late neogene (developments in palaeontology and stratigraphy)

THE LATE NEOGENE

BIOSTRATIGRAPHY, GEOCHRONOLOGY AND PALEOCLIMATOLOGY OF THE LAST 15 MILLION YEARS IN MARINE AND CONTINENTAL SEQUENCES

This Page Intentionally Left Blank

Developments in Palaeontology and Stratigraphy, 2

THE LATE NEOGENE

BIOSTRATIGRAPHY, GEOCHRONOLOGY AND PALEOCLIMATOLOGY OF THE LAST 15 MILLION YEARS IN MARINE AND CONTINENTAