene biostratigraphy of DSDP sites 111 & 116 froa the North Atlantic Ocean and the age of the Northern Heaisphere glaciation. Stockh. Contr. 6eol. 32, 115-137. BACKMAN J. 1980: Miocene - Pliocene nannofossils and sediaentation rates in the Hatton - Rockall Basin, NE Atlantic Ocean. Stockh. Contr. 6eol. 36, 1-93. BACKMAN J. 1984: Cenozoic calcareous nannofossil biostratigraphy froa the North Eastern Atlantic Ocean, DSDP Leg 81. IRDSDP, Bl, 403-428. BACKMAN J. 1986: Accuaulation patterns of Tertiary calcareous nannofossils around extinctions. 6eol. Rdsch., 75, 185-196 .BACKMAN J. It SHACKLETON N.J. 1983: Quantitative biochronology of Pliocene and Early Pleistocene - calcareous nannofossils froa the Atlantic, Indian and Pacific oceans. Mar. Nicropaleont., 8, 141-170. BACKHAN J. fc SHACKLETON N.J. & TAUXE L. 1983: Quantitative nannofossil correlation to open ocean deep sea sections froa the Plio-Pleistocene boundary at Vrica, Italy. Nature, 304, 156-158. BARBER A . J , TJ0KR0SAP0ETR0 S. i CHARLTON T .R . 1986: Mud v o lc a n o e s , s h a le d i a p i r s , wrench f a u l t s , and a§langes in accretionary coaplexes, eastern Indonesia. Bull. Aa. Ass. Petrol. 6eol., 70, 1729-1741. BAUMANN P. It ROTH P.H. 1969: Zonierung des Obereozans und Oligozans des Monte Cagnero (Zentral Appenin) ait planktonischen Foraainiferen und Nannoplankton. Eclog. geol. Helv., 62, 303-323. BERBERIAN F. h BERBERIAN M. 1981: Tectono-plutonic episodes in Iran. In, GUPTA H.K. It DELANEY F.M. (eds.) ■ Zagros, Hindu Kush, Hiaalaya, geodynaaic evolution*, Geodynaaics Ser. Aa. geophys. Union, 3, p .5 - 3 2 . BERGEN J.A. 1984: Calcareous nannoplankton froa DSDP leg 78A: Evidence for iabricate underthrusting at the Lesser Antillean active aargin. IRDSDP, 78A, 411-447. BERGGREN W .A ., KENT D .V ., FLYNN J . J . It Van COUVERING J . A . 1986: C e n o zo ic g e o c h ro n o lo g y . B u l l . g e o l. S oc. Aa., 96, 1407-1418. BERGGREN W.A. It Van COUVERING J . A . 1974: The L a te Neogene - b io s t r a t ig r a p h y , g eo ch ro n o lo g y and paleocliaatology of the last 15 aillion years in aarine and continental sequences. Palaeogeogr., Palaeocliaatol. Palaeoecol., 16, 9-35. BI0LZI H., PERCH-NIELSEN K. It RAMOS I; Triquetorhabdulus - an Oligocene - Miocene calcareous nannofossil genus. Newsl. Int. Nannoplankton Ass., 3, 89-92. BLACK M. 1963: The fine structure of the aineral parts of coccolithophoridae. Proc. Linn. Soc. Lond., 174, 41-46. BLACK H. 1967: New naaes for soae coccolith taxa. Proc geol. Soc. Lond. 1640, 139-145. BLACK M. 1968: Taxonoaic probleas in the study of coccoliths. Palaeontology, 11, 793-813. BLACK M. 1971: The systeaatics of coccoliths in relation to the palaeontological record. In,FUNNELL B.M. It RIEDEL W.R. (Eds.): 'The Micropalaeontology of the Oceans*. Caabridge University Press(Caabridge), p . 611-624. BLACK M. 1972: Crystal developaent in discoasteraceae and braarudosphaeraceae (planktonic algae). Palaeontology, 15, 476-489.

250A BLACK H. 1972-1975: British Lover Cretaceous coccoliths, 1. Gault Clay. Palaeontogr. Soc. Honogr., 126-127, 1-142. BLACK N. It BARNES B. 1961: Coccoliths and discoasters froi the floor of South Atlantic Ocean. J. R. •icrosc. Soc., 80, 137-147. BONEY A.D. 1970: Scale bearing phytoflagel1ates: An inter is review. Oceanogr. ear. Biol. A. Rev., B, 251-305. BORSETTI A.N. & CATI F. 1972: II nannoplancton vivente nel Tireno centro-ieridionale. Giorn. Geol., 3B, 385-452. BORSETTI A.H. It CATI F. 1976: II nannoplancton vivente nel Tireno centro-»eridionale parte II. Giorn. Geol., 40, 209-240. BOUDREAUX J.E. 1974: Calcareous nannoplankton ranges, DSDP Leg 23. IRDSDP, 23, 1073-1090. BOUDREAUX J.E. It HAY W.W. 1969; Calcareous nannoplankton of the late Pliocene - Pleistocene - Recent sediients in the Subiarex cores. Revta. esp. Micropalaeont., 1, 249-292. BRAARUD T., GAARDER K.R., NARKALI J. fc N0RDLI E. 1952: Coccolithophorids studied in the electron iicroscope. Nytt. Hag. Bott., 1, 129-134. BRAHLETTE H.N. It RIEDEL W.R. 1954: Sratigraphic value of discoasters and soae other aicrofossils related to recent coccolithophores. J. Paleont., 28, 385-403. BRAMLETTE H.N. It NILC0X0N J.A. 1967: Hiddle Tertiary calcareous nannoplankton of the Cipero Section, Trinidad, Nest Indies. Tul. Stud. Geol. Paleontol., 5, 93-131, It 220. BRASIER H. 1986: Why do lover plants and aniials bioiineralize? Paleobiology, 12, 241-250. BREHERET J.G. 1978: Foraes nouvelles quaternaires et actuelles de la faaille des GAphyrocapsaceae (Coccolithophorides). C.r. hebd. SAanc Acad. Sci. Paris, D2B7, 447-449. BREITZHAN L.L., BIRNIE R.W., JOHNSON G.D. 1983: Fission-track ages of the Chagai intrusives, Baluchistan, Pakistan. Bull. geol. Soc. Aa., 94, 253-258. BUKRY D. 1971a: Discoaster evolutionary trends. Hicropaleontology, 17, 43-52. BUKRY D. 1971b: Coccolith stratigraphy Leg 6, DSDP. IRDSDP., 6, p.965-1004. .BUKRY D. 1971c: Cenozoic calcareous nannofossils froa the Pacific Ocean. Trans. S.Diego Soc. nat. Hist., 16, 303-328. BUKRY D. 1971d: Coccolith stratigraphy, Leg 7 DSDP. IRDSDP, 7, 1513-1518. BUKRY D. 1973a: Lov latitude coccolith biostratigraphic zonation. IRDSDP, 15, 685-703. BUKRY D. 1973b: Coccolith stratigraphy, East Equatorial Pacific, Leg 16 DSDP. IRDSDP, 16, 653-711. BUKRY D. 1973c: Phytoplankton stratigraphy, DSDP Leg 20, Western Pacific Ocean. IRDSDP, 20, 307-317. BUKRY D. 1974a: Coccoliths as paleosalinity indicators - evidence froa the . In, "The Black Sea - Geology, Cheaistry, and Biology'. Hen. An. Assoc, petrol. Geol., 20, 353-363. BUKRY D. 1974b: Coccolith and si 1icoflagel1 ate stratigraphy, Eastern Indian Ocean, DSDP Leg 22. IRDSDP, 22, 601-605. BUKRY D. 1974c: Coccolith stratigraphy, Arabian and Red Seas, DSDP Leg 23. IRDSDP, 23, 1091-1093. BUKRY D. !974d: Coccolith zonation of cores froa the Western Indian Ocean and the Gulf of Aden, DSDP Leg 24. IRDSDP, 24, 995-996. BUKRY D. 1974e: Phytoplankton stratigraphy, offshore East Africa, DSDP Leg 25. IRDSDP, 25, 635-646. BUKRY D. 1975: Coccolith and silicoflagellate stratigraphy, North West Pacific Ocean, DSDP Leg 32, Sites 303-313. IRDSDP, 32, 677-701. BUKRY D. 1976: Coccolith stratigraphy of Hanihiki Plateau, Central Pacific, DSDP Site 317. IRDSDP, 33, 493-501. BUKRY D. 1978: Neogene coccolith stratigraphy, Hid-Atlantic Ridge, DSDP Leg 45. IRDSDP, 45, 307-317. BUKRY D. 1981: Pacific Coast coccolith stratigraphy between Point Conception and Cabo Corrientes, DSDP Leg 63. IRDSDP, 63, 445-471. BUKRY D. 1985: Hid-Atlantic Ridge coccolith and silicoflagellate biostratigraphy, DSDP Sites 55B It 563. IRDSDP, 82, 591-602. BUKRY D. It BRAHLETTE H.N. 1968: Stratigraphic significance of two genera of Tertiary Calcareous nannofossils. Tul. Stud. Geol. Paleontol., 6, 149-155. BUKRY D. It BRAHLETTE H.N. 1969: Soae new and stratigraphically iaportant calcareous nannofossils froa the Cenozoic. Tul. Stud. Geol. Paleontol., 7, 131-142. BUKRY D. It PERCIVAL S.F. 1971: New Tertiary Calcareous nannofossils. Tul. Stud. Geol. Paleontol., 8, 123-146.

251 BURNS D.A. 1973: Structural analysis of flanged coccoliths in sediients fro* the South West Pacific Ocean. Revta. esp. Micropalaeont. 5, 147-160. BURNS D.A. 1975: Distribution abundance and preservation of nannofossils in Eocene to Recent Antarctic sediaents. N.Z. J. 6eol. Geophys., 18, 583-595. BURNS D.A. 1977: Phenotypes and dissolution aorphotypes of the genus Sephyrocapsa Kaiptner and E» iiiasia kaxleyi (Lohaann). N.Z. J. Geol. Seophys. 20, 143-156. BYBELL L. & 6ARTNER S. 1972: Provincialisa aaong aid-Eocene calcareous nannofossils. Hicropaleontology, 18, 319-336. CATI F. It BORSETTI A.H. 1970: I discoasteridi del Miocene delle Harche. Giorn. Geol., 36, 617-652. CATI F, It BORSETTI A.H. 1972: Nuove noae per un Discoasteride del Hiocene del 1e Marche. 6iorn. Geol., 38, 373. CEPEK P. & HIND F.H. 1979: Neogene and Quaternary calcareous nannofossils froa DSDP Site 397 (NU African aargin). IRDSDP, 47, 289-315. CHRISTENSEN T 1962: A lg e r . In, BOCHER T . H . , LANGE H. It SORENSEN T. (EDS): " B o ta n ik , ( S y s te a a tik B o t a n ik ) " , 2, p .1-178. CHRISTENSEN T. 1978: Annotations to a textbook of phycology. Bot. Tidsskr., 73, 65-70. CLOCHIATTI M. 1969: Contribution a l ’Atude de Helieosphaera carteri (Uallich) Kaaptner (Coccolithophoridae). Revue aicropalAont., 12, 75-83. CLOCHIATTI M. 1971: Sur l ’existence de coccosphfcres pourtant des coccolithes de 6ephyrocapsa oceaaica et de [ajliaaia haxleyi (Coccolithophorid^s). C. r. hebd. S6anc. Acad. Sci. Paris, D273, 318-321. COHEN C.L.D. 1964: Coccolithophorids froa two Caribbean deep-sea cores. Hicropalaeontology, 10, 231-250. COHEN C.L.D. & REINHARDT P. 1968: Coccolithophorids froa the Pleistocene Caribbean deep-sea core CP-28. Ns. Jb. Geol. Palaont, Abh., 131, 289-304. C0VIN6T0N M. 1985: New aorphological inforaation on Cretaceous nannofossils froa the Niobara Foraation (Upper Cretaceous) of Kansas. Geology, 13, p.683-686. CRAME J.A. 1984: Neogene and Quaternary aollusca froa, the Makran coast, Pakistan. In, HAQ It MILL IMAN (eds.): "Marine geology and oceanography of Arabian Sea and coastal Pakistan", Van Nostrand Reinhold (New York), p.45-61. DEFLANDRE G. 1942: CoccolithophoridGes fossilies D’Oranie. Genres Scyphospbaera Lohaann et Ihoracospbaera Ostenfeld. Bull. Soc. Hist. Nat. Toulouse, 77, 125-137. DEFLANDRE 6. 1952: Classe de coccolithophoridAs (Coccolithophoridae Lohaann, 1902). In, GRASSE P.P. (ed.), "TraitA de zoologie", Toae 1, fasc.l, p.339-364. DEFLANDRE G. It FERT 1954: Observations sur les coccolithophorides actuels et fossilts en aicroscopie ordinaire et dlectronique. Annls. Padont., 40, 115-176. DIXON H.H. 1900: On the structure of coccospheres and the origin of coccoliths. Proc. R. Soc., B64, 305-315. DODGE J.D. 1973: "The fine structure of algal cells". Acadeaic Press (London), p .1-261. DYKSTRA J.D. & BIRNIE R.W. 1979: Segaentation of the Quaternary subduction zone under the Baluchistan region of Pakistan and Iran. In, FARAH A. It DeJONG K.A. (eds.): "Geodynaaics of Pakistan", Geol. Surv. Pakistan, p.319-323. EDWARDS A.R. 1973: Key Species of New Zealand calcareous nannofossils. N.Z. J. Geol. Geophys., 16, 68-89. EDWARDS A.R. & PERCH-NIELSEN K. 1974: Calcareous nannofossils froa the Southern South West P acific, DSDP Leg 29. IRDSDP, 29, 469-539. ELDRED6E N. It 60ULD 5.J. 1972: Punctuated equilibria: An alternative to phyletic gradualisa. In, SCH0PF T.J.M. (Fd.): "Models in Paleobiology*. Freeaan Copper (San Francisco), 82-115. ELLIS C.H. 1982: Calcareous nannoplankton biostratigraphy, DSDP Leg 60. IRDSDP, 60, 507-535. ELLIS C.H. It L0HMANN W.H. 1979: Neogene calcareous nannoplankton biostratigraphy in Eastern Mediterranean deep sea beds (DSDP Leg 42A, Sites 375 It 376). Mar. aicropaleont., 4, 61-84. ELLIS C.H., L0HMANN W.H. It WRAY J.L. 1972: Upper Cenozoic calcareous nannofossils froa the Gulf of Mexico, DSDP Leg 1, Site 3. Colo. Sch. Hines Q, 67, 1-103. FALCON N.L. 1974: An outline of the geology of the Iranian Hakran. Geogrl. J., 140, p.284-291. FARHOUDI 6. It KARIG D.E. 1977: Makran of Iran and Pakistan as an active arc systea. Geology, 5, 664-668. FARINACCI A. 1971: Roundtable on calcareous nannoplankton Roaa. In FARINACCI A. (ed.), Proc. II Planktonic Conf., Roaa 1970, Vol. 2, 1343-1369. FILEHICZ H.V. 1985: Calcareous nannofossil biostratigraphy of the Middle Aaericas Trench and Slope, DSDP Leg

252 84. IRDSDP, 84, 339-361. FLEISHER R.L. 1974a: Cenozoic planktonic foraainifera and biostratigraphy, Arabian Sea, DSDP Leg 23A. IRDSDP, 23, 1001-1072. FLEISHER R.L. 1974b: Preliiinary report on late Neogene Red Sea foraainifera, DSDP Leg 23B. IRDSDP, 26, 985-1012. FLORES J.A. L S1ERR0 F.J. 1987: Calcareous nannoplankton in the Tortonian / Messinian transition series of northeastern edge of the Guadalquivir Basin. In, STRADNER H It PERCH-NIELSEN K. (Eds.): Procs. Int. Ass. Nannoplankton Ass. Meeting, Vienna 1985. Abh. Geol. Bundesanst. 39, 67-84. FONSECA B. 1976: Notes sur la gAologie et palAontologie du HiocAne de Lisbonne, XVII. Coccolithas taganas, nouvelle espAce de coccolithophoridA du MiocAne de Lisbonne. Boll. Soc. geol. Port., 20, 29-32. FREUND R., ZAK I. & 6ARFUNKEL Z. 1968: Age and rate of the sinistral loveient along the Dead Sea Rift. Nature, 20, 253-255. FROST B.W. 1980: Grazing. In MORRIS I. (ed.): "The physiological ecology of phytoplankton'. Univ. of California Press (Berkeley), p.465-492. FUTTERER D. 1976: Kalkige dinoflagellaten (Calciodinelloideae*) und die systeiatische stellung der Thoracosphaeroideae. Ns. Jb. Geol. Palaont., Abh., 151, 119-141. FUTTERER D. 1978: Distribution of calcareous dinoflagel1ates in Cenozoic sediients of Site 366, Eastern North Atlantic. IRDSDP, 41, 709-738. 6AARDER K .R . 1970: Three neu ta x a o f c o c c o lit h in e a e . N y tt Hag. B o t t . , 17, 113-126. 6AARDER K .R . & HEIMDAL B .R . 1973: L ig h t and sc an n in g e le c t r o n lic r o s c o p y o b s e r v a t io n s on Rhabdothorax regale (6aarder) Gaarder n.coib. Noru. J. Bot., 20, 89-97. GAARDER K .R . & HEIMDAL B .R . 1977: A r e v is io n o f th e genus Syracosphaera Lohaann (Coccolithineae). 'Meteor' Forsch. Ergebnisse, D24, 54-71. SANSSER A. 1971: The Taftan Volcano (S.E. Iran). Eclog. geol. Helv., 64, 319-334. 6ARD G. 1987: O b s e rv a tio n s on a d ia o r p h ic c o cc o s p h e re . In, STRADNER H. fc PERCH-NIELSEN K. ( E d s .) : P r o c s . Int. Ass. Nannoplankton. Meeting, Vienna 1985. Abh. geol. Bundesanst., 39, 85-87. GARTNER S. 1967a: Nannofossil species related to Cydococcolithus leptoporus (Hurray & Blackaan). Paleont. Cont. Univ. Kansas, Paper 28, 1-4. GARTNER S. 1967b: Calcareous nannofossils froa the Neogene of Trinidad, Jaaaica and Gulf of Mexico. Paleont. Cont. Univ. Kansas, Paper 29, 1-7. GARTNER S. 1969: Correlation of Neogene planktonic foraainifera and calcareous nannofossil zones. Trans Gulf Cst. Ass. geol. Socs., 19, 585-599. GARTNER S. 1971: Calcareous nannofossils froa the J0IDES Blake Plateau cores and revision of Paleogene nannofossil zonation. Tul. Stud. Geol. Paleontol., 8, 101-122. GARTNER S . 1974: N a n n o fo s s il b io s t r a t ig r a p h y , Leg 22, DSDP. IRDSDP, 22, 577-601. GARTNER S. 1977: Calcareous nannofossil biostratigraphy and revised zonation of the Pleistocene. Har. ■ icropaleont., 2, 1-25. GARTNER S . It BUKRY D. 1969: T e r t i a r y h o lo c o c c o lit h s . J . P a la e o n t ., 43, 1213-1221. 6ARTNER S . It BUKRY D. 1974: Ceratolitbus acutas 6artner & Bukry n.sp. and Ceratolithas atplificus Bukry & Percival noienclatural clarification. Tul. Stud. 6eol. Paleontol., 11, 115-118. GARTNER S. It BUKRY D. 1975: Morphology and phylogeny of the coccolithophycean faaily Ceratolithaceae. J. Res. U.S. geol. surv., 4, 451-465. GIRDLER R.W. It STYLES P. 1974: Two stage Red Sea floor spreading. Nature, 247, 7-11. GIRDLER R.W. It STYLES P. 1978: Sea floor spreading in the western Gulf of Aden. Nature, 271, 615-617. GIR0D M. It CONRAD G. 1975: Les foriations volcanologiques rAcentes du Sud de 1f Iran (Kouh-e-Shahsararan): donnAes pAtrologiques prAliainaires; iaplications structurales. Bull, volcan., 39, 1-17. 60Y G. 1981: Nannofossiles calcaires des schistes carton (Toarcien InfArieur) du Bassin de Paris. Doc. RCP 459, editions Bur. Rech. gAol. ain., 1-86. GREEN J.C. 1986: Bioaineralization in the algal class Pryanesiophyceae. In, LEADBEATER B.S.C. It RIDING R. (eds.): 'Bioaineralization in lower plants and aniaals'. Systeaatics Assoc. Spec. Vol., 30, p .1-401.

GRUN W., PRINS B. It ZWEILI F. 1974: Coccolithophoriden aus dea Lias epsilon von Holzaaden (Deutschland). Ns. Jb. Palaont., Abh., 147, 294-328. HAMILTON G .B . It HQJJATZADEH M. 1978: C e n o zo ic c a lc a r e o u s n a n n o f o s s ils - A r e c o n n a is s a n c e . In , LORD A .R . (ed.): 'A stratigraphical index of calcareous nannofossils", Br. aicropalaeont. Soc. Series, E llis

253 Horwood (Chichester), p .136-167. HAQ B.U. 196S: Electron licroscope studies of soie Eocene calcareous nannoplankton froi Syria. Stockh. Contr. Geol. 15, 23-38. HAQ B.U. 1968: Studies on Upper Eocene calcareous nannoplankton froi Northwest Genany. Stockh. Contr. 6eol. 18, 13-74. HAQ B.U. 1971a: Palaeogene calcareous nannoflora. 1, The Palaeocene of Uest Central Persia and the Upper Palaeocene - Eocene of Uest Pakistan. Stockh. Contr. 6eol. 25, 1-56. HAQ B.U. 1971b: Palaeogene calcareous nannoflora. II, Oligocene of Uestern Geraany. Stockh. Contr. Geol. 25, 57-9 7 . HAQ B.U. 1971c: Palaeogene calcareous nannoflora. Ill, Oligocene of Syria. Stockh. Contr. Geol. 25, 57-129. HAQ B.U. 1973: Evolutionary trends in the Cenozoic coccolithophore genus Helicopostosphnera. Hicropaleontology, 19, 32-52. HAQ B.U. 1976: Coccoliths in cores froa the Bellinghausen abyssal plain and antarctic continental rise, (DSDP Leg 3 5 ). IRDSDP, 35, 557-568. HAQ B.U. 1978: Calcareous nannoplankton. In, HAQ B.U. It BOERSKA A. (Eds.): 'Introduction to Narine Hicropaleontology*. Elsevier (New York), p.79-109. HAQ B.U. 1980: Biogeographic history of Hiocene calcareous nannoplankton and paleo-oceanography of the Atlantic Ocean. Hicropaleontology 26, 414-443. HAQ B.U. 1984: Jurassic to Recent nannofossil biochronology: an update. In HAQ B.U. (Ed.); 'Nannofossil biostratigraphy*, Benchiark Papers in Geology, 78, Hutchinson Ross (Stroudsberg PA), p.359-378. HAQ B.U. b BER66REN U.A. 1978: Late Neogene calcareous nannoplankton biochronology of the Rio 6rande Rise (S. Atlantic). J. Paleont., 52, 1167-1194. HAQ B . U . , HARDENBOL J . , VAIL P . R . , URIGHT R . C . , STOVER L . E . , BAUM G . E . , LOUTIT T . , G0MB0S A ., DAVIES T . , PFLUN C., ROHINE K., POSAHENTIER H. in press: Chronology of fluctuating sea levels since the Triassic (250 Billion years ago to the present). HAQ B.U. & LIPPS J.H. 1971: Calcareous nannoplankton. IRDSDP, 8, 777-789. .HAQ B.U. It LOHMANN G.P. 1976: Early Cenozoic calcareous nannoplankton biogeography of the Atlantic ocean. Har. Hicropaleont., 1, 119-194. HAQ B.U., LOHMANN 6.P. & UISE S.U. 1976: Calcareous nannoplankton biogeography and its ’ paleocliaiatological iaplications: Cenozoic of the Falkland Plateau, (DSDP 36), and Hiocene of the Atlantic Ocean. IRDSDP, 36, 745-759. HARHS J.C ., CAPPEL H.N. It FRANCIS D.C. 1982: Geology and petroleu# potential of the Hakran coast, Pakistan. In, Procs. Offshore South East Asia conference, Singapore, p .1-9. HARHS J.C ., CAPPEL H.N. It FRANCIS D.C. 1984: The Hakran Coast of Pakistan: It's stratigraphy and hydrocarbon potential. In, HAQ It NILLIHAN (eds.): "Narine geology and oceanography of Arabian Sea and coastal Pakistan*, Van Nostrand Reinhold (New York), p.3-26. HART H.B. 1980: A water depth aodel for the evolution of planktonic foraainiferida. Nature, 286, 252-254. HAY H.W. 1968: Coccoliths and other calcareous nannofossils in aarine sediaents in Cyrenaica. In BARR F.T. (ed.): "Geology and archaeology of Northern Cyrenaica, Libya.* Petroleua Exploration Society of Libya, p .149-157. HAY U.U. & BEAUDRY F.H. 1973: Calcareous nannofossils, Leg 15 DSDP. IRDSDP, 15, 625-684. HAY U .U ., HOHLER H . P . , ROTH P .H ., SCHHIDT R .R . k BOUDREAUX J.E. 1967: Calcareous nannoplankton zonation of the Gulf Coast and Caribbean - Antillean area and transatlantic correlation. Trans. Gulf Cst. Ass. geol. Socs., 17, 428-480. HAY U.U., HOHLER H.P. It HADE H.E. 1966: Calcareous nannofossils froa the Nal'Chik (NH Caucasus). Edog. geol. Helv. 59, 379-401. HAY H.W. & T0HE K.H. 1962: Electronaicroscopic exaaination of soBe coccoliths froa Donzacq (France). Edog. geol. Helv., 55, 497-517. HEEZEN B.C., LYNDE R.P. It F0RNARI D.J. 1978: Geologic Bap of the Indian Ocean. In, "Geological Horld Atlas*, UNESCO, sh eet 21. HEIHDAL B.R. 1982: Validation of the naaes of soae species of Zygosphaera Kasptner. Newsl. Int. Nannoplankton. Ass., 4, 52-56. HEIHDAL B.R. It GAARDER K.R. 1980: Coccolithophorids froB the northern part of the eastern Atlantic (I) Hoiococcoliths. "Heteor* Forsch. Ergebnisse, D32, 1-14. HEIHDAL B.R. It GAARDER K.R. 1981: Coccolithophorids froa the northern part of the eastern Atlantic (II)

254 Heterococcoliths. "Meteor" Forsch. Ergebnisse, D33r 3 7 -6 9 . HEKEl H. 1973: Nannofossil biostratigraphy, Leg 20 DSDP. IRDSDP, 20, 221-247. HIBBERD D.J. 1976: The ultrastructure and taxonoiy of the Chrysophyceae and Pryinesiophyceae (Haptophyceae): A survey with soie new observations on the ultrastructure of the Chrysophyceae. J. Linn. Soc., Bot., 72, 55-80. HIBBERD D.J. 1980: Pryinesiophytes (=Haptophytes). In, COX E.R. (Ed.): "Phytoflagellates". Elsevier (North Holland), 273-318. HOJJATZADEH M. 1978: Discoasters of the Blue Clay (Middle Miocene) of Malta and Gozo. 6eol. Mag. 115, 1-19. H0LLI6AN P.M., VIOLLIER H., HARBOUR D.S., CAMUS P. t CHAMPAGNE-PHILLIPPE N. 1983: S atellite and ship studies of coccolithophore production along a continental shelf edge. Nature, 304, 339-342. HOLM HANSEN 0. 1984: Phytoplankton dynaiics in the aarine food web. In, HAQ l HILLIHAN (eds.): "Marine geology and oceanography of Arabian Sea and coastal Pakistan", Van Nostrand Reinhold (New York), p .32 7-3 3 5 . H0NJ0 S. 1976: Coccoliths: production, transportation, and sediaentation. Mar. Micropaleont., 1, 65-79. HUNTING SURVEY CORPORATION LIMITED 1960: R e con n aissan ce g e o lo g y o f p a rt o f Hest P a k is ta n . 6 ove rna en t of Canada (Toronto), p .1-550. HUTCHINSON G .E . 1967: "A T r e a t is e on L ia n o lo g y , v o l . 2 . I n tr o d u c tio n to la k e b io lo g y and th e lia n o p la n k t o n " . J.W iley It Sons (New York) p .1-1115. INTERNATIONAL CODE OF BOTANICAL NOMENCLATURE 1983: Sydney 1981, S t a f le u F .A . ( e d .) . Regnua v e g ., I l l , 1-472. IN0UYE I. It PIENAAR R.N. 1984: New observations on the coccolithophorid lltbilicosphaera sibogae v a r . foliosa (Pryanesiophyceae) with reference to cell covering, cell structure, and flagellar apparatus. Br. phycol. J., 19, 357-369. JACOB K.H. It QUITTMEYER R.L. 1979: The Makran region of Pakistan and Iran: Trench-Arc systeas with active plate subduction. In, FARAH A. It DeJONG K.A. (eds.): "Geodynaaics of Pakistan", Geol. Surv. Pakistan, p.305-318. JAFAR S.A. 1975a: Calcareous nannoplankton froa the Miocene of Rotti Indonesia. Verh. K. ned. Akad. wet., Eerste Reeks 28, 1-99. JAFAR S.A. 1975b: Soae coaaents on the calcareous nannofossil genus Scyphosphaera and the neotypes of Scypbosphaera spp. froa Rotti, Indonesia. Senckenberg. Leth., 56, 365-381. JAFAR S.A. & MARTINI E. 1975: on the validity of the nannoplankton genus Helicosphaera. Senckenberg. leth., 56, 381-397. JANIN M.C. 1981: Essai de datation de concretions polya£talliques et evolution Quaternaire du coccolithe Cycloeoccolithus leptoporus - >achtyrei. Bull. Soc. geol. Fr., 23/3, 287-296. JERK0VIC M .L . 1970: Hoelaerhthdus nov. gen. type d'une nouvelle faaille de Coccolithophoridees fossiles: Noelaerhabdaceae du Miocene sperieur de Yougoslavie. C.r. hebd. Seanc. Acad. Sci. Paris, 270, 468-470. JIANG M.J. It GARTNER S. 1984: Neogene and Quaternary calcareous nannofossil biostratigraphy of the Walvis R id g e . IRDSDP, 74, 561-595. KAMPTNER E. 1950: Uber den subaikroscopischen Aufbau der Coccolithen. Anz. ost. Akad. Hiss. Math. Naturw. Klasse, 87, 152-158. KAMPTNER E. 1954: Untersuchungen uber den Feinbau der Coccolithen. Arch. Protistenk., 100, 1-90. KAMPTNER E. 1955: Fossile Coccolithineen - Skelletreste aus Insulinde. Verh. K. Ned. Akad. wet., 50, 1-105. KAHPTNER E. 1963: Coccolithineen - Skeletreste aus Tiefseeablagenungen des Pazifischen Ozeans. Annin, natur. Mus. Wien, 66, 139-204. KAMPTNER E. 1967: Kalflagellaten - Skeletreste aus Tiefseeschlaaa des Sudatlantischen Ozeans. Annin, natur. Mus. Wien, 71, 117-198. KLAVENESS D. 1972a: Coecolithus huxleyi (Lohaann) Kaaptner I. Morphological investigations in the vegetative cell and the process of coccolith formation. Protistologica, 8, 335-346. KLAVENESS D. 1972b: Coccolitbas haxleyi (Loha.) Kaaptn. II. The flagellate cell, aberrant cell types, vegetative propagation and life cycles. Br. phycol. J., 7, 309-318. KLAVENESS D. 1973: The aicroanatoay of Calyptrosphaera sphaeroidea, with soae suppleaentary observations on the aotile stage of Coccolitbus pelagicus. Norw. J. Bot., 20, 151-162. KLAVENESS D. 1976: Eiiliania haxleyi (Lohaann) Hay St Mohler III. Mineral deposition and the origin of the aatrix during coccolith foraation. Protistologica, 12, 217-224.

255 KLAVENESS D. k PAASCHE E. 1979: Physiology of coccolithophorids. In, LEVANDOWSKY H. k HUNTER S .H . ( e d s .) : "Biochemistry and physiology of protozoa*, 2nd.Edn. V o l.l. Academic Press (London), p .191-213. KNUTTEL 5. 1986: Calcareous nannofossil biostratigraphy of the Central East Pacific Rise, DSDP Leg 92. IRDSDP, 92, 255-290. LAMBERT B. 1986: La notion d’esp&ce chez la genre Braarudospbaera Deflandre 1947. Hythe et rAalitA. Revue Hicropaldont., 28, 255-264. LANG T.H. 1986: Calcareous nannofossils froi the Nankai Trough and Japan Trench, DSDP Leg 87. IRDSDP, 87, 573-585. LAVINE L.S. k ISENBER6 H.D. 1962: Intracellular calcification in a coccolithophorid. Nature, 196, 78. LEADBEATER B.S.C. 1970: Preliainary observations on differences of scale lorphology at various stages in the life cycle of * Apistoneaa- Syracosphaera' sensu von Stosch. Br. phycol. J, 5, 57-69. LEADBEATER B.S.C. k HORTON C. 1973: Ultrastructural observations on the external morphology of soie members of the Haptophyceae froi the coast of Jugoslavia. Nova Hedwigia, 24, 207-233. LEADBEATER R. A RIDING B.S.C. (Eds.) 1986: 'Biomineralisation in lower plants and animals*. Procs. Bioain. Conf., Birainghaa 1985. Systeaatics Assoc, spec. Publ. 30, Oxford University Press (Oxford), p . 1-401. LECAL J. 1965: Havisolenia n .g . aprili n.sp. Bull. Soc Hist. nat. Toulouse, 100, 422-426. LEF0RT F. 1972: 0'ielques caractAres aorphologiques de deux espAces actuelles de Braarudosphaera (ChrysophycAes, CoccolithophoracAes). Botaniste, 55, 81-93. LEGGETT J.K., A0KI Y. k T0BA T. 1985: Transition froa frontal accretion to underplating in a part of the Nankai Trough accretionary complex off Shikoku (SW Japan) and extensional structures in the lower trench slope. Har. Petrol. Geol., 2, 131-141. LEGGETT J.K. k PLATT J.P. 1984: Structural features of the Hakran fore-arc on Landsat iaagery. In, HAD It NILLIHAN (eds.): 'Harine geology and oceanography of Arabian Sea and coastal Pakistan*, Van Nostrand Reinhold (New York), p.33-43. LEH0TAY0VA R. 1985: Helicospbaera porosa - new species of calcareous nannofossil from Oligocene / Miocene boundary. Zapad Karp. SAr. Paleont., 10, 103-104. LEH0TAY0VA R. A PRIEWALDER H. 1981: Cycloperfolithas, eine neue Nannofossil gattung aus dem Badenien der Zentralen Paratethys. In, PAPP A. (Ed.): 'Chronostratigraphie und Neostratotypen Hiozan H4 Badenien*, p.486-489. LIPPS J.H. 1969: Triquetorbabdulus and similar calcareous nannoplankton. J. Paleont., 3, 1029-1032. L0EBLICH A.R. It TAPPAN H. 1966: Annotated index and bibliography of the Calcareous nannoplankton. Phycologia, 5, 1966. L0EBLICH A.R. It TAPPAN H. 197B: The coccolithophorid genus Calcidiscus Kamptner and its synonyms. J. Paleont. 52, 1390-1393. L0HMANN H. 1902: Die coccolithophoridae, eine Honographie der Coccolithen bildenden Flagellaten, zugleich ein Beitrag zur Kenntnis des Mittelmeerauftreibs. Arch. Protistenk., 1, 89-165. LQHHANN G.P. It CARLSON J.J. 1981: Oceanographic significance of Pacific Late Miocene calcareous nannoplankton. Mar. micropaleont., 6, 553-579. McINTYRE A. It BE A.W.H. 1967: Modern coccolithophores froa the Atlantic Ocean - I. Placoliths and cyrtoliths. Deep-Sea Res., 14, 561-597. McINTYRE A., BE A.W.H. It PREIKSITAS R. 1967: Coccoliths and the Pliocene - Pleistocene boundary. Progr. Qceanogr., 4, 3-25. McINTYRE A., BE A.W.H. It ROCHE M.B. 1970: Modern Pacific coccolithophorids: A palaeontological thermometer. Trans. N.Y. Acad. Sci., 32, 720-731. MANT0N I. 1986: Functional parallels between calcified and uncalcified periplasts. In, LEADBEATER B.S.C. It RIDING R. (Eds.): 'Bioaineralisation in lower plants and animals*. Systematics Assoc, spec. Vol., 30, p .157-172. MANT0N I . , BREMER 6. It OATES K. 1984: N anoplankton from th e 6 a la p a g o s I s la n d s : Hichaelsarsia elegans Gran and Halopappas adriaticas Schiller (coccolithophorids) with special reference to coccoliths and their unaineralized coaponents. Phil. Trans. R. Soc., B305, 183-199. MANT0N I. & LEADBEATER B.S.C. 1974: Fine structural observations on six species of Cbrysochroiulina from wild Danish marine nanoplankton, including a description of C.catpatulifera s p . n ov. and a preliminary survey of the nanoplankton as whole. Biol. Skr. K. danske Vidensk. Selsk., 20/5, 1-26. MANT0N I. It LEEDALE G.F. 1963: Observations on the micro-anatomy of Cristallolithas hyalinus G aarder It

256 Harkali. Arch. Hikrobiol., 47, 115-136. HANTON I. & LEEDALE 6.F. 1969: Observations on the aicroanatoay of Coccolithus pelagicus and Cricosphaera c a r t e r a e , with special reference to the origin and nature of coccoliths and scales. J. aar. biol. Ass. UK, 49, 1-16. HANTON I. It OATES K. 1982: Nanoplankton froa the 6alapagos Islands: Two genera of spectacular coccolithophorids (O p h i a s t e r and Calciopappas ) with special eaphasis on unaineralized periplast coaponents. Procs. R. Soc., B300, 435-465. HANTON I. It OATES K. 1985: Calciosoleniaceae (coccolithophorids) froa the 6alapagos Islands: Unaineralized coaponents and coccolith aorphology in fooplosoleiia and Calciosolesia, with a comparative analysis of equivalents in the unaineralized genus Navisolenia (Haptophyceae = Pryanesiophyceae). Phil. Trans. R. Soc., B309, 461-477. HANTON I. It PARKE H. 1962: Preliainary observations on scales and their node of origin in ChrysochroMuIipa p o l y l e p i s sp. nov. J. aar. biol. Ass. U.K., 42, 565-578. HANTON I . , SUTHERLAND J . It OATES K. 1976: A r c t i c c o c c o lit h o p h o r id s : two s p e c ie s o f 7 urrisphaera gen. nov. froa West Greenland, Alaska and the Northwest Passage. Proc. R. Soc., 8194, 179-194. HANTON I . , SUTHERLAND J . k OATES K. 1977: Arctic coccolithophorids: Jf iguana arctica gen. It sp. nov. froi Greenland and Arctic Canada, H.aitMlifera sp. nov. froa S. Africa and Caiciarc

257 Neogene boundary stratotype*. 6iorn. Seol., 44, 140,142. HUNK W.H. k RILEY 6. 1952: Absorption of nutrients by aquatic plants. J. ear. Res., 11, 215-240. HURRAY 6 . k BLACKHAN V.H. 1898: On the nature of the coccospheres and rhabdospheres. Phil. Trans. R. Soc., B190, 427-441. HUZA J . P . , WISE S.W. k C0VIN6T0N J.H. 1987: Neogene calcareous nannofossils froa DSDP Site 603, lover continental rise vestern North Atlantic; biostratigraphy and correlations with lagnetic and seisaic stratigraphy. IRDSDP, 93/11, 593-616. NA8YHAR0SY A. 1987 H e asu re ae n ts on Love r H io ce n e Spheoolithus p o p u la tio n s . In, STRADNER H ., PERCH-NIELSEN K. (Eds.): Procs. Int. Nannoplankton Ass. Heeting, Vienna 1985. Abh. Geol. Bundesanst., 39, 217-229. N1SHIDA S. 1971: Nannofossils froa Japan IV. Calcareous nannofossils froa the Tonohaaa Group, Shikoku, SW Japan. Trans. Proc. palaeont. Soc. Japan, N.S., 83, 143-161. NISHIDA S. 1979a: Atlas of Pacific nannoplanktons. Nevs Osaka Hicropalaeontol., Spec. Paper, 3, 1-26. NISHIDA S. 1979b: Restudies of the calcareous nannoplankton biostratigraphy of Tonohaaa 6roup, Shikoku, Japan. Bull. Nara Univ. Education, 28, 97-110. NISHIDA S. 19B0: Calcareous nannoplankton biostratigraphy around the Pliocene - Pleistocene boundary in the southern part of Okinava - jiaa, Japan. Chishitsugaku Zasshi, 86, 525-536. NISHIDA S. 1981: Nannoplankton biostratigraphy of offshore Hokurikai and San-in districts, Japan, especially concerned to the stratigraphic leanings of Coccolithus pelagicus and Iraarudosphaera bigelouii . Earth Science, 35, 204-210. NISHIDA S. 1982: Has Coccolithus pelagicus (Hallich) Schiller a cold water species? Bull. Nara Univ. Education, 31, 63-77. NISHIDA S. & K0NDA I. 1974: Planktonic foraminifera and calcareous nannoplankton in bottoi saiples obtained froa off the Kii Peninsula and Shikoku Island, Japan. Bull. Nara Univ. Education, 23, 51-91. NOEL D. 1973: Nannofossiles calcaires de sddiaents jurassiques fineaent laainAs. Bull. Hus. natn. Hist, nat. Paris, Ser.3, 75, 95-155. NORRIS R .E . 1965: L iv in g c e l l s o f Ceratolithus cristatus (Coccolithophorineae). Arch. Protistenk., 108, 19-24. NORRIS R.E. 1971: Extant calcareous nannoplankton froa the Indian Ocean. In, FARINACCI A. (Ed.): Procs. II Planktonic Conf., Roaa, Vol. 2, p.899-910. NORRIS R . E . 19B3: The f a i i l y p o s it io n o f Papposphaera Tangen and P a p p o a o v a s Hanton k O ates (Pryanesiophyceae) vith records froa the Indian Ocean. Phycologia, 22, 161-169. NORRIS R.E. 1984: Indian Ocean nannoplankton I. Rhabdosphaeraceae (Pryanesiophyceae) vith a review of extant taxa. J.Phycol., 20, 27-41. NORRIS R.E. 1985: Indian Ocean nannoplankton II. Holococcoli thophor ids (Calyptrosphaeraceae, Pryanesiophyceae) vith a reviev of extant taxa. J.Phycol., 21, 619-641.

0KADA H. k BUKRY D. 1980: Suppleaentary aodification and introduction of code nuabers to the low latitude coccolith biostratigraphic zonation (Bukry 1973, 1975). Har. Hicropaleont., 5, 321-325. 0KADA H. k H0NJ0 S. 1975: Distribution of coccolithophores in aarginal seas along the Western Pacific Ocean and in the Red Sea. Mar. Biol., 31, 271-285. 0KADA H. !t HcINTYRE A. 1977: Modern coccolithophoridae froa the Pacific and North Atlantic Oceans. Micropaleont., 23, 1-55. 0UTKA D.E. & WILLIAMS D.C. 1971: Sequential coccolith aorphogenesis in Hyaenoaonas carterae. J. Protozool., 18, 285-297. PAASCHE E. 1968: Biology and Physiology of Coccolithophorids. A. Rev. Microbiol., 22, 71-86. PAASCHE E. fc KLAVENESS D. 1970: A physiological coaparison of coccolith foraing and naked cells of Coccolithus huxleyi. Arch. Mikrobiol., 73, 143-152. PARKE M. 1961: Soae reaarks concerning the class Chrysophyceae. Br. phycol. B ull., 2, 47-55. PARKE M. 1971: The production of calcareous eleaents by benthic algae belonging to the class Haptophyceae (Chrysophyceae). In, FARINACCI A. (Ed.) Procs. II Int. Plankton. Conf., Roaa 1970, 2, 929-937. PARKE M. k ADAMS I. I960: The aotile (Crystallolithus hyaliaus Gaarder L Harkali) and non-aotile phases in the life history of Coccolithus pelagicus (Wallich) Schiller. J. aar. biol. Ass. UK, 39, 263-274. PARKE H. & HANTON I. 1962: Studies on aarine : VI Chrysochroauliaa prisgsheiaii sp. nov. J. aar. biol. Ass. UK, 42, 391-404. PERCH-NIELSEN K. 1971: Elektronenaikroskopische untersuchungen an Coccolithen und Verwand^ten Foraen aus dea

258 Eozan von Danemark. Biol. Skr. K. Danske Vidensk. Selsk., 18, 1-76. PERCH-HIELSEN K. 1972: Remarks on late Cretaceous to Pleistocene coccoliths froa the North Atlantic. IRDSDP., 12, 1003-1070. PERCH-NIELSEN K. 1977: Albian to Pleistocene calcareous nannofossils froa the western South Atlantic, DSDP Leg 39. IRDSDP 39, 699-825. PERCH-NIELSEN K. 1984: Validation of new coabinations. Newsl. Int. Nannoplankton. Ass, 6, 42-43. PERCH-NIELSEN K. 1985a: M e so zo ic c a lc a r e o u s n a n n o f o s s ils . In , BOLLI H .H ., SAUNDERS J . k PERCH-NIELSEN K. (eds.): 'Plankton Stratigraphy". Caabridge Univ. Press (Cambridge), p.329-426. PERCH-NIELSEN K. 1985b: Cenozoic calcareous nannofossils. In, BOLLI H.H., SAUNDERS J. k PERCH-NIELSEN K. (eds.): "Plankton Stratigraphy*. Caabridge Univ. Press (Caabridge), p.427-554. PLATT J.P. k LEG6ETT J.K. 1986: Stratal extension in thrust footwalls, Makran accretionary prisa; iaplications for thrust tectonics. Bull Aa. Ass. Petrol. Geol., 70, 191-204. PLATT J.P., LE66ETT J.K., YOUNG J.R., RAZA H, & ALAH S. 1985: Large-scale sediaent underplating in the Makran accretionary prisa, southwest Pakistan. Geology, 13, 507-511. PRINS B. 1971: Speculations on relations, evolution and stratigraphical distribution of discoasters. In, FARINACCI A (Ed.), Procs. II Plkt. Conf., Roaa 1970, 2, 1017-1037. PRINS B. 1979: Notes on nannology - 1. Clausicoccus a new genus of fossil coccolithophorids. Newsl. Int. Nannoplankton Ass., 1, N2-4. PR0T0-DECIMA F., MEDIZZA F. k T0DESC0 L. 1978: South Eastern Atlantic Leg 40 calcareous nannofossils. IRDSDP 40, 571-635. PUJ0S-LAMY A. 1973: Repartition bathymAtrique des foraminiffcres benthiques pronds du Golfe de Gascogne. Coaparison avec d’autres aires oceaniques. Revta. esp. Micropalaeont., 5, 213-234. PUJ0S-LAMY A 1977. Eiiliania e t Sephyrocapsa (nannoplancton calcaire): Bioaetrie et int£ret biostratigraphique dans le Pleistocene supGrieur Marin des Azores. Revta. esp. Micropalaeont., 9, 69-84. PUJ0S A. 1985a: Quaternary nannofossils frees the Goban Spur, eastern North Atlantic Ocean DSDP Holes 548-549A. IRDSDP, 80/11, 797-792. PUJOS A. 1985b: Nannofossils from Quaternary deposits in the high productivity area of the central E q u a t o r ia l P a c i f i c , DSDP Leg 85. IRDSDP 85, 553-578. PUJOS A. 1985c: Cenozoic nannofossils from the central equatorial Pacific, DSDP Leg 85. IRDSDP 85, 5B1-607. PUJOS A. 1987: Late Eocene to Pleistocene medium-sized and small-sized "Reticulofenestrids". In, STRADNER H. & PERCH-NIELSEN K. (Eds.): Procs. Int. Nannoplankton Ass. Meeting, Vienna 1985. Abh. geol. Bundesanst., 39, 239-277. QUENNELL A.M. 1959: Tectonics of the Dead Sea Rift. In, Procs. Congresso geologico internacional XX session - Mexico 1956, p.385-405. RADE J. 1975: Scyphosphaera evolutionary trends with special reference to eastern Australia. Micropaleontology, 21, 151-464. RADE J. 1977: Tertiary biostratigraphic zonation based on calcareous nannoplankton in Eastern Australian nearshore basins. Micropalaeontology, 23, 270-296. RADRIZZANI C.P. k VALLERI G. 1977: New data on calcareous nannofossils from the Pliocene of the Tyrrhenian Basin, Site 132, DSDP Leg 13. Riv. ita l. Palaeont. Stratigr., 83, 897-924. RAFFI I k RIO D. 1979: Calcareous nannofossil zonation of DSDP Site 132 - Leg 13 (Tyrrhenian Sea - Uestcin Mediterranean). Riv. Ital. Palaeont. Stratigr., 85, 127-172. RAFFI I. k RIO D. 1981: Coccolithas pelagicus (Wallich): a palaeoteaperature indicator in the late Pliocene Mediterranean deep sea record. In, WEZEL E.C. (ed.), "Sedimentary basins of mediterranean margins’ , C.H.R. Italian project of oceanography, Bologna, p.187-190. RAFFI I. k SPROVIERI R. 1985: Calcareous nannoplankton in DSDP Hole 125, Leg 13 (Ionian Sea - Eastern Mediterranean): a review. Boll. Soc. Pal. Ital., 23, 53-73. RAYM0NT J.E.6. 1980: "Plankton and productivity in the oceans". 2nd. Edn., Pergamon, (New York), p .1-489. REID F.M.H. 1980: Coccolithophorids from the North Pacific central gyre with notes on their vertical and seasonal distribution. Micropalaeontology, 26, 151-176. RIO D. 1982: The fossil distribution of coccolithophore genus Sepbyrocapsa Kamptner and related Pliocene - P le is t o c e n e c h r o n o s tra t ig r a p h ic p rob lem s. IRDSDP 68, 325-346. R0MEIN A.J.T. 1979: Lineages in early Palaeogene calcareous nannoplankton. Utrecht. Micropalaeont. Bull., 22, 1-231.

259 ROTH P.H. 1970: Oligocene calcareous nannoplankton biostratigraphy. Eclog. geol. Helv., 63, 799-881. ROTH P.H. 1973: Calcareous nannofossils - Leg 17, DSDP. IRDSDP 17, 695-795. ROTH P.H. 1974: Calcareous nannofossils from the north west Indian Ocean, Leg 24 DSDP. IRDSDP 24, 969-994. ROTH P.H., FRANZ H.E. & WISE S.W. 1971: Morphological study of selected members of the genus Sphe»olithn<, Deflandre (incertae sedis, Tertiary). In, FARINACCI A (Ed.), Procs. II PIkt. Conf., Roma 1970, 2, 1099-1119. ROTH P.H., HULLIH M.H. & BERGER W.H. 1975: Coccolith sedimentation by fecal pellets: laboratory experiments and field observations. Bull. geol. Soc. Aa., 86, 1079-1084. ROTH P.H. & THIERSTEIN H.R. 1972: Calcareous nannoplankton: Leg XIV of the Deep Sea Drilling Project. IRDSDP 14, 421-487. R0WS0N J.D., LEADBEATER B.S.C. h 6REEN J.C. 1986: Calciua carbonate deposition in the aotile (Crystallolithus) p h a s e of Coccolithus pelagic us (Pryanesiophyceae). Br. phycol. J., 21, 359-370. RYAN W.B.F., CITA H.B., RAWS0N H.D., BURCKLE L.H. i SAIT0 T. 1974: A paleoaagnetic assignaent of Neogene Stage boundaries and the developaent of isochronous datua planes between the Mediterranean, the Pacific, and the Indian Oceans, in order to investigate the response of the world ocean to the Mediterranean "Salinity Crisis". Riv. ital. Palaeont. Stratigr., 80, 631-688. SAMTLEBEN C. 1979: Pliocene to Pleistocene coccolith asseablages froa the Sierra Leone Rise, Site 366, Leg 41. IRDSDP 41, 913-9311. SAMTLEBEN 1980: Die evolution der Coccolithophoriden Sattung Sephyrocapsa Nach Befunden ia Atlantik. Palaont. Z., 54, 91-127. SCHILLER J. 1930: Coccolithinae. In, RABENHORST L. (ed.), "Kryptogamen-Flora", vol.10, pt.2, Akadeaisches Verlag, (Leipzig), p.89-267. SHEARMAN D.J. 1976: The geological evolution of Southern Iran. Geogrl. J., 142, 393-410. SIEBURT J.M. 1979: Haptophytes and Prasinophytes. In, SIEBURT J.M. "Sea Microbes". Oxford University Press (Oxford, New York), p.211-224. SIMPSON E.S.W., SCHLICK R. et al 1974: Initial Reports of the Deep Sea Drilling Project, Volume XXV, Washington (U.S. Governaent Printing Office). IRDSDP 25. SMAYDA T. 1970: The suspension and sinking of phytoplankton in the sea. Oceanogr. aar. biol. A. Rev., 8, 353-414. SMAYDA T. 1980: Phytoplankton species succession. In, MORRIS I. (ed.): "The physiological ecology of phytoplankton". Univ. of California Press (Berkeley) p.493-570,. STEINMETZ J. & STRADNER H. 1984: Cenozoic calcareous nannofossils froa DSDP Leg 75, Southeast Atlantic Ocean. IRDSDP, 75, 671-753. ST0NELEY R. 1974: Evolution of the continental margins bounding a former southern Tethys. In, BURKE C.A. L DRAKE C.L. (eds.): "Geology of continental margins", Springer Verlag (New York), p.889-903. STRADNER H. 1959: First report on the discoasters of the Tertiary of Austria and their stratigraphic use. Proc. 5th World Petroleuu Congress, New York, 1959, 1, 1081-1095. STRADNER H. 1973: Catalogue of calcareous nannofossils froa the sediaents of Neogene age in the Eastern Atlantic and Mediterranean Sea. IRDSDP, 13, 1137-1201. STRADNER H. & ALLRAH F. 1981: The nannofossil assemblages of DSDP Leg 66, Middle America Trench. IRDSDP 66, 589-639. STRADNER H. It EDWARDS A.R. 1968: Electron microscopic studies on Upper Eocene coccoliths from the Oauiaru diatoaite, New Zealand. Jb. geol. Bundesanst. Wien, Sonderband 13, 1-66. STRADNER H. fc FUCHS R. 1980: Uber nannoplanktonvorkommen ia Sarmatien (Ober Miozan) der Zentralen Paratethys in Niederostereich und ia Burgenland. Beitr. Palaont. Ost., 7, 251-279. STRADNER H. & PAPP A. 1961: Tertiare discoasteriden aus Osterreich und deren Stratigraphische bedeutung. Jb. geol. Bundesanst, Wien, Spec. No. 7, 1-160. TAKAYAHA T. 1973: On the distribution of calcareous nannoplankton in the youngest Cenozoic of Japan. Geol. soc. Japan, Memoir 8, 45-63. TAN S.H. 1927: Discoasteridae incertae sedis. Proc. Sect. Sci. K. Akad. Wet., Amsterdam, 30, 411-419. TAN S.H. 1930: Discoasteridae, Coccolithinae and Radiolaria. Leid. Geol. Meded, 5, 92-95. TAPPAN H. 1980: "The paleobiology of plant protists". W.H. Freeman (San Francisco) p.1-1028. TANGEN K. 1972: Papposphaera lepida gen. nov, n.sp., a new marine coccolithophorid from Norwegian waters. Norw. J. Bot., 19, 171-178. TANGEN K., BRAND L.E., BLACKWELDER P.L. & GUILLARD R.R.L. 1982: Jboracosphaera h e u n (Lohmann) Kamptner is

260 a dinophyte: Observations on its aorphology and life cycle. Nar. Hicropalaeont. 7, 193-212. TAYLOR F.J.R. 1980: Basic biological features of phytoplankton cells. In, MORRIS I. (Ed.): 'The physiological ecology of phytoplankton*. Univ. California Press (Berkeley), p.3-55. THE0D0RIDIS S.A. 1983: On the legitisacy of the generic naae Discoaster Tan, 1927 ex Tan 1931. Int. Nannoplankton Ass. Newsl., 5, 15-21. THEODDRIDIS 5. 1984: Calcareous nannofossil biostratigraphy of the Hiocene and revision of the helicoliths and discoasters. Utrecht Hicropaleont. Bull., 32, 1-271. THIERSTEIN H.R. 1974: Calcareous nannoplankton Leg 26 DSDP. IRDSDP 26, 616-689. TOTTEN D.K. 1976: Coccolitbas pelagicasi a clarification. In, FRITSCHE A.E., BEST H., WORNANBT W.W. (eds.), 'The Neogene Syiposiua, SEPH Pacific Section', San Francisco, p.155-160. TONE K.H. 1979: Variation and systeiatics in the calcareous nannofossils of the genus Spheaolithus. Aa. Zool., 19, 55-572. Van HECK S.E. & PRINS B. 1987: A refined nannoplankton zonation for the Banian of the Central North Sea. In, STRADNER H. it PERCH-NIELSEN K. (Eds.): Procs. Int. Nannoplankton Ass. fleeting, Vienna 1985. Abh. geol. Bundesanst., 39, 285-303. VAROL 0. 1982: Calcareous nannofossils froa the Antalya Basin, Turkey. Ns. Jb. Geol. Palaont., Mb#, 244-256. VAROL 0. 1984: New Neogene calcareous nannofossil taxa froa Malta and southern Turkey. Ns. Jb. Geol. Palaont., Hh., 375-384. VAROL 0. 1985: Hiocene calcareous nannofossils froa the Hut Basin, southern Turkey. J. Hicropalaeont., 4, 127-139. VREDENBURG E. 1301: A geological sketch of the Baluchistan desert and part of eastern Persia. Mea. geol. Surv. India, 31, 179-302. Be VRINB-de JONG E.W., BORMAN A.H., THIERRY R., WESTBROEK F., GRUTFR M. & KAHERLING J.P. 1986: Calcification in the coccolithophorids £•iliania haxleyi and Plearochrysis carterae II. B i oche aica l as pect s. In, LEABBEATER B.S.C. & RIDING R. (Eds.): "Bioaineralization in lover plants and animals", p.204-220. WALLICH G.C. 1877: Observations on the coccosphere. Ann. Hag. nat. Hist., Ser.4, 19, 342-350. WALLICH G.C. 1861: Reaarks on soae novel phases of organic life, and on the boring powers of ainute annelids, at great depths in the sea. Ann. Hag. nat. Hist., Ser.3, 8, 52-58. WALSBY A.F. & REYN0LBS C.S. 1980: Sinking and floating. In, MORRIS I. (ed.): 'The physiological ecology of phytoplankton'. Univ. of California Press, (Berkeley), p.371-412. WATA8E N. & WILBUR K.H. 1966: Effects of teaperature in growth, calcification and coccolith fora in Coccolithus haxleyi (Coccolithineae). Hanoi. Oceanogr., 11, 567-575. WELLS A.J.1969: The crush zone of the Iranian Zagros Mountains, and itsiaplications. Geol. Hag., 106, 38 5-394. WESTBROEK P., de JONG E.W., van der HAL P., BORMAN A.H., de VRINB J.P.H., K0K D., de BRUIJN W.C., St PARKER S.B. 1384: Hechanis# of calcification in the aarine alga Etiliania baxleyi. Phil. Trans. R. Soc., B304, 435-444. WESTBROEK P., van der WAL P., van EMBURG P.R., de VRINB-de JONG E.W., l de BRUIJN W.C. 1986: Calcification in the coccolithophorids fiiliania baxleyi and Plearochrysis carterae I. Ultrastructural aspects. In, LEABBEATER B.S.C. & RIDING R. (Eds.): "Bioaineralization in lower plants and aniaals*, p . 189-203. WHITE R.S. 1973: Deforaation of the Hakran continental aargin. In, FARAH A. & BeJONG K.A. (eds.): "Geodynaaics of Pakistan8, Geol. Surv. Pakistan, p.235-305. WHITE R.S. 1982: Beforaation of the Hakran accretionary sediment prisst in the Gulf of Oaan (north-west Indian Ocean). In, LEGGETT J.K.: "Trench fore-arc geology', Spec. Publ. geol. Soc. Lond., 10, p .357-372. WHITE R.S. in press: Active and passive plate boundaries around the Gulf of Oaan, north-west Indian Ocean. WHITE R.S. St KLITG0RB K.B. 1976: Sediaent deforaation and plate tectonics in the Gulf of Oaan. Earth Planet. Sci. Lett., 32, 199-209. WHITE R.S. St L0UBEN K.E. 1982: The Hakran continental margin: Structure of a thickly sediaented convergent plate boundary. In, WATKINS J.S. St BRAKE C.L. (eds.): "Studies in continental aarine geology". Mea. Aa. Ass. Petrol. Geol., 34, p.499-518. WHITE R.S. & ROSS B.A. 1973: Tectonics of the Western Gulf of Oaan. J. geophys. Res., BB4, 3479-3489. WHITHARSH R.B., WESER O.E., ROSS B.A. et al 1974: Initial Reports of the Beep Sea Drilling Project, Voluae XXIII, Washington (U.S. Governaent Printing Office). IRBSBP 23.

261 WILBUR K.M. I UATABE N. 1962: Experimental studies on calcification in lolluscs and the alga Coccolithas huxleyi. Ann. N.Y. Acad. Sci., 103, 82-112. WINTER A., REISS Z., It LUZ B. 1978: Distribution of living coccolithophore assemblages in the Gulf of Eilat (’Aqaba). Mar. Hicropaleont4, 197-223. WINTER A., STOCKWELL D. & HARGRAVES P.E. 19B6: Tintinnid agglutination of coccoliths: A selective or random process? Har. Hicropaleont., 10, 375-379. WISE S.W. 1973: Calcareous nannofossils from cores recovered during leg 18 DSDP: Biostratigraphy and observations on diagenesis. IRDSDP 18, 569-616. WISE S.W. 1983: Hesozoic and Cenozoic calcareous nannofossils recovered by DSDP leg 71 in the Falkland Plateau, SW Atlantic Ocean. IRDSDP, 71/11, 481-551. WISE S.W. & WIND F.H. 1977: Mesozoic and Cenozoic calcareous nannofossils recovered by DSDP Leg 36 drilling on the Falkland Plateau, Southwest Atlantic section of the Southern Ocean. IRDSDP, 36, 269-493. YENTSCH C.S. 1980: Light attenuation and phytoplankton photosynthesis. In, MORRIS I. (Ed.): "The physiological ecology of phytoplankton*. Univ. California Press (Berkeley), p.193-233. YOUNG J.R. 1987: Possible functional interpretations of coccolith morphology. In, STRADNER H. It PERCH-NIELSEN K. (Eds.): Procs. Int. Nannoplankton Ass. Meeting, Vienna 1985. Abh. geol. Bundesanst., 39, 305-313. ZOBEL 1974: Quaternary and Neogene foraminifera: Biostratigraphy. IRDSDP, 25, 573-578.

262 INITIAL REPORTS OP THE DEEP SEA DRILLING PROJECT Relevant chapters in the 'Blue Books' fora the aost important single body of information on Neogene nannofossils; this list gives bibliographic inforaation on thea in a aore convenient foraat than in the aain reference list. LEG SITES DATE Co-chief scientists AREA Papers on Neogene nannofossils*. 1 1-7 1969 E W I N G H. k W0RZEL J.L 6.of Mexico B U K R Y k BRAMLETTE, 369-387 P. 6 4 4 - 6 0 1971 F I S H E R A . 8. C.Pac. BUKRY D.,965-1004, P. 7 6 1 - 6 7 1971 WINTERER E.L. C.Pac. MARTINI E. It HORSLEY T. 1471-1511, PT. 8 6 8 - 7 5 1971 TRACEY J.I. C.Pac. HAQ B.U. It LIPPS J.H., 777-789, P. 9 76-84 1972 HAYS J.D. C.Pac. BUKRY D. 817-832. 12 111-119 1972 LAUGHTON A.S. It BER68REN U.A. N.Atl. PERCH-NIELSEN K., 1003-1070, P. 13 1 2 0 - 1 3 3 1973 RYAN U.B.F. k H S U R.J. Medit. STRADNER H., 1137-1201, PT. 14 135-144 1972 HAYES D.E. k P I H M A.C. C.Atl. ROTH P.H. It THIERSTEIN H.R, 421-487, P. IS 146-154 1973 ED6AR N.T. k SAUNDERS J.B. Caribbean HAY W.W. It BEAUDRY F.H., 625-684, PT. BUKRY D., 685-703, Z. 16 155-163 1973 VAN ANDEL T.H. k HEATH 6.R. C.Pac. BUKRY D., 653-711, PTZ. 17 164-171 1973 WINTERER E.L. k EWING J.I. C.Pac. ROTH P.H., 695-795, PT. 20 19 4- 2 0 2 1973 HEEZEN B.D. k HacGREGOR I.D. C.Pac. HEKEL H., 221-247, P.: BUKRY D., 307-317, PT. 22 2 1 1 - 2 1 8 1974 Von der B0RCH C.C It SCLATER J.8. N.E.Ind.Oc. GARTNER S., 577-601. 23 219-230 1974 WHITHARSH R.B. WESER O.E. ROSS D.A. I.0./Red Sea BOUDREAUX J.E., 1073-1090. 24 231-23B 1974 FISHER R.L. It BUNCE E.T. N.W.Ind.Oc. ROTH P.H., 969-994. 25 239-249 1974 SIMPSON E.S.W. It SCHLICH R. W. Ind.Oc. MULLER C., 579-635, PT. 26 2 5 0 - 2 5 8 1974 DAVIES T.A. It LUYENDYK B.P. S.Ind.Dc. THIERSTEIN H.R., 616-689. 27 259-263 1974 VEEVERS J.J k HEIRTZLER J.R. S.E.Ind.Oc. PROTO-DECIMA F., 589-623, P. 28 2 6 4 - 2 7 5 1975 HAYES D.E. It FRAKES L.A. Ind/Ant.Oc. BURNS D.A., 589-59B. 29 276-284 1974 KENNETT J.P. It H0UTZ R.E. S.W.Pac. EDWARDS A.R. k PERCH-NIELSEN K., 469-539, PT. 32 303-313 1975 LARSON R.I. It H0BERLY R. N.W.Pac. BUKRY D., 777-701, PTZ. 33 314-318 1976 SCHLAN6ER S.0. It JACKSON E.D. C.Pac. MARTINI E., 383-425, PT.: BUKRY D., 493-501, PT, 35 322-325 1976 HOLLISTER C.D. It CRADDOCK C. Ant.Oc. HAB B.U.,557-568, PT. 36 3 2 6-33 1 1977 BARKER P.F. k DALZIEL I.W.D. S.Atl. WISE S.W It HIND F.H., 269-493, P. 38 336-352 1976 TALWANI M. It UDINTSEV 6. N.Atl. MULLER C., 823-841, P. 39 3 5 3 - 3 6 0 1977 SUPKO P.R. It PERCH-NIELSEN K. C.Atl. PERCH-NIELSEN K., 699-825, PT. 40' 3 6 1 - 3 6 5 1978 BOLLI H.N. It RYAN W.B.F. S.E.Atl. PROTO-DECIMA F. et al., 571-635, P. 41 3 6 6 - 3 7 0 1978 L A N C E L O T Y. k S E I B O L D E. E.Atl. FUTTERER D., 709-739, PT. 1979 SAMTLEBEN C . , s u p p l e s t 913-931, PT. 42A 371-378 1978 HSU K. It MONTADERT L. Medit. MULLER C., 727-751. 42B 3 7 9-38 1 1978 ROSS D.A. 1. NEPROCHEV Y.P. Bl ack Sea PERCIVAL S.R., 773-781. 43 382-387 1979 TUCHOLKE B.E. It VOGT P.R. N.N.Atl. OKADA H. St THIERSTEIN H.R., 507-573. 45 3 9 5 - 3 9 6 1978 MELSON W.G. It RUBINOWITZ P.D. C.Atl. BUKRY D., 307-317, PT. 47 3 9 7 - 3 9 8 1979 Von RAD U. It RYAN W.B.F. E.C.Atl. CEPEK P. It WIND F.H., 289-315, P. 4 8 3 9 9 - 4 0 6 1979 MONTADERT L. It ROBERTS D.8. N.Atl. MULLER C., 589-637, P. 51-3 417-418 1979 DONNELLY T. k F R A N C H E T E A U J. C a r i b b e a n SIESSER W.G., 823-845 P. 56-7 434-441 19B0 Von HUENE R. It NASU N J a pan Tr. HAQ B.U. It GOREAU H., 867-873, P. 59 4 4 7-45 1 1981 KROENKE L. It SCOTT R. C.Pac. MARTINI E., 547-565, PT. 60 4 5 2-46 1 1982 HUSS0N8 DM. It UYEDA S. C.Pac. ELLIS C.H., 507-535, Z. 63 4 6 7 - 4 7 3 1981 YEATS R.C. It HAQ B.U. E.C.Pac. BUKRY D., 445-471, PTZ. 66 486-493 1981 WATKINS J.C. St MOORE J.C. E.C.Pac. STRADNER H. It ALLRAM F., 589-639, PT. 67 494-501 1982 AUBOUIN J It Von HUENE R. E.C.Pac. MUZYLOV N., 383-399, P. 68 502-503 1982 PRELL W.L. It GARDNER J.V. E.C.Pac. RIO D., 325-346, PT. 71 5 1 1 - 5 1 4 1983 LUDWIG W.J. It KRASHENIKOV V.A. S.Atl. WISE S.W., II, 484-551, PT. 74 525-529 1984 MOORE J.C. k RABINOWITZ P.D S.E.Atl. JIANG M.J. It GARTNER S., 475-500. 75 530-532 1984 HAY W.W. k SIBUET J.C. S. E.Atl. STEINMETZ J. It STRADNER H., 671-753, P. 78A 541-543 1984 BIJU-DUVAL B. it MOORE J.C. C a r i b b e a n BERGEN J.A., 411-447, P. 80 548-551 1985 De GRACIANSKY P.C. It POAG C.W. N. E.Atl. PUJOS A., II, 767-792, PT. 81 552-555 1984 ROBERTS D,G k SCHNITKER D. N.Atl. BACKHAN J., 403-428. 82 556-564 1985 BOUGAULT H. It CANDE S,C. N.Atl. PARKER M.E., 559-590.: BUKRY D., 591-602, PT. 84 5 6 5 - 5 7 0 1985 Von HUENE R. k A U B O U I N J. E. C.Pac. FILEWICZ M.V., 339-361, PT. 85 5 7 1 - 5 7 5 1985 MAYER L. It THAYER F. C.Pac. PUJOS A., 553-579 It 581-607, PT. 86 57 6-58 1 1985 H E A T H 6. R k BURCKLE L.H. N.W.Pac. MONECHI S., 301-336. 90 586-594 1986 KENNETT J.P. k Von der BORCH C.C. S.W.Pac. MARTINI E., 747-761, PT. 92 597-602 1986 LEINEN M. k REA D.K. E.C.Pac. KNUTTEL S., 255-290, PT. 9 3 6 0 3 - 6 0 5 1987 Van HINTE J.E. It WISE S.N. N.W.Atl. MUZA J.P. et al, 593-616, P.

*. Codes after papers: P - plates; T - new taxonoay; Z - new zonation scheae. N.B. These only refer to the Neogene.

263 APPENDIX 1 - NANNOFOSSIL DISTRIBUTION TABLES See Chapter 8 (p-95-108) for discussion, and further explanation. ABBREVIATIONS & SYMBOLS

PRINSIACEAE SPHENOLITHACEAE Geph lg Gepbyrocapsa la r g e S.abies Spbeaolithas abies Geph s» Gephyrocapsa s a a l 1 S.belei S.beleaaos P , i . l a c Pseadoeailiaaia lacaaosa v a r . lacaaosa S.ciper S.ciperoensis P.I.ovt P.l.ovata S.delph S.delpbix R.doroa Reticalofeaestra dorooicoides S.heter S.heteroaorphosas R.p.tia R.pseadoatbilica v a r . i iaata S.aorif S.aorifortis R.p.hag R.p.hagii R.p.pab R.p.pseadoatbilica DISCOASTERACEAE R.p.rot R.p.rotaria N.U.D 6 Six rayed discoasters, unidentified C.fiord Cyclicargolitbas florid an as D.defl. D.deflandrei D.sciss Dictyococcites scissor a d.drugg l.draggii D.kaglr d.kagleri COCCOLITHACEAE D.bolli D.bollii C.pelag Coccolitbas pelagicas D.exils D . e x i l i s C.p.anp C.p.naanopelagicas D . c f ex D.cf exilis C.p.tio C .p.iiapelagicas D.cf br D.cf broaueri C.p.pel C.p.pelagicas D.b.bru D.b.broaueri Cl.fens Ciaasicoccas feaestratas D.b.asa D.b.asyaaetricus Cl.pral C.pritalis D.b.taa D.b.taaalis Cy.carl Cycloperfolithas carlae D.b.tri D.b.triradiatas S.petra Solidopoas petrae D.calc. D.calcaris C.l.lpt Calcidiscas leptoporas v a r . leptoporas D.alias D.alias C.l.pat C.l.pataecas D.trist D.tristellifer C.l.tac C.l .aacintyrei D.v.var D.variabilis v a r . variabilis C.l.ptc C.l.preaaciatyrei D.v.dec D.v.decorus C.aitsc Coroaocyclus aitesceas D.v.Ibl D.v.loeblichii G .r o t u l 6eainilitbella rotula D.v.paa D.v.paasas l l . j a f a r (labilicospbaera jafari HUD S Five rayed discoasters, U .J a f .A U.jafari v a r . A D.bell. D.bellus U.sibog (l.sibogae D.baaat D.batatas H.prplx Hayaster perplexas D.p'pnt D.prepeataradiatas O.fragl Oolithotas fragiI is D.pent. D.peataradiatas D.g.gag D.gaiagaeratas v a r . gaingaeratas HELICOSPHAERACEAE D.g.brg D.g.berggretii H.cart. Relicosphaera carteri C.calyc Catiaaster ealyculas H.c.crt H.c.carteri C.coalt C.coal Ha s H.c.gra H.c.graaalata "C.iexic' C.texicanus H.c.hyl R.c.hyaliaa H.c.sel H.c.sellii CERATOLITHACEAE fc TRIOUETORHABDULACEAE H.c.ulc H.c.uallicbii A.area Aagalolithiaa area H.i.iat H.i.iateraedia A.p.prt Ataarolithas pritas v a r . pritas H.i.eup H.i.eapbratis A.p.atp A.p.atplificas H.iaver H .ia v e r s a A.trie A.tricoraicalatas R.cot pc H.coapacta A.t.trc A.t.tricornicalatus H.oblig H.obligaa (s.l.) A.t.del A.t.delicatus H.o.ori H.orieatalis var. oriental is C.araat Ceratolithas artatas H.o.pac H.o.pacifica C.atlat C.atlaaticas H. recta H.recta C.crist C.cristatas H.stals R.stalls H. eonvl Aiaylithita cotvallis O. serrt Orthorhabdas serratas PQR1QSPRAERACEAE I. cari). Irigaetorhabdalas caritatas Poatsph Poatospbaera, undifferentiated I.cball I.challengeri Scy. sp Scyphospbaera, unidentified l.tilou I.tilouii S.apsta. S.apsteiiii 7 .ragos J.ragosas S.globl S.globulata l.r.rag J.r.ragosas S.in ter S.iateraedia l.r.str l.r.striatas S.lagea S.lagena S.polch S.pulcherriaa DINOPHYTA, THORACOSPHAERACEAE (Only recorded where abundant). MISCELLANEOUS Ihoracs Jhoracospbaera, undifferentiated R.clavg Rbabdospbaera clavigera f . b e i n ' I. heiaii Scaphol Scapbolithas, undifferentiated Syracs. Syracosphaera, undifferentiated ABUNDANCE SYMBOLS I.syteo letralithoides syaeoadesii (percentages are of total asseiblage) B.bigel Braaradospbaera bigelouii S - Doain an t >25Z F - Few 0.2-1X A - Abundant 5-25Z i - Rare 0.2-.04Z C - Coaion 1-5Z * * Very rare <.04X NANNOFOSSIL DISTRIBUTION, SITE 219 (ARABIAN SEA).

SAMPLE 14-3 14-2 13-5 13 -1 12-6 12-3 11-6 11-2 10-6 10-2 9-2 8-3 7-4 7-2 6-6 6-3 6-1 5-2 4-5 4-1 3-3 2-3 1-4 1-3 1-1 REF NO 1 2 3 4 5 6 7 8 9 10 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 ZONE 3/4 4 4 4 111 111 lit 111 111 111 12 13+ 13+ 15 16 16 16 17 19 19 19 20 20 20 21

6 e p h si C F 8 8 8 8! S e p b Ig C 8 A 8 P . l . l a c c A CFF F P . l . p a c c AAF F P . d o r o i 8 A 8 F F F C k . p . i i i AAA A AA 8 0 AAFFAA A R . p . h a q F A C F CCF 8 A C AA R . p . p i b F i F F I ft 1 F i 1 C ft R . p . r o t C i I C . AA f i o r d A C i i D . s c i s• s

C . p . n p CF C . p . p e l CFF FF F F 1 ft ft ft 0 ft 1 CF C l . p r i l FF FF F f i C f . e a r l 0 i 1 1 I ft 1 S. pe t r a ft 1 , F C . l . l p t c c F C CF F F FFC F F F c F C F C F F F i C. C . l . p a t c c F F 1 1 1 F 1 FFF C . l . t a c F C CF F F C FF F F F F F F 1 1 C . B i t s c 1 F i S . r o t e l 1 FF ■ F , i 1 F F C c FF F. U . j a f a r A c A A F 1 1 F F c t l . j a U CFF F 1l . s i b o g FF FFC C. H . p e r p l IF C C 1 ft I ft 1 l O . f r a g l F C.

H . c . c r t FF F FCF c C F c c c C F F C F 1 1 F c! H.c.gri c c c C H . c . h f l 1 F F H . c . s e l 1 F 1 F 1 1 F H. c .h Ic ft 1 1 F F. H. i . e a p 1 F i H . i . i i t ' ' H . i i v e r F i. H . o b l i g F F I F

P o i t s p h 1 * l l 1 1 » o F I I 1 Scjf s p . 1 S . a p s t i » S . g l o b l 1 1 S . ptilch ■ •

B . b i g e l ft R . c l a v g , ft ft 1 1 0 , i. S c a p b o l 1 ft ft S j r a c s . , 1 1 1 F. T . s p i e o ft

S . b e l e i F S . b e te r c A A A S . i o r i f ACC C S . a b i e s AAA A A c AC c c c A t c b . d e f l C 1 1 D . c f . d f AC A B . b . b r u F 1 F F FF F F 1 F c F F c F 1 1. B . b . a s i ft 1 ( » ft ft l B . b . t a i l D.b.tri ft I ft ' B . c a l c . 0 l • B . s a r c . i i 1 1 l t I F ft I 1 B . v . v a r I 1 l 1 ft 1 F B . v . d e c 1 B . v . l b l 1 B . e . p a i ft B . p e i t . i i F F F F i c F C c c C c B . g . g i g c FC C c C c l . q . b r g 1 ft F F

P . a r c a ft P . p . p n ft 1 P . p . a i p ft ft ll.t . l r c ft A . t . d e l ft C . a n a t 1 C . cr i s t « 1 1 « 1 ft 0 ft. l .r a g o s 1 ft ft ft

SAMPLE 14-3 14-2 13-5 13 ■1 12-6 12-3 11-6 11-2 10-6 10-2 9-2 8-3 7-4 7-2 6-6 6-3 6-1 5-2 4-5 4-1 3-3 2-3 1-4 1-3 1-1

2 6 4 A NANNOFOSSIL DISTRIBUTION SITE 223 (ARABIAN SEA)

SAMPLE 27-2 27-1 25-2 23-2 22-2 20-3 20-2 19-3 18-5 16-6 16-5 13-3 10-2 7 - 2 6 - 4 5 - 5 5 - 2 4 - 6 4 - 3 3-1 1-2 R E F NO. 1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 23 ASE 4/5 4/5 5 5 6 / 7 6 / 7 8 9 9 10 10 ell 11 111 15 16 16 17 17 19 2 0

Geph Ig i D . Geph st F 0 P.l.liC i FF 0 i R.l.oet C 0 F i R.dor on D DSA F R.p.tin F C F SD C A R.p.haq SD A D A A D SAA F AAA R.p.ptb 0 i A D A AA A 0 i i F C.fiord D F A A 1 0 0 Ji.sciss i

C.p.tio 0 F C.p.pel i F F C F C F F CCC C F FFF 0 Cl.prtl F i F 1 Cy.carl F 0 C.l.lpt l C i F 0 i i 0 F i C i F F F . C.l.pat l F C.l.tac. 1 F i F 1 C F F FFFFF F FF F i i F i . C.l.pnc l i G.rotal 1 1 C 0 1 1 F i il.jafar F FF D 1 1 0 A il.jaf.fi F 1 U.sibog C F . H.perpI 1 l

H . c . c r t 1 1 0 F 1 1 F l l 1 i C F . H . c . g r n 1 F 1 0 0 H.c.sel 0 H.i.int

Pontsph i 0 1 Scy spp 0

R.clavg Syracs. 1 0

S.abies A C A c AA S.heter FF F c S.torif F FF A 1 F

HUD. 6 1 F 1 C 1 FF 0 D.defl FFF D.boll F 1 D.exils F C F 1 l 1 0 D.cf br F I 1 0 B.b.brn 1 F C C C 1 l i F 1 C D.b.ast i 0 0 D.b.tan i 0 D . c a J c . 1 i D.surc. 1 i 0 D.v.var , 0 F i 0 0 G.v.dec F 0 0 i 0 M U D - 5 F F 1 0 D.bell. C D.baiat 1 0 D.p'pnt 1 i 0 G.pent. 1 F l F i l i 'G.q.qnq 1 F C C . c o a l . 1 F

C.crist 0 0 J.tilou I T.ragos 0

SAMPLE 2 7 - 2 2 7 -1 25-2 23-2 22-2 20-3 20-2 19-3 18 - 5 1 6 - 6 16 - 5 13-3 10-2 7-2 6-4 5-5 5-2 4 - 6 4 - 3 3-1 1-2

2 6 5 NANNOFOSSIL DISTRIBUTION SITE 225 (RED SEA) SITE 227 (RED SEA)

SAMPLE 25.2 22.3 20.2 17.2 14.1 13.2 12.2 R E F N O 41 42 43 44 45 46 47 ZONE ?12 13+ 13+ 13+ 13+ 13+ 15

R.p.iin C A A A C A A . R.p.baq i C C C C A F . R.p.pib 1 1 F l .

C.pelag 1 * 0 I C.l.lpt • 1 C.l.tac 1 + 1 F . S.rottil 1 U.jafar 1 C A F H.perpl •

H.c.crt F F F C I C F . H.inter 1 1

Pantspb o 1 l F F . S c y sp. o F l . S.apstn 1 o I I . S.globI 0 IDO. S.pulch 0

R.clavg 0 1 E.bigel

S.abies D A A A D 0 0 '

D.b.brv i t i i . V.b.asi i D.surc. 0 d.pent. + F l 1 C .

A.t.trc 0 C.crist • 1

1 .beiii 1 1 1 1 1 F . Ihoracs 1 1 1 1 F .

SAMPLE 25.2 22.3 20.2 17.2 14.1 13.2 12.2

2 6 6 NANNOFOSSIL DISTRIBUTION, SITE 231 (GULF OF ADEN)

SAMPLES in cm in cnj rom — incNicsicom uJ*3rp c^rpCN<^rr^incr>rocr3rocorocNjm<>Ju3 » i i i i i • i uo uo in in in in in5^! ^in ^ co cnj ^ co ^ ^ in n o cor-r»-.u3inco—«<=>

REF No. 1 2 3 4 5 G 7 8 9 10 U 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 AGE S 7 7 ?7 ?7 ?8 9 10 10 10 10 10 ell ell 111 111 12 12 13* 1 3 M 3 M 3 + 15 15 16 16 16 16 16 17 17 17 18 19 19

B e p h l g A. B e p h s i A i C F A A. P . l . l a c F F A F F C C P . l . o x t F F A A C C R . d o r o t A fi 9 B fi fi C C R . p . i i i E D fi S A A fi fi 0 fi 0 A R . p . h a g A A A f C A C A C A fi A A A A A R . F p . p iA b A S A A A A A F C C . f i o r d

C . p . i i o F F F C . p . p e l C F C C C C F C C C F C I F F A C C l . p n l C / . c a r l S . p e t r a C . l . l p t l F F F F F F F F F C . l . i a c F C C C F F I I B . r o t a l I F F F F ll.j a f a r C C A A C C U . s i b o g C C H . p e r p l

H . c a r t . I I F F F F F F l l F F I F I F F F I I F F F F F l l F F l C I F C H . i . i i t I I • F R . o . o r i R . ob l i g P o i t s p h ®oio Scy s p . S . ap s t i S . g l o b l S . l a g e i S . p a l c b

R . c o i v l • * F F R . c l a v g S f r a c s . S . a b i e s A A A C A A A A A C C A C C S . i o r i f c c c c c C A A A

MUD. 6 C i i i B . d e f l . B . k a g l r B . e x i l s B . c f e x i i i B . c f b r iiii C F B . b . b r u C F B . b . a s i B . b . t a i D . b. t r i B . c a l c . B . s a r c . I I F I B . v . v a r B . v . d e c till B . v . p a i M UD-5 B . b e l l . B . b a i a t B . p ’ p i t B . p e i t . Fill I C C F F B . g . g i g o i B . q . b r g C. coaJ. R . a r c a C . c r i s t 7 . r u g o s

REF No. 1 2 3 4 5 6 7 3 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

267 NANNOFOSSIL DISTRIBUTION, SITE 242 (MADAGASCAR CHANNEL)

SAMPLES 9-4 9-1 -5 8-4 8-3 8-2 8-1 7-5 7-4 7-4 7-3 7-3 7-2 7-1 6-CC 6"3 6-3 6-3 6-2 5-4 5-2 4-6 4-3 3-5 3-4 3-3 3-1 2-4 2-4 2-2 2-1 2-1 1-4 REF NO. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 ZONE NP25 1 2 2 2 34 6 6 6 6 ?6 7 7 9 10 ell ell ell 111 111 12 13+ 13+ 15 15 15 16 17 17 17 18 21

S e p h Ig S e p h si c c P . l . l a c A A A P . l . o e t A A R . d o r o n A A R . p . i i i F C C C A A C A C A C A fi A C A R . p . h a q C 0 A A A A A C C A A C R . p . p i b l l F I l F R . p . r o t R . d a e i s C . f i o r d D D A A A

C . p . n p F F C . p . i i o C . p . p e l F F F F F c c C F c c C F F F F F F C l . f a s t C l . p n l C C C C C c c c c I C C F C C F C j . c a r l I I ■ I I S . pe t r a C . l . l p t F F F F F C C C C F C C F F F F F F A C C . l . p a t C . l . i a c F F F I F F F I F F C C C C F C F C . i i t s c F F F I F F S . r o t a l C C F F C F F F C C F F l C F C F F F U . j a f a r A A A A A A C C A C C C F F C C F A A A A O . j a f J C C C F F U . s i b o g f). p e r pi 0 . fr ag l

H . c . c r t F F F C F C F C F F C C C C F F F F F F I F F C H . c . g n F F F F F F F F F F R . c . h f l F l F I I F F I F R . c . s e l l F H.c.hIc I H.i.etip I I F H . i . i i t F F i l I ■ i i H . i e d i t F F H . c o i p c o o H . o . o r i « F F C F F I H . o . p a c H . o b l i q H. recta

P o n t s p h F F F S e p sp. I S . a p s t i S . g l o b l S. in t e r S . l a g e r S . p n l c h

R . cl a n g F C c c c c S c a p h o l F F F F S f r a c s . c c c c 1 . s y i e o

S . b e l e i S . c i p e r S . d e l p h S . b e t e r S . i o r i f A A A C C C C C C C C S . a b i e s C A A A A A A C C C C

V . a d a i . F F B . d e f l . C A C C A ti.drtigg B . h t g l r D. be l l i B. exils i . e f ex C C C V . c f br F I C l . b . b r a C C C C F c c c c S . b. a s i i > I l . b . t a i D . b . t r i l D . c a l c . l I F F I b . s a r c . I 1 D . a l t u s D.v.var F F F C F t . v . d e c D . v . l b l I o D . v . p a n I).bell, l . b a i a t b . p ’p i t b . p e n t . F F F F C C C F D . q . q n q F C C b . q . b r g F C . calj/c C.coal.

A.area d . p . p r i A . p . a i p P . t . t r c A . t . d e l C . a n a t C . a t l a t C . cr i s t 0 0 0 0 0 0 K . c o a v . F F F D . s e r r t L e a n I I I F T. chall I o I l . i i l o H T . r . r u g Fife lilt I . r . s t r

SAMPLES 9-4 9-1 8-5 8-4 8-3 8-2 8-1 7-5 7-4 7-4 7-3 7-3 7-2 7-1 6_cc 5~3 6-3 6‘3 6'2 5‘4 5'2 4-6 4-3 3*5 3'4 3'3 3-1 2'4 2"4 2‘2 2-1 2-1 **4 26 8 SITE 214 SITES 239, 238 AND 241 SITE 249 SITE 251A (ARABIAN SEA) SITE 258 (WEST CENTRAL INDIAN OCEAN) (SW INDIAN OCEAN) (SW INDIAN OCEAN) (SE INDIAN OCEAN)

SAMPLE 6-3 6-2 SAMPLE 8.4 5.2 4.4 11.5 11.2 7.5 7.1 SAMPLE 16.1 13.3 12.6 4.5 4.1 SAMPLES 26-2 16-2 15-2 15-1 13CC 269 13-5 13-4 13-2 10CC 10-6 10-5 4-5 4-2 SAMPLE 8.2 8.1 A4CC 6.1 REF NO 41 42 45 46 47 42 43 49 50 R E F NO. 3 7 3 6 38 41 4 3 REF. NOS 44 45 46 47 REF NO. 48 49 50 51 52 53 54 55 56 REF No 57 58 59 60 ZONE 15 15 ZONE 9 ell 111 15 15 15 16 ZONE 7 10 10 12 12 ZONES 4/5 9 9/10 9/10 10 ell ell ell 111 111 12 15 15 R . p . i i i B A AA ZONE 12 12 12 16 , R . p . h a q ACC AC P . l . l t c S e p h si F F R . p . i i i A F fi A A A C C C D R . p . p i b A 1 1 F F fi . R . d o r o i A . R . p . i i i C C P . l . l a c FF C . R . p . h a q AA CCC A A A fi D fi A A . R . p . i i i D D fi R . p . h a q A A P . l . o v t C . C . p . i i o , R . p . p i b fi A A A 1 C C C C C C F F . R . p . p i b f C R . p . h a q C C C R . d o r o i D . C . p . p e l C C C FF C . f i o r d C R . p . p i b F F C R . p . i i i C C C C C C C l . p r i l 1 F I C . p e l a g R . p . h a q AA A AA A C f . c a r l I FF 1 ft C . p . p e l A CA A C A A A C F C F F . C . p e l a g i F F F . C . l . i p t f C R . p . p i b C 1 ft F • F ft . C . l . l p t c c c C F C l . p r i l F i F F 1 C l . p r i l i C . l . i a c F C R . p . r o t C C f . c a r l F FF 1 • C . I . p a t F C . l . l p t c c c C . S . r a t a l C i F C . l . l p t FF C C c C A C F C C C C . C . l . i a c c 1 F | F C . I . p a t F F F . ll.jafar 1 C . p e l a g C C c F . C . I . p a t F F . C . l . p i c 1 C . I . i a e F 1 • F . C l . p n l ft F C . l . i a c ft F , CF F F l 1 ft o S . r a t a l 1 F FFF S . r o t a l F F F 1 . H . c a r t . C F C f . e a r l ft 1 1 C . l . p i c 1 ll.jafar c F A F C ll.jafar C F C 1 . C . l . i p t F AA F . G . r o t u l F , FF c c H . p e r p l ft F ■ 1 F 1 1 F . H . p e r p l ft P o a t s p h I C . l . i a c 1 F F F F F . ll.jafar c F 1 ft ' 1 * ft S e p s p . 1 S . r o t u l ft 1 F i 1 1 . H . c . c r t c C F , H . p e r p l ft H . c a r t . F F F F . S . g l o b l I « ll.jafar A F FF H . c . g r i F H . i i t e r ft S . l a g e i 1 H . i i t e r ft 1 ft H.c.crt ft Fill 0 1 S . p a l c h H . c . c r t , C F F , F . H . c . g r i 1 ' c P o i t s p h 1 ft • ft H . c . g r i F F P o i t s p h ft 1 ft H . c . u l c • S c y . s p Scy s p . ft ft S p r a c s . , H . c . u l c ft 1 . ft P o i t s p h ft S . a p s t i ft ft S . a p s t i ft ft S . a b i e s C F H . i n t e r 1 i S . a p s t i t S . g l o b l ft S . g l o b l ft H . s t a l s ft S . g l o b l ft S. it ter B . b . b n C C ft R . c l a v g F l S . l a g e i ft B . b . a s i 1 F P a i t s p h F 1 R . c l a v g 1 S c a p h o l F• ft S . p a l c h ft ft ft B . b . t r i 1 Scy sp. 1 1 . s j i e o ft S j r a c s . & B . s a r c . ■ c S . a p s t i i 1 i 1 I . s f i e o ft R . c l a v g l F l F . B . v . v a r F 1 S . g l o b l S . a b i e s CCC C • 1 S c a p h o l 1 B . v . d e c « S. li t e r ft ft S . i o r i f F S . h e ter F S f r a c s . ft B.peit. C F S . l a g e n 1 i ft S . i o r i f F B . c f ex F 1 B. b i g e l ft ft S . p a l c h ft ft 1 1 S . a b i e s 1 1 F ft 1 C . c r i s t 1 1 B . c f br FF ft B . b . b r u CC 1 1 S . a b i e s F F C R . c l a n g C « B . b . t r i ft B . c f d f C 1 SAMPLE 6-3 6-2 B . b o l l i 0 B . c a l c . 1 HUD. 6 ft 1 S . a b i e s AA C C F B . c f e x ft 1 B . s a r c . ft 0 B . b . b r u ft ft | 1 . S . i o r i f C B . v . v a r F B . c f b r F 1 1 1 F F F 1 B . b . a s i ft B . v . d e c ft ft B . b . b r u F « 1 F F F F l F 1 B . b . t a i ft B . d e f l . 1 1 B . v . p a i I ft B . b . a s i 0 ft 1 B . s a r c . 1 * 1 . B . e x i l s c B . b e l l . ft 1 B . b . t a i « ft B . h a i a t B . v . v a r ft ft 6 B . c f br F F 1 B . b . t r i 1 1 ft 1 1 ft 0 B . p ’p i t ft ft B . v . d e c 1 ft B . b . b r u 1 FC A C C A . B.calc. ft ft ft B . p e i t . ft F 1 1 MUD-5 ft ft ft B . b . a s i i 1 B.sarc. | ft 1 • ft B . q . q i q F F B . b e l l . ft ft B . b . t r i 1 F B.v.var i FF 1 F c c 1 1 1 ft B . h a i a t ft B . s a r c . 1 F F F F 1 . B . v . d e c » It.area ft 1 1 0 B . p e i t . ft ft ft ft B.v.var F C F F A . p . a i p ft B . v . I b l ft ft B . v . d e c ft I F F 1 . R . t . t r c ft B . v . p a i ft ft A . p . p r i ft ft 1 H . c o i v l AC B . b e l l . ft ft ft B.v.pai 1 A . t . d e l ft F l . r a g o s ft ft B . h a i a t 0 B.bell. » A . t . t r c ft ft D.haiat 1 B . p ’p i t ft ft B . p e i t . 1 F AA F C . SAMPLE 16.1 3.3 12.6 4.5 4.1 B . p e i t . ft 1 ft 1 1 1 1 1 SAMPLE 8.2 8.1 A4CC 6.1 B . q . q a q F C B . q . q i q 1 1 ft ft B . q . b r g 1 1 B . q . b r g 1 ft 0 C . c a l p c F ft C . ca l f c ft ft C . c o a l . 1 C . c o a l t 1 ' C . i e x i c ' 1 A . a r e a ft A . a r e a 1 . A . t . d e l ft ft A . t . d e l ft H . c o i v l A A c A c c i T . r a g o s A . t . t r c 1 1 1 1 1 1 0 ft ft C . c r i s t i i 1 1 . I . r . s t r ft K.convl ft l . r a g o s « • ' SAMPLES 26-2 16-2 15-2 15- 1 13CC 13-5 13-4 13-2 10CC 10-6 10-5 4-5 4-2

SAMPLE 3 .4 5.2 4 .4 11.5 11.2 7.5 7.1 APPENDIX 2 - COMPUTER PROGRAMS

INTRODUCTION The thesis was produced on a BBC Master 128 microcomputer. Vordvise Plus, and Spellraaster, were used for the text. ViewSheet was used to create datafiles for: the distribution charts (Appendix 1, and Fig.42); coccolith size variation diagrams (Figs.20-22,41); and plots of biometric data (Figs.9,13-14,44). The graphs from this data were produced from short BASIC programs. The drawings of coccoliths and discoasters (Figs.10,12-13,39,43,47,51-53,55), were produced by two BASIC programs. The discoasters (and cross-section views of coccoliths) were drawn with a geometrical drawing program (Listing 1), detailed notes are given on this program since it is a useful introduction to the general technique. The coccolith drawings were produced from a greatly modified version of this program (Listing 2), the notes on this program cover general points only, and are intended for people with experience of BBC BASIC. The theoretical basis for this program is discussed in Chapter 4. For both programs the primary data is the shape of a single ray, (which is then geometrically reproduced in a variety of positions and orientations, at variable scale). This data was produced manually; by drawing a single element on graph paper and recording the cartesian co-ordinates of selected points around its outline. An alternative technique would be to use a digitising tablet.

DISCOASTER DRAWING PROGRAM: LINE NOTES 10-250 PROGRAM 20. Asks if print out is wanted. If so diagram needs to be slightly stretched to compensate for the distortion produced by the printer. 30. Clears the screen, sets graphics mode, and draws border. 50. Sets up array co-ords for DATA. 60-200. Main loop, repeated for each constituent sub-diagram. 70. READ from DATA the origin (OX%, 07%), scale (P%), orientation (ang), no of rays (R%), and name (name$), of sub-diagram. Set graphics origin for sub-diagram. 80-190. Sub-loop drawing each element of sub-diagram, (e.g. ray outlines, or central knob). 90. READ from DATA the number of steps (S%), and PLOT mode (K%), of element. 100-120. READ co-ordinates of each step from DATA into array (co-ords). N.B. BOC$ is simply a separator in the data statements (usually £).

270 130-180 Plotting loop, repeated for each ray. A is the angle of the ray.

140 Hove (PLOT mode 4 ) to beginning of element. 150 Draw first half-ray of element. 160-170 Draw second half-ray of element (ommited for asymmetrical ray elements). 210. Print screen, if required. NB SDUMP is a screen dumping program held on disc. 230-250 Procedure, PROCdr, called from the program, for drawing from current point to the next. The co-ordinates of the new point are calculated for the current ray angle (A), from the co-ordinates for the first ray (p,g); this uses the general equation for rotation of cartesian co-ordinates. A line is then drawn from the current point to the new point, in PLOT mode K% (53 dotted line, 5 solid line, 4 no line). 260-360 DATA 260 Basic DATA for first sub-diagram (READ by line 70). 270 DATA for first element of sub-diagram (READ by lines 90 and 110). The first two figures are the number of points (S%) and the PLOT mode (K%). Subsequent numbers are the x and y co-ordinates of each point, separated by the filler BOG$. BOG$ is also used to note what each data set draws, and to indicate the end of sub-diagrams [STOP), and of the entire diagram (END).

COCCOLITH DRAWING PROGRAM, GENERAL NOTES. The listing given is simplified, omitting procedures for varying element form around the coccolith, printing, saving screens, drawing side views, only drawing part of the coccolith etc. The remaining procedures are concerned with: determining the positions and orientations of elements around the proto-coccolith ring (genangle); recalculating their co-ordinates for this position and orientation (recalc); drawing the elements (draw, plot); and preventing overlapping parts of elements from being drawn [condplot, partray, setwipe, wipe).

SELECTED LABELS Al angle from centre to current element start position, ang angular orientation of element. BOG$ String array holding plot commands for corresponding element points in coozds. coords Numeric array for the coordinates of the elements. dist distortion, cancels out printing effect. E ellipticity; of ring of element start positions.

271 G% current element point number. Ll distance between element start positions. offset ray precession. P% scale. S% total number of element points. T% counter. X% x coordinate of start position of element. Y% y coordinate of start position of element.

PROCEDURES genangle: The first step in plotting each element is to define the start angle (ang), and start position (X%,Y%). This is done by an iterative process, working around proto-coccolith ring until the correct distance from the previous point is reached, then the local tangent is calculated.. recalc: Computes the cartesian co-ordinates of the element points for the start angle and position, given by genangle. partray: Procedure controlling condplot, and acting on its results. In conjunction these procedures work through parts of the element which may overlap with the previous element, until the previous element is encountered then plot to it. condplot: Procedure for conditionally plotting between two points (G% and H%). Every graphics pixel from G% to H% is examined (via POINT) until a foreground pixel is found. In addition one pixel above is examined, in order to prevent problems with diagonal lines. plot(Kl,G%): Service procedure to plot in plot mode K% from current position to element point G%. setvipe & wipe: Setwipe draws a simplified version of the first element, to enable condplot to work for subsequent elements. Wipe removes this element later. readcoords: Reads the values in the second DATA statement into the arrays coords and BOG$. Some changes can be made at this stage to the element shape, and position relative to its start position. draw: Works through the element points, carrying out the appropriate plots between them, as determined by the corresponding BOG$ values.

272 LISTING 1 - DISCOASTER DRAWING PROGRAM

10 REM DISCOASTER DRAWING PROGRAM 20 INPUT print$: IF print$="Y" THEN dist=l.193 ELSE dist=l 30 M0DE128: MOVES,5: PLOTS,5,1010: PLOTS,1274,1010: PLOTS,1274,5: PLOTS,5,5 40 50 DINco_ords(25,2) 60 REPEAT 70 READ OXZ,OY7.,PX,anq,RX,nase$: VDU29,0XX;0YX; 80 REPEAT 90 READ Sl , U 100 FOR step = 1 TO Sl 110 READ co_ords(step,1), co_ords(step,2), B0G$ 120 NEXT step' 130 FOR A=2*PI/RX+RAD(ang)TO 2*PI+RAD(ang) STEP 2*PI/RX 140 N=K7.: K7.=4: PR0Cdr(co_ords(l,l),co_ords(l,2)): K7.=N: 150 FOR TX=1 TO Sl: PROCdr(co_ords(TX,1 ),co_ords(TX,2)): NEXT TX 160 IF BOG$=BASYMR GOT0180 ELSE N=KZ: K7.=4: PRQCdr(-cojjrds(SX,l),co_ords(SX,2)):KX=1 170 FOR TX=SX TO 1STEP -1: PROCdr (-co_ords(T7.f 1),co_ords(TZ,2)): NEXT VL 180 NEXT A 130 UNTIL BQ6$=BST0P" OR BOG$="END* 200 UNTIL BOG$="END“ 210 VDU26: IF print$="Ya THEN *RUN SDUMP 220 END 230 DEFPROCdr(p,q) 240 PLOT K7., PX/10*(p*-COS(A)+q*SIN(A)),PX/10*dist*(p*SIN(A)+q*C0S(A)) 250 ENDPROC 260 DATA 600,50,25,0,1, BOLLII SIDE 270 DATA 5,5,12,50,D.BOSS,11,64,£,14,72,£,10,80,£,0,78,£ 280 DATA 3,53,63,33,Prox Ridges,26,28,£,22,16,£ 290 DATA 10,5,0,50,BODY,50,50,£,65,45,£,120,42,£,120,38,£,63,33,£,52,23,£,22,16,£,12,7, £,0,7,STOP 300 DATA 960,600,25,0,6, BOLLII PROX 310 DATA 2,5,6,11,P.KNOB,0,13,£,3,53,1,19,£,3,30,£,3,60,£ 320 DATA 10,5,23,40,BODY,17,43,£,11,52,£,6,70,£,5,100,£,12,110,£,14,120,£,11,120,£,5,113,£,0,111,STOP 330 DATA 320,600,25,0,6, BOLLII DISTAL 340 DATA3,5,13,22,KNOB,10,22,£,0,14,£ 350 DATA 2,53,23,40,SUTURES,13,22,ASYH 360 DATA 10,5,23,40,BODY,17,43,£,11,52,£,6,70,£,5,100,£,12,110,£,14,120,£,11,120,£,5,113,£,0,111,END LISTING 2 - COCCOLITH DRAWING PROGRAM

10 REN SIMPLIFIED VERSION OF COCCOLITH DRAWING PROGRAM 20 MODE128: DIHcoordsZ(25,4): DINB06$(25) 30 FOR p a r t d i ag-1 TO 4 40 A3=PI/2: raycount=0 50 READ OXX,OYX,piotdirn$,naae$,offset,no ofrays,scale,rad,E 60 VDU29,QXZjOYZ;: PZ=scaIe: radius=rad 70 dA=2*PI/no_ofrays: Ll=SQR((l-C0S(dA))A2+SIN(dA)A2) 80 PROCreadcoords: 90 REPEAT 100 PROCgenangle 110 PR O Crecalc 120 IF raycount=0PROCsetwipe: S0T0140 130 PROCdraw 140 raycount=raycount+l: IFraycount=3 PROCwipe 150 UNTIL Al<=-i.5*PI OR A1>=2.5*PI 160 REMdA=2*PI/no_ofrays:A3=PI/2:PROCgenangle: PRDCrecalc : PROCdraw 170 NEXT p a rtd ia g 180 END 190 : 200 DEFPROCgenangle 210 IF plotdirnSO ’ left": A2=A3-dA: x=C05(A2)-C0S(A3): y=E*SIN(A3)-E*SIN(A2) 220 IF p lo t d ir n $ =, l e f f : A2=A3+dA: x= C0S(A3)-C0S(A2): y= E*SIN(A2)-E*SIN(A3) 230 L=SQR(xA2+yA2) 240 IF L>L1 OR L<.95*L1 THEN d A = d A + d A * ((Ll-L )/ L l): GOTO 210 250 A1=.5KA3+A2): A3=A2: ang=ACS(x/L)*-SGN(ASN(y/D) 260 XZ=COS(Al)*radius: YX=SIN(Al)*E*radius*dist 270 ENDPROC 280 : 290 DEFPROCrecalc:A=ang-RAD(offset) :IFplotdirn$=,left"THEN A=ang-RAD(offset) 300 cosA=COS(A): sinA=SIN(A) 310 FOR TZ=0 TO SZ 320 coordsX(TZ,3)=XZ-PZ/10*(-cosA+coordsZ(TZf2)*sinA) 330 coordsZ(TX,4)=YZ+PZ/10*(sinAKoordsZ(TX,2)*co5A) 340 NEXT TZ 350 ENDPROC 360 : 370 DEFPROCpartray: REN with condplot draws froi C to next line, or to EC. 380 start=6I: end=6Z: REPEAT: end=endH: UNTIL B0G$(end)='EC' 390 REPEAT 400 PROCcondplot(GZ,GZ+l): GZ=GZ+1 410 UNTIL found=TRUE OR GZ=end 420 PR0Cplot(4,start): FOR TZ=start+l TO GZ: PROCplot(5,TZ): NEXT TZ 430 GZ=end 440 ENDPROC 450 : 460 DEFPROCcondplot(GZ,HZ) 470 AZ=coordsZ(GZ,3): BZ=coordsZ(GZ,4): CZ=coordsZ(HZ,3): DZ=coordsZ(HZ,4) 480 UZ=SQR((AZ-C7.)A2+(B7.-DX)A2) 490 TZ=-1: found=0 500 REPEAT 510 T7.=TZ+2: 520 EZ=AZ+TZ»(CZ-AZ)/UZ: FZ=BZ+TZ*(DZ-BZ)/UZ 530 IF POINT(EZ,FZ)=1 OR P0INT(EZ+2,FZ)=1 THEN TZ=UZ+1: fou n d= -l 540 UNTIL TZ>=UZ 550 IF found THEN coordsZ(HZ,3)=EZ: coordsZ(HZ,4)=FZ 560 ENDPROC 570 : 580 DEFPR0Cplot(KX,6Z) 590 PL0TKX|CoordsX(6Z,3)l coord5X(6Zl 4) 600 ENDPROC 610 : 620 DEFPROCsetwipe 630 P R O C p lo t( 4 ,1 ) : FOR TZ=1 TO SZ: P R 0 C p lo t( 5 ,T Z ) ; NEXT VU oldang= ang: ENDPROC 640 DEFPROCwipe 650 oldA3=A3: dA=2*PI/no_ofrays: A3=PI/2: PROCgenangle: PROCrecalc 660 P R O C p lo t( 4 ,1 ) : FOR TZ=1 TO SZ: P R O C p lo t( 7 , T Z ) : NEXT TZ 670 A3=oldA3: ENDPROC 680 : 690 DEFPROCreadcoords 700 READ SZ: FOR TZ=1 TO SZ: READ X v a l, Y v a l,B 0 6 $ (T Z ) 710 coordsZ(TZ,l)=Xval*l: coordsZ(TZt2)=Yval: NEXT TZ 720 ENDPROC 730 : 740 DEFPRQCdraw: 750 FOR GZ=1 TO SZ 760 IF B06$(8Z)=“P* THEN P R O C p lo t(5 f GZ) 770 IF B06$(6Z)=’,I1" THEN P R O C p lo t( 4 ,GZ) 780 IF B06$(GZ)="D" THEN P R O C p lo t( 2 1 ,GZ) 790 IF BOGS(GZ)=*C" THEN PROCpartray 800 NEXT 61 810 ENDPROC 820 REH Data for drawing Coronocyclus nitescens (Figure 39/H, p.165). 830 DATA 960,512,right, C.nitescens PRQXinner cycle,10,40,60,200,1 840 DATA 12,-9,6,C,-10,3,EC,0,0,C,-10,2,EC,0,0,M ,l,3,P,-l,4,P,-2,7,P,-4,6,P,-9f6,P,0,0,H,-9,-4,P 850 DATA 960,512,le ft, C.nitescens PROXouter cycle,10,40,40,200,1 860 DATA 7,-14,16,C,0,0,EC,-10,15,C,0,17,EC,-14,16,H,-12,16,P,-10,15,P 870 : 8B0 DATA 320,512,right, C.nitescens DISTinner cycle,10,40,40,200,1 890 DATA 12,0,0,C,-3,4,EC,17,14,C,0,11,EC,17,14,M,18,3,P,0,0,P ,16,13,11,10,11,P ,12,5 ,P ,17,5 ,P ,16,13,P 900 DATA 320,512,right, C.nitescens DISTouter cycle,10,40,40,200,1 910 DATA 7,15,20, C, 14,12,EC, 12,18,C, 0,17,EC, 12,18,N, 13,19,P, 15,20,P

275 PLATE 1 - COCCOLITHACEAE A

No. Identification, notes Saaple Age Negative nuaber 1. Coccolithas pelagicas var.tiopelagicas 231-57-2 NN7 15/21 ph 2. Coccolithas pelagicas a 242-8-1 NN4 8/22 ph, 3. Coccolithas pelagicas var.pelagicas 242-7-1 NN7 5/5 ph 4. Coccoiithfls pelagicus var.pelagicas 231-56-5 NN7 15/28 ph 5. Coccolithas pelagicas var.postas y«r 223-6-4 NN15 13/26 ph 6-7. CJafliicoccus feaestratus 242-9-1 NN1 7/2-3 ph & xpl 8-9. Claasicoccas fenestratus 242-9-1 NN1 21/2-3 ph & xpl 10-11. Clausicoccus fenestratas b 242-9-1 NN1 21/2-3 ph l xpl 12-13. Claasicoccas fesestratus 242-8-2 NN3 4/32-33 ph k xpl 14-15. Clausicoccus fenestratus c 242-8-1 NN4 8/36-37 ph k xpl 16-17. Claasicoccas pritalis 223-27-2 NN5 13/31-32 ph k xpl 18-19. Claasicoccas priaalis 219-12-6 NN11 13/8-9 ph k xpl 20. Setioilithella rotala 242-6-CC NN9 6109 dst x3200 21. 6e*inilithella rotala 242-4-1 NN12 6266 side x4400 22. Uabilicosphaera jafari var. A 242-8-4 NN2 S192 dst x8000 23. Utbilicosphaera jafari var. A 242-8-4 NN2 S188 dst xBBOO 24. Coccolithas pelagicas d 242-6-C NN9 S109 prx x3000 25. Coccoiithas pelagicas 242-6-CC NN9 6218 prx x4300 26. Coccolithas pelagicas e 242-6-CC NN9 647 prx x2300 27. C.p.aannopelagicas uourrv?£ 242-8-5 NN2 S1&9 dst x5000 28. Calcidiscus leptoporus var. leptoporus f Chal. 128 Recent S3 dst x3100 29. Calcidiscus leptoporus var. leptoporas 9 242-5-4 1NN11 S263 dst x3000 30. 1l.sibogae var. sibogae h 242-1-4 NN20 6200 dst x5000 31. Seainilithella rotala 242-6-CC NN9 6221 dst x6300

NOTES - Light aicrographs are all xl500. a. Large C.peiagicas speciaens such as this occur throughout the Early Miocene, I have restricted C.p.iiopelagicas to the even larger specimens that occur in the Middle Miocene. b. Second speciaen froi the saae field of view as 8,9. c. Specimen inter*ediate between typical C.feaestratas and C.priialis, this for* occurs throughout the range of C.fevestratus. o. Note that the outer cycle of the proxieai shield continues beneath the inner cycle. e. Siall elliptical coccoliths are R.pseadouabilica (coapare with plate 3/21). Saall round coccolith is probably ll.jafari, with clay over central area. f. Saaple of Recent ooze, Challenger Site 128 (S.Atlantic). f. Probably C.leptoporus var. A. g. Note siailarity to ll.jafari varJ, including relative shield size. Saall coccolith is Sephyrocapsa ericsonil.

276 PLATE PLATE 2 - COCCOLITHACEAE B

No. Identification, notes Saaple Age Negative nuaber 1,6. Calcidiscas leptoporas varJ a 219-9-2 NN11 12/11-12 ph. 2. Calcidiscus leptoporas var.leptoporas 219-7-4 NN13/15 12/21 ph 3. Calcidiscus leptoporas var.pataecus 219-12-6 NN11 13/7 ph 4. Cycloperfolithus carlae 242-6-CC NN9 12/7 ph 5. Cycloperfolithas carlae b 223-13-3 NN11 14/12 ph 7. Caleidiscus leptoporas var.*acintyrei 242-7-1 NN7 5/5 ph 8,13. Calcidiscas leptoporas var.tacin tyrei 223-16-5 NN10 14/13-14 ph h xpl 9,14. Calcidiscas leptoporas var.preiacintyrei c 251A-26-2 NN5 19/5-6 ph & xpl 10,15. Calcidiscas leptoporas Yar.preaacintyrei c 242-7-4 NN6 21/26-27 ph fc xpl 11. Calcidiscas leptoporas sat.leptoporas d 219-6-3 NN15/16 12/31 xpl 12. Calcidiscas leptoporas var.Mac in tyrei d 219-6-3 NN15/16 12/32 xpl 16y\?,l? Coronocyclus nitescens var. nitesceis e 242-8-4 NN1 21/11-13 xpl 1 ph 17. Hayaster perplexus 219-13-1 NN5 13/35 ph 20-21. Coronocyclus nitescens var. »ite5ceM 242-8-1 NN4 8/39-40 xpl & ph 22-23. Corojiocyclas sitescets var. nitescens 219-13-5 NN5 14/6-7 xpl k ph. 24-25. Cororocyclus nitescens var. ellipticas f 242-7-4 NN6 21/29-30 xpl & ph 26-27. (Jtbilicosphaera sibogae var. foliosa g 219-1-3 NN20 23/19-20 xpl i ph 28-29. Ihbilicosphaera jafari h 231-57-2 NN7 15/22-23 ph & xpl 30. Geiinilithella rotula i 251A-26-2 NN5 19/3 ph 31. Solidopons petrae 219-14-3 NN4+ 15/11 xpl 32-33. Solidoposs petrae 219-13-5 NN5 14/8-9 ph It xpl 34. Coccolithus pelagicus j 219-6-6 NN15 12/30 ph 35. Coccolithus pelagicas var. pelagicus k 231-56-5 NN7 15/28 xpl

NOTES - Light micrographs are all xl5Q0. a. Note radial partitions in the central area. b. Note grill visible in central area. c. Note that only central areas are strongly elliptical; shields are much less elliptical than those of Coccolithas pelagicas, or Cycloperfolithas carlae. d. Isolated proximal shields, very pale in phase contrast. e. Probably proximal view, 16 is low focus. f. Holotype, more strongly elliptical types also occur, relatively narrow ria is typical. g. Note that only the ssaller proxiaal shield is strongly birefringent. h. Speciaen with both shields, shields are usually detached. i. Larger than usual specimen. j. Rare variety of C.pelagicas with open central area, note similarity to S.petrae. k. Typical C.pelagicas (saae speciaen as PI.1/4), note siailarity of extinction cross to that of S.petrae.

277 PLATE 2 PLATE 3 - PRINSIACEAE

No. Identification, notes Saaple Age Negative nuaber

1 . Bictyococcites scissara 242-9-4 NP25 12/2 xpl

2. Reticalofenestra pseadoanbilica v a r . r a t a r i a 219-11-6 NN11 13/12 xpl

3. Reticalofenestra pseadoanbilica v a r . h a q i i 2 42-3-3 NN15 23/8 xpl

4 . Reticalofenestra pseadoanbilica var. aiaata 231-42-2 NN11 15/37 xpl 5. Bephyrocapsa Caribbeanica 223-3-1 NN19 14/26 xpl

6 . Bephyrocapsa oceanica 2 1 9-1 -3 NN20 23/17 xpl

7 -8 . Cyclicargolithas florid anas 242-9-1 NN1 7/6-7 xpl It ph

9 -10 . Victyococcites daviesii 242-8-2 NN3 4/24-25 xpl & ph

11-12. R.p.pseadoanbilica 242-6-CC NN9 22/1-2 xpl It ph

13-14. Reticalofenestra pseadoanbilica a 231-58-2 NN7 15/15-16 xpl & ph

15. Reticalofenestra pseadoanbilica a 231-57-2 NN7 15/24-25 xpl

16-17. Reticalofenestra pseadoanbilica a 242-9-1 NP25 21/2-3 ph It xpl

18-19. Reticalofenestra pseadoanbilica a 219-12-6 NN11 12/37, 13/1 ph & xpl

20. Cyclicargolithas floridanas 242-8-2 NN3 6215 dst x5600

21. Reticalofenestra pseadoanbilica b 242-6-CC NN9 6216 dst x5000

22. Bephyrocapsa oceanica 242-1-4 NN20 6201 p rx l d st x5000

23. Eniliania haxleyi c C h a l.1 2 8 Recent S5 d st x8800

24. Cyclicargolithas floridanas 242-8-5 NN1 6206 prx <3600

25. R.p. pseadoanbilica 242-5-4 1NN11 6264 d st x3000

26. 6ephyrocapsa oceanica c C h a l.1 2 8 Recent S8 dst x6000

27. 76ephyrocapsa oceanica C., d C h a l.128 Recent S2 prx x6400

28. Pseudoeniliania lacunosa var. ovata 242-5-2 1NN11 S182 dst x8000

29. Reticalofenestra pseadoanbilica e 242-4-1 NN13-14 6265 prx x4O0G

30. Bephyrocapsa oceanica c Chal.128 Recent S15 dst x5000

31. IBephyrocapsa oceanica C| dChal.128 Recent S32 prx x6400

NOTES - L ig h t a ic ro g r a p h s are a l l xl5 Q 0 . a. Speciaens of R.pseadoanbilica lacking all or part of their rias. These have been described as a separate genus,Pyrocyclas, b. Saaller speciaens of R.pseadoanbilica froa the saae saaple are shown in plate 1/26. c. Saaple of Recent ooze, Challenger Site 128 (S.Atlantic). d. Proxiaal views of 76.oceanica. fi.oceajn'ca and E . h a x l e y i are the only reticulofenestrid species occurring in this eaterial and so even though the bridge is not visible these speciaens can be reasonably confidently identified as 6.oceanica. e. Speciaen of R.pseadoanbilica with auch of ria aissing, this produces the * P y r o c y c l a s * type speciaens

278 PLATE 3 PLATE 4 - HELICOSPHAERACEAE

No. Identification, notes Sample Age N e g a tiv e number

1. Helicosphaera carteri a 227-17-2 NN13/15 23/21 xpl side

2. Helicosphaera carteri var. granulata 219-13-1 NN5 14/5 xpl

3. Helicosphaera carteri v a r . c a r te r i 2 4 2-6 -3 NN11 6/5 ph

4. He!icosphaera innersa 2 1 9 -1 -3 NN20 23/18 xp l

5 -6 . Helicosphaera carteri v a r . u a l l i c h i i b 2 4 2-1-4 NN21 23/11-12 xpl & ph

7-8 Helicosphaera obiiqua 242-8-1 NN4 21/20-21 xpl It ph.

9-10 Helicosphaera oriemtalis var.oriental is 2 31-58-2 NN7 15/17-18 ph fc xpl

11-12 Helicosphaera intermedia v a r . i n t e r m e d i a d 242-7-3 NN6 4/14-15 ph & xpl

13-14 Helicosphaera intermedia v a r .intermedia 242-9-1 NN1 7/13-14 ph & xpl

15-16. Helicosphaera intermedia c 2 42-8-2 NN3 4/28 ph & x p l.

17. Helicosphaera intermedia y a r . i n t e r m e d i a d 227-20-2 NN13-14 14/31 ph

18. Helicosphaera intermedia v a r . e u p h r a t i s 242-8-1 NN4 7/30 xpl

19-20 Helicosphaera intermedia v a r . i n t e r m e d i a d 227-20-2 NN13-14 14/27-28 xpl & ph

21. Helicosphaera orientalisv a r . oriental is e 242-7-2 NN7 S144 dst x7200

22. Helicosphaera carteri v a r . g r a n a l a t a 242-6-CC NN9 6226 dst x3400

23. Helicosphaera carteri v a r . g r a n a l a t a f 2 42-7-2 NN7 S152 dst x2150

24. Helicosphaera carteri var. carteri 9 C h a l.1 2 8 Recent Sll dst x4000 25. Helicosphaera carteri 242-6-CC NN9 6218 prx x4300

26. Helicosphaera carteri h 242-1-4 NN21 652 prx x4000

27. Helicosphaera carteri 242-6-CC NN9 648 prx x2150

28. Helicosphaera carteri i 2 42-3-4 NN15 6246 prx xl850

NOTES - L ig h t m icro graph s are a l l xlSOO. a. Side view showing the asymmetric profile, and also optical discontinuity of the jmwMil plate. b. Note that in cross-polars the apertures appear larger and less obliquely orientated than they do in phase contrast. This suggests that the basal plate is not showing birefringence; and that the pores in the basal plate do not precisely correspond to those in the blanket. c. Specimen with interaediate aorphology, between H.i.intermedia and H.i.euphratis ; bar is distinctly oblique and does not fill the central area. d. Speciaens froa the Early Pliocene of the Red Sea, H.intermedia has rarely been reported above the Kiddle Miocene. e. Note orientation of pores, towards the flange teraination, as in H.obliqua (and H.inversa), reverse of

ca se in H.intermedia, and H . o b l i q u e . f. Also speciaens of C.l .mac-in tyre i and C.pelagicus (both distal). g. Sample of Recent ooze, Challenger Site 128 (S.Atlantic). h. Etched speciaen, showing details of the proximal plate construction, speciaens in figs 27 and 28 are increasingly well preserved. i. Also proximal view of C.l.macintyrei.

279 PLATE 4 PLATE 5 - PONTOSPHAERACEAE, AND MISCELLANEOUS

No. Identification, notes Sample Age N e g a tiv e number

1 . Pontosphaera sp. 231-46-4 NN11 15/31 xpl

2. IScyphospbaera apsteiaii a R227-20-2 NN13-14 14/32 xpl

3 . Poatosphaera sp. R225-14-3 NN16 15/6 xpl

4 -5 . Ihoracosphaera heiaii R 225-20-2 NN13-14 15/2-3 xpl

6 . Braarudosphaera bigelonii b M1CC NN21 14/35 xpl

7. Scyphosphaera palcherriaa 242-6-3-104 eNNll 22/30 xpl

8-9 Scyphosphaera palcherriaa c 2 4 2-5-4 1NN11 23/13-14 xpl 1 ph

10. Scyphosphaera iateraedia R225-14-3 NN16 15/6 ph

12. Scyphosphaera apsteinii v a r . a p s t e i a i i R 227-13-2 NN13/14 23/26 xpl

13. Scyphosphaera apsteiaii v a r . a p s t e i a i i c R227-13-2 NN13/14 23/25 xpl

14. IScyphosphaera apsteiaii var. apsteinii d R225-19-2 NN13/14 15/4 xpl

15. IScyphosphaera apsteiaii e 2 19-9-2 NN11 12/14 ph

16. Scyphosphaera iateraedia 242-3-4 NN15 19/29 ph

17. Scyphosphaera pulcherriaa 242-3-3 NN15 23/6 xpl

18-19. Scyphosphaera apsteiaii var. recurvata f 242-6-3-104 eNNll 23/35 xpl & ph

20. Scyphosphaera lageaa R225-14-3 NN16 15/10 ph

2 1 ,2 5 . Scyphosphaera palcherriaa g 2 4 2-4 -5 NN12 6272 x2000 & 4000

22. Scyphosphaera apsteiaii 242-3-1 NN15 6241 x 1300

23. Scyphosphaera global ata 242-3-4 NN15 6245 x3000

24. Scapholithas sp. 242-1-4 NN20 655 x6400

26. Paatasphaera sp h C h a l.1 2 8 Recent S4 x3100

27. Ihoracosphaera heiaii 242-1-4 NN20 6202 xl400

28. Rhabdosphaera clavigera Chal.128 Recent SI x3400

NOTES - L ig h t m icrograph s are a l l x l5 0 0 . R Specimens from Red Sea DSDP sites: Pontosphaeraceae, and Thoracosphaeraceae, were unusually common in these samples, a. Basal view of a specimen similar to that in figure 4; correlation of views based on racking focus, and

use of mobile mounts. Taller Scyphosphaera coccoliths have similar basal plates. b. Specimen from piston core of Pleistocene sediments from the Bosphorus, Turkey. c. These specimens show well the bilamellar construction of the wall. d. Low walled variety, inward curvature of the walls is indicative of Scyphosphaera , rather than

Postosphaera. e. Rarely encountered small lopadolith, possibly Scyphospbaera graphica MULLER 1974. f. Note discrete base, this is quite often seen in well preserved specimens.

280 PLATE 5 plate 6 - m m s m defl/ih m i - exilis - m m iu s group

No. Identification, notes Sample Age N e g a tiv e number

1. Discoaster v a r i a b i l i s v a r . v a r i a b i l i s 2 1 9-6 -3 NN16 12/33 b f

2. Discoaster variabilis v a r . d e c o r a s 2 2 3 -5 -5 NN16 14/21 ph

3. Discoaster exilis v a r . e x i l i s a 242-6-CC NN9 4/8 ph

4. Discoaster exilis var. exilis a 242-7-1 NN7 5/2 ph

5 -6 . Discoaster exilis v a r . e x i l i s b 242-7-1 NN7 5/10-11 ph mid-high f.

7. Discoaster d e f l a n d r e i 242-8-2 NN3 4/26 ph

8. Discoaster deflandrei 242-8-1 NN4 8/32 ph

9. Discoaster deflandrei - 5 rayed c 242-8-1 NN4 8/19 ph 11. Discoaster deflandrei 242-8-1 NN4 8/38 ph

12. Discoaster k a g l e r i 242-7-1 NN7 5/16 b f

13. Discoaster variabilis v a r . v a r i a b i l i s 2 4 2-3 -4 NN15 6255 p rx x l7 0 0

14. Discoaster exilis v a r . e x i l i s de 242-6-CC NN9 6217 x 1200

15. Discoaster deflandrei 242-8-2 NN3 6213 x2200

16. Discoaster kagleri 242-7-1 NN7 S91 x3300

17. Discoaster variabilis v a r . p a n s u s 2 4 2-3 -4 NN15 6231 prx x2000

18. Discoaster exilis v a r . e x i l i s e 242-6-CC NN9 S ill prx x2200

19. Discoaster deflandrei 2 4 2-8 -2 NN3 6210 x2200

20. Discoaster kagleri 242-7-1 NN7 S97 x2400

21. Discoaster variabilis v a r . p a n s u s 242-6-3-1104 e N N ll 659 p rx x 1300

22. Discoaster exilis v a r. e x i l i s f 242-7-1 NN7 S38 dst x2400

23. Discoaster exilis v a r. e x i l i s x4 f 242-7-1 NN7 S101 x3000

24. Discoaster deflandrei 242-7-1 NN7 S84 d st x3000

NOTES - A l l l i g h t m icro graph s are xlSOO a. Note distinctly dark bifurcations on all the Discoaster exilis speciiiens, this a result of the ray

tips thinning, and is not usually seen in Discoaster variabilis. b. Speciaen in upper left is probably etched Discoaster exilis, n o te id e n t i c a l c e n tr e s on the

cnor i mane c. This variety is soaetiaes termed Discoaster noorei, coaparison with specimen in fig .8 shows that rays are identical. d. Caccolithus pelagicas specimen at bottoa, etched Discoaster exilis to right of main specimen. e. Specimens of D . e x i l i s showing transitional morphology toward D.variabilis. f. Typical, Middle Miocene, D . e x i l i s specimens.

281 PLATE 6 PLATE 7 - DISC0DS7ER BP. OH HE HI AND SIMCULUS GROUPS

No. Identification, notes Sample Age Negative nuaber

1 -2 . Discoaster haaatus 242-6-CC NN9 21/33-34 high-aid f.

3. Discoaster haaataslbellas 242-6-CC NN9 21/33 ph

4. Discoaster bell as 242-6-CC NN9 21/32 ph

5. Discoaster prepeutaradiatas 2 4 2-6 -3 -1 4 2 NN10 22/26 ph

G. Discoaster exilis a 242-7-1 NN7 5/15 ph

7. Discoaster exilis - 5 rayed a 242-7-1 NN7 5/6 ph

8. Discoaster b r o a u e r i v a r . triradiatas 2 42-6-3 NN11 6/25 ph

9. Discoaster broaueri v a r . b r o a u e r i b 2 19-8-3 NN12 12/18 ph

10. Discoaster variabilis ? v a r . l o e b l i c h i i c 219-B-3 NN12 12/17 ph

11. Discoaster brouuerivar. broaueri 242-6-3-1104 eNNll 6/23 ph

12. Discoaster broaueri var. asynetricus d 2 1 9-9 -2 NN11 12/13 ph

13. Discoaster broaueri v a r . t a i a l i s 2 23-5-5 NN16 14/23 ph

14. Discoaster broaueri v a r . triradiatas 2 19-7-4 NN13-15 12/21 ph

15. D i s c o a s t e r variabilis var. l o e b l i c h i i c 242-5-4 1NN11 11/21 ph

16 Discoaster sarculas 219-12-6 NN11 13/4 ph b f.

17. Discoaster haaatas 242-6-CC NN9 22/0 ph

18. ?Discoaster broaueri e 2 23-5-5 NN16 14/22 ph

19. D i s c o a s t e r c a l c a r i s 242-6-CC NN9 22/7 ph 2 0 ;? 4 . Discoaster surculas - stereo-pair 242-5-4 1NN11 6257-258 prx xl700

21. Discoaster surculus 242-3-4 NN15 6252 dst x4000

22. ?Discoaster broaueri e 242-3-4 NN15 6251 prx x2400

23. Discoaster calcaris 242-6-CC NN9 S75 prx x 1700

25. Discoaster sarcalas f 242-3-4 NN15 6242 prx x2200

26 Discoaster broaueri v a r . b r o a u e r i g 242-3-4 NN15 6254 p rx x2000

27 Discoaster haaatas 242-6-CC NN9 SI 12 p rx x2000

NOTES - A l l l i g h t m icro graph s are x 1500 a. Specimens such as these which reseible Discoaster broaueri are coaaon before the first true occurrence

of the specieSj they don't show the typical ray curvature of Discoaster broaueri, or well formed ray t i p s . b. Relatively robust Discoaster broaueri, c a r e is sometimes needed not to confuse these with Discoaster

a l t a s , but they occur throughout the range of Discoaster broaueri. c. Both these specimens coae from significantly higher than the generally reported range of this variety. d. Typical fora of Discoaster b.asynetricas, although froa earlier than acae. e. Large specimens with truncated ends, Discoaster broaueri, Discoaster v.decoras and Discoaster sarcalus can all give specimens siai1ar to these, and can be d ifficult to separate. f. Speciaens such as this with very sharply reflected ray ends are very si a i1ar to soae D i s c o a s t e r

v.decorus speciaens. R.pseadoaibilica at bottom. g. Typical Discoaster broaueri f o r a . Spbenolithas abies and Helicosphaera carteri below .

282 PLATE 7 PLATE B - MISCELLANEOUS d l S C M S l E R AND C / H M S 1 E R SPECIES

No. Identification, notes Saaple Age Negative nuiber

1 -3 . C a t i n a s t e r c o a l i t u s 242-6-CC NNS 10/30-32 bf low-sid-high f

4 -5 . Catinaster coalitas/calycalas 242-6-CC NN9 10/17-18 bf high-aid f.

6 -7 . Catinaster coalitas/calycalas 242-6-CC NN9 21/35 fc 22/0 ph high-aid f.

8 -9 . C a t in a s t e r coalitas/calycalas 242-6-CC NN9 22/8-9 ph h ig h - a id f .

10. C a t i n a s t e r c o a l i t u s 242-6-CC NN9 22/13 ph side

11. Catinaster coalitas 242-6-CC NNS 22/6 ph side

12. d i s c o a s t e r pentaradiatus a 242-6-3-104 NN11 22/29 ph

13. d i s c o a s t e r pentaradiatus- 4 rayed a 242-3-1 NN15 19/33 ph

14. d i s c o a s t e r pentaradiatus - 3 rayed 219-11-6 NN11 13/17 ph

15. d i s c o a s t e r pentaradiatus - single ray 2 19-7-4 NN13-15 12/20 ph

16. d i s c o a s t e r quinqaeraaus 242-6-3-104 e N N ll 6/3 ph

17-19 discoaster alias 2 4 2-3-3 NN15 23/2-4 ph low-aid-high f

20. discoaster alias 2 4 2-3 -3 NN15 23/7 ph, aid f.

21. d i s c o a s t e r a l t a s 242-3-4 NN15 19/20 ph, a id f .

22. discoaster altas 242-3-4 NN15 G244 p rx x2200

23-24 discoaster alias - stereo-pair b 242-3-4 NN15 6233-4 o b liq u e p rx x 1700

25. d i s c o a s t e r flJtflS b 2 42-3-4 NN15 6248 side/prx x2600

26. d i s c o a s t e r pentaradiatas 242-3-4 NN15 6236 dst xl850

27. d i s c o a s t e r quinqueranus c 242-5-4 1NN11 6256 dst x2000

28. discoaster qainqueranus c 2 42-5-2 1NN11 S181 d st x3500

29. d i s c o a s t e r quinqueranus c 242-6-3-104 e N N ll 660 dst x2600

NOTES - A l l l i g h t a ic ro g r a p h s are x 1500

C a t i n a s t e r s p e c ia e n s Figs. 1 and S are focused on the proxiaal/basal surface. Figs. 3,4 and 8 on dst/top surface. The side views, figs. 10 and 11, are arranged with proxiaal surface downwards. a. These varieties are flat and do not show birefringence, unlike the typical 5-rayed variety. b. Note that although the distal surface is flat (as described by MULLER, 1974a) a long distal knob is present in the centre of it (best seen in Fig.25). c. The variation between these speciaens is mainly preservational, the speciaen in fig .27 is etched, that in fig. 29 slightly overgrown.

283

PLATE 9 - CERATOLITHACEAE AND TRIDUETORHABDULACEAE

No. Identification, notes Sample Age N e g a tiv e number

1. Triquetorhabdulus rugosus v a r . r u g o s u s 242-7-1 NN7 5/1 ph

2. Iriquetorhabdulus rugosus aff.var.striatus a 242-6-3-104 NN11 22/28 ph

3 -4 . Triguetorhabdulus rugosus var. striatus 2 42-5-4 1NN11 23/31-32 ph low-high f.

5. Ceratolithus a t l a n t i c u s b 242-6-CC NN9 17/5 xpl

6 . Angulolitbina a r e a 21 9-1 2 -6 NN11 13/6 ph

7. Iriquetorhabdulus rugosus v a r . r u g o s u s 2 4 2-7-4 NN6 21/28 ph

8. Iriquetorbabdulus rugosus aff.var.striatus a 2 4 2-5 -4 INN 11 23/15 ph

9. Iriquetorbabdulus rugosus v a r . s t r i a t u s 2 4 2 -6 -3 -1 4 2 NN10 23/30 ph

10. fiaaurolithus pritus v a r . p r i a u s 2 1 9-11-2 NN11 13/23 ph

11. Qrthorhabdus serratus c 2 42-7-4 NN6 21/22 ph

12. Triquetorhabdulus carinatus 2 4 2-9-4 NP25 21/5 ph.

13. Iriquetorbabdulus challengeri d 242-8-4 NN1 21/11 ph

14. Hinylithina convallis 242-6-3-142 NN10 22/15 ph p la n view

15. Kinylitbina convallis 2 4 2-6 -3 -1 4 2 NN10 22/14-15 ph s id e view

16. Ceratolithus a n a t u * 2 1 9-8 -3 NN12 12/16 ph

17,22. Orthorhabdus serratus 242-8-1 NN4 21/18-19 ph & x p l.

1 8 ,23 . Iriquetorhabdulus ailouii e 242-8-1 NN4 21/16-17 xpl high-low f.

19,24. Iriquetorbabdulus cballengeri 2 4 2-8-4 NN1 21/12-13 xpl low-high f.

20. Kinylithina convallis 242-6-3-142 NN10 22/16,25 side side view

21. Ceratolithus cristatus 2 19-1-4 NN20 23/29 x p l.

25. flinylithina convallis 2 4 2-6 -3 -1 4 2 NN10 S174 p la n view x4000

28. Ceratolithus cristatus 242-3-4 NN15 6235 upper surface x3200

27. Triquetorhabdulus carinatus f 2 4 2-8 -2 NN2/3 SI 15 x6000

28. Iriquetorbabdulus carinatus 2 42-8-5 NN1 6203 2400

29. Iriquetorhabdulus rugosus v a r . r u g o s u s g 242-6-CC NN9 6223 x2200

NOTES - A l l l i g h t m icro graph s are x 1500 a. These specimens with weakly developed subsidiary ridges are intermediate in morphology, between

l.r.rugosus and I.r.striatus. b. This occurrence of Ceratolithus atlan tic us is highly anomalous, see text. c. Plan view of unusually broad specimen, possibly intermediate to I . r u g o s u s . d. Same specimen as figs. 19 and 24. Round coccolith on the right is i/ibiiicospbaera J a f a r i v a r . A. e. Note in lower fo cu s view ( f i g . 23) th e s l i g h t w ings or e a rs on th e l a t e r a l b la d e s , STRADNER & ALLRAN

(1981) proposed these as the main distinguishing feature of l . a u r i t u s . f. See Plate 10/15 for light micrograph of a similar specimen. Circular coccoliths are Cyclicargolithus

floridanus. g. Note characteristic asymmetric form with broad and pointed ends and curved median blade, the extent to which this asymmetry is developed is variable but the form is constant. Circular coccolith on the

l e f t i s an is o la t e d s h ie ld o f Seninilithella rotula.

284 PLATE 9 PLATE 10 - SPHEHOLITHACEAE

No. Identification, notes Saaple Age Negative nuaber

1 - 2 ,6 . Spheaolithas aoriforais v a r . d i s s i a i l i s 242-8-1 NN4 8/8-10 xpl k ph

3 . Spheaolithas aoriforais 2 4 2-8 -2 NN3 4/27 xp l

4 . Spheaolithas abies a 242-6-3-142 NN1 14/16 xpl

5 . Spheaolithas abies a 227-13-2 NN13/15 23/24 xpl

7. Spheaolithas aoriforais 242-3-1 NN1 7/15 xpl

8 . Spheaolithas aoriforais 2 42-8-2 NN3 4/29 xpl

3. Spheaolithas aoriforais b S/KB/10 E .M io . 12/3 x p l.

10. Spheaolithas heteroaorphosas c 242-8-1 NN4 21/14 xpl

11-13. Spheaolithas beleaaos 242-8-2 NN3 4/18-20 ph k x p l, 0° k 45°

14. Spheaolithas beleaaos 2 42-8-2 NN3 4/31 xpl

15. Iriquetorhabdalas cariDatus d 242-8-2 NN2-3 21/10 xpl 45°

16. Spbeaolithas ciperoeasis e 213-15-1 NP25 23/16 xpl 17. Spheaolithas capricoraatas 242-9-1 NP25 21/1 xpl

18. Spheaolithas delphix 242-9-1 NP25 21/2 xpl

13. Spheaolithas beleaaos 242-8-2 NN2-3 4/31 xpl

20. Spheaolithas beleaaos f 2 4 2-8-2 NN3 SI 13 x4400

21. Spheaolithas aoriforais 242-6-CC NN9 G49 x2150

22. Spheaolithas abies 242-6-3-104 e N N ll S64 x4000

23. Spheaolithas heteroaorphosas 242-8-1 NN4 8/18 xpl 45°

24. Spheaolithas beleaaos 242-8-2 NN2-3 G207 x5200

25. Spheaolithas aoriforais 242-8-2 NN2-3 G211 x3700

26. Spheaolithas aoriforais 2 4 2-8 -2 NN2-3 G203 x3700

27. Spheaolithas aoriforais 2 4 2-8 -2 NN2-3 G208 x3700

NOTES - A l l l i g h t n ic r o g ra p h s are x 1500 In figs. 13,15, and 23 the polarization direction is at 45° to the plate aargins, in all others it is parallel to the plate Kargins. a. Fig.5 shows large specinens with characteristic 5.abies aorphology; the sphenolith in fig .4 a saall

cn*ofi«oc HistinnnichoH ac S.n rather ncn-descript fcrsj J------pnahiet.--- b. Saaple froi the Hakran. c. Both sphenoliths are S.heteroaorphosas speciaens, left hand one has spine coapletely in extinction. d. Long speciaen, fig .3/27 is an Sen of a siailar speciaen, froa the saae saaple. e. Speciaen with well developed bifurcations, note that spine is not in extinction in the vertical p o s i t i o n .

f . S.beleaaos, this speciaen shows reasonably well that the spine is foraed of three laths.

285 PLATE 10 INDEX A - TAXA DESCRIBED IN THE SYTEMATICS SECTION (Chapters 11-17)

Amauroli thus 236 D.b.asyaaetricus 221 Pontosphaera 193 A .p r ia u s 236 D.b.taaalis 221 PRINSIACEAE 170 A.p.aspli ficus 237 D.b.triradiatus 222 Pseudoeailiania 172 A.tricorniculatus 237 D . c a l c a r i s 222 P.lacunosa 172 A.t.bizzarus 237 D.deflandrei 212 P . l . o v a t a 173 A.t.delicatus 238 D.d.druggii 213 "Pyrocyclus" 175 A . t . n in a e 238 D . e x i l i s 214 Reticulofenestra 173 A n g u lo li th in a 243 D.e.petaliforais 214 R.doronicoides 173 A .a re a 243 D.e.subsurculus 214 R.pseudouabilica 174 Calcidiscus 153 D.giganteus 222 R.p.haqii 174 C.leptoporus 153 D .h a a a tu s 223 R .p .a in u t a 174 C . l . A 154 D.kugleri 213 R.p.pseudouabilica 174 C.l.aacintyrei 154 D .B U s icu s 215 R.p.rotaria 175 C.l.pataecus 155 D.neorectus 223 Scyphosphaera 193 C.l.presacintyrei 155 D.pentaradiatus 223 S . a p s t e i n i i 193 C a tin a s te r 227 D.prepentaradiatus 224 S . a . d i l a t a t a 193 C.calyculus 228 D.quinqueraaus 225 S.a.recurvata 194 C . c o a l i t u s 228 D.q.berggrenii 225 S .a .A 194 C.aexicanus 228 D .s u r c u lu s 216 S . ' c y l i n d r i c a " 195 CERATOLITHACEAE 239 D.variabilis 216 S .g lo b u la t a 195 Ceratolithus 239 D.v.decorus 217 S .in t e r a e d ia 195 C .a r a a tu s 239 D.v.loeblichii 218 S.lagena 196 C.atlanticus 239 D .v .p a n s u s 218 S.pulcherriaa 196 C .c r i s ta t u s 240 D.v.pseudovariabilis 218 S . t u b if e r a 197 C.c.separatus 240 E i i l i a n i a 171 S.ventriosa 197 C . c . t e l e s i u s 240 E .h u x le y i 171 S o lid o p o n s 162 Clausicoccus 155 Gesinilithella 160 S .p e tr a e 162 C.fenestratus 156 G .r o t u la 161 SPHENOLITHACEAE 203 C . p r i i a l i s 157 Gephyrocapsa 172 Sphenolithus 203 COCCOLITHACEAE 153 H ayaster 161 S.abies 203 Coccolithus 157 H.perplexus 161 S.beleanos 203 C.pelagicus 157 HELICOSPHAERACEAE 182 S .c a ly c u lu s 203 C .p .a io p e la g ic u s 158 Helicosphaera 182 S.capricornutus 204 C.p.nannopelagicus 158 H . c a r t e r i 182 S.ciperoensis 202 C .p .p o n t u s 158 H.c.granulata 183 S .d e lp h ix 204 Coronocydus 159 H.c.hyalina 183 S.heteroaorphosus 204 C.nitescens 159 H . c . s e l l i i 184 S . a o r i f o r a i s 205 C . n . e l l i p t i c u s 159 H.c.w ailichii 184 S.a.dissiailis 205 Cyclicargolithus 170 H.interaedia 184 TRIQUETORHABDULACEAE 241 C . f lo r id a n u s 170 H.i.euphratis 185 T riq u e to rh a b d u lu s 241 C y c lo p e r f o lit h u s 160 H .in v e r s a 185 T . c h a lle n g e d 241 C . c a r la e 160 H .o b liq u a 185 T .c a r in a t u s 241 Dictyococcites 171 H.orientalis 186 T.finifer 243 D . d a v ie s ii 171 H.o.pacifica 186 T . a i l o w ii 242 D .s c is s u r a 171 H .r e c t a 187 T .ru g o s u s 242 Discoaster 212 Hinylithina 243 T.r.extensus 242 D .a lt u s 219 H.convallis 244 T.r.striatus 243 " D .a s t e r " 226 O o lit h o t u s 162 Uabilicosphaera 162 D.bellus 219 O . f r a g i l i s 162 ll.s ib o g a e 162 D . b o l l i i 214 Orthorhabdus 241 U . s . f o l i o s a 163 D.braarudii 220 O .s e r r a t u s 241 U . j a f a r i 163 D.brouweri 220 P0NT0SPHAERACEAE 193 U . j . A 164

286 INDEX B - Names of taxa (with some non-taxonomic references & synonyms)

a b e le i 196 Calyptrosphaera: d is te n t u s 202 a b ie s 203 23, 39, 59, 146 divaricatus 212 a b is e c t u s 170 caapan u la 196 doronicoides 173 Acanthoica: 36, 147 canescens 196 d ru g g ii 213 a c u ta , H. 182 c a n t h a r e llu s 196 e le g a n s 197 a c u tu s , C. 239 capricornensis 226 e l l i p t i c u s 159 ad aaan teu s 226 capricornutus 204 e lo n g a ta , H. 185 a d r i a t i c u s 147 c a r in a t u s 241 elongatus, S. 203 aequatorialis 1% c a r la e 160 E a i l i a n i a : aequiscutua 159 c a r t e r a e 147 17, 35, 46, 66, 171 albatrosiana 146 c a r t e r i 182 e u p h r a tis 185 a l t u s 219 Catinaster: 227-8 e x i l i s 214 Aaaurolithus: 235-6 centrovallis 154 e x te n su s, S. 213 aaphora 196 CERAT0LITHACEAE 239 extensus, T. 242 a i p la 194 Ceratolithus: 105, 236-9 fa rn s w o r th ii 242 aspliaperta 181 challengeri, 0. 214 fe n e s tr a tu s 156 a a p li f ic u s 237 challengeri, C. 241 f i n i f e r 243 anconitanus 223 chaabrayensis 226 flo r i danus 170 Angulolithina 243 Chrysochroiulina: 17, 148 f o lio s a 163 an n u la 172 Chrysotila: 25, 147 fo ra o s u s , 5. 214 A n o p lo s o le n ia 146 ciperoensis 202 f o r io s u s , D. 226 a n t i l l a r u r 162 Clausicoccus: 149 153, 155 f r a g i l i s 162 a n t i l l e a n a 193 c la v ig e r a 147 fu scu s 155 Apistoneaa: 17, 147 coalitus 228 g alean a 195 a p s t e i n ii 193 C0CC0LITHACEAE 153 Geainilithella: 149, 160 a re a 243 Coccoli thus: Gephyrocapsa: 57, 65, 172 a r c t i c a 148 23, 32, 58, 149-153, 157 giganteus 222 a rg u tu s 226 c o h e n ii 193 g i l l i i 226 a r a a tu s 239 coapacta 181 gladstonensis 197 a s te r 226 co a p a ctu s 205 g lo b u la ta 195 asyaaetricus 221 c o n ic a 196 g lo b u lo s a 194 a t l a n t ic u s 239 c o n ic u s 204 g o zo e n sis 214 a u la k o s 212 c o n v a l1i 5 244 g ra n u la ta 183 a u r i t u s 242 Coronocyclus: 150, 153, 159 h a l l d a l i i 196 b e le a n o s 203 c r i b e l l u n 156 Halopappus: 36, 147 b e l l u s 219 cricotus 161 h a ia tu s 223 b e r g g r e n ii 225 cristatus 240 h a q ii 174 b ig e lo w ii 146 crouchii 181 H a y a ste r: 149, 161 bireticulata 156 Cyc1i cargoli thus: 56, 170 h e i a i i 146 b is e c t a 171 C y c lo p e r f o l i t h u s : HELICOSPHAERACEAE 19? b iz z a r u s 237 149, 153, 160 Helicosphaera: b o l l i i 214 c y l i n d r i c a 195 32, 58, 63-5, 104, 178-182 b r a a r u d ii 220 D a k ty le t h r a : 39, 146 heraosus 176 BRAARUDOSPHAERACEAE 146 d a n icu s 171 heteroaorphosus 204 Braarudosphaera: 105, 146 d a rra g h i 196 hulbertiana 163 b r o u v e r i 220 d a v i e s ii 171 h u x le y i 171 b u k r y i 170 d e co ru s 217 h y a lin a 183 b u r k e i 182 d e f la n d r e i, S. 197 HYHEN0H0NADACEAE 147 calcaris 222 deflandrei, D. 212 Hyaenoaonas 147 C a lc ia r c u s : 23, 148 d e lic a t u s 238 ic a r u s 218 Calcidiscus: 104, 149-153 d e l p h ix 204 intercalaris 226 CALCIOSOLENIACEAE 146 d e n ta tu s 237 in t e r a e d ia ,H . 184 Calciosolenia: 36, 63, 146 Deutschlandia: 39, 147 i n t e r i e d i a , S . 195 californiana 181 Oictyococcites: 56, 171 in v e r s a , H. 185 calyculus, C. 228 d i l a t a t a 193 inversiconica 194 calyculus, S. 203 d i l a t u s 212 in v e rs u s , P. 176 Calyptroli thina: 39, 146 Discoaster: 105, 208 -212 j a f a r i 163 Calyptrolithophora: 39, 146 Oiscosphaera: 39, 147 japonica 173 CALYPTROSPHAERACEAE 146 d i s s i a i l i s 205 k a a p tn e r i, H. 182 287 kaaptneri, S. 197 perch-nielseniae 185 s e l l i i 184 krejcigrafii 163 p e rp le x u s 161 separatus 240 k u g le r i 213 p e t a l i f o r e i s , U. 164 serratus, C. 159 la cu n o sa 172 p e t a l i f o r a i s , D. 214 serratus, 0. 241 lag en a 196 p e tr a e 162 s ib o g a e 163 la a e l l o s a 147 philippinensis 186 s ig n u s 214 la u tu s 226 p i r i f o r a i s 194 s i ap 1 ex 240 le p id a 148 Pleurochrysis: 17, 46,, 147 S o lid o p o n s 162 le p to p o ru s 153 p ii opelagicus 157 S p h a e r o c a ly p tr a 146 l o e b l i c h i i 218 p o ly le p is 148 sphaeroidea 146 l o r d i i 161 PONTOSPHAERACEAE 193 SPHENOLITHACEAE 203 *ac i ntyrei 154 Pontosphaera: Sphenoli thus: 104, 201-3 aartinii, S. 197 59, 104, 189, 193 s t a l i s 181 nartinii, T. 241 p on tu s 158 s t e l 1 u lu s 226 a e x ica n u s 228 p orosa 181 stradneri, C. 164 M ic h a e ls a r s i a: 36, 147 predistentus 202 s t r a d n e r i, D. 223 a i l o u i i 242 preoaci ntyrei 155 s t r i a t u s 243 n in u ta , H. 181 prepentaradiatus 224 subdisticha 156 a in u t a , R. 174 pr iatal i s 157 subdi s t ic h u s 156 a in u t u lu s 173 p r ia u s 236 subsurculus 214 Minylithina 243 p r i ngshei a i i 148 s u r c u lu s 216 ffliopelagicus 158 PRINSIACEAE 170 syaeonjiesi i 146 a i r a b i l i s 162 p r i s a a t i c a 214 SYRACOSPHAERACEAE 147 aisconceptus 223 p ro c e ra 194 Syracosphaera: 36, 39 , 147 a o o re i 212 P ry a n e s iu a 148 t a a a l i s 221 n o r i f o r a i s 205 Pseudoeailiania: 51-5 5 , 172 ta s a a n ia e 156 a u lt ip o r a 189 pseudouabilica 174 te le s a u s 240 a u s ic u s 215 pseudovariabilis 218 Tetralithoides 146 nannopelagicus 158 pugnosa 214 TH0RAC0SPHAERACEAE . 145 Navisolenia 146 pulcherriaa 196 Thoracosphaera: 105;, 145 n e o ab ie s 203 p u lc h r a 147 t o r a lu s 226 neogaaaation 170 P y r o c y c lu s 175 tricorniculatus 237 neohaaatus 222 q u a si tu b i( e r a 197 trinidadensis 212 n e o re c tu s 223 q u a ttr o s p in a 147 TRIQUETORHABDULACEAE 241 nephados 212 queenslandensis 197 Triquetorhabdulus: 233,, 241 n in a e 238 quinqueraaus 225 triradiatus 222 n ite s c e n s 159 q u in t a tu s 225 t r i s t e l 1i fe r 219 Noelaerhabdus: 57 r a d ia n s 201 tr u n c a ta 187 o b liq u a 185 r a d ia t u s 154 tu b e r i 215 o b ru ta 156 r e c t a , H. 187 t u b if e r a 197 o b ru tu s 156 recta, S. 194 Turrisphaera: 23,, 148 ob tu su s 226 re c u r v a ta 194 t u r r i s 196 Ochrosphaera: 17, 147 Reticulofenestra: 82-94, 173 Uabellosphaera: 39, 147 Oolithotus: 149, 162 RHABDOSPHAERACEAE 147 Uabilicosphaera: o ra n g e n s is 176 Rhabdosphaera 147 22, 25, 46, 68, 162 o r i e n t a l i s 186 rhoaba 185 v a r i a b i l i s 216 Orthorhabdus: 234, 241 r o t a r i a 175 ve d d e ri 181 o va ta 173 r o t t i e n s i s 197 v e n t r io s a 197 pacific*, H. 186 r o t u l a 161 v e r e n s is 203 pacifica, R. 173 r u fu s 226 walbersdorfensis 181 pacifica, Sc. 197 ru g osu s 242 v a l l i c h i i 184 pacificus, S. 205 r u t e l l u s 220 w a ltra n s 181 p a le o c a r t e r i 183 sanaiguelensis 215 U ig v a a ia : 23, 148 pansus 218 s a u n d e r s ii 212 v o o d r in g i 226 Papposphaera: 39, 148 Scapholi thus 146 zaaaitaaeapeli 226 p a ta e cu s 155 s c is s u r a , H. 181 p e la g ic u s 157 s c is s u r a , D. 171 penna 195 Scyphosp^cra; pentaradiatus 223 24, 59, 104 . 189, 193 2 8 8

Top View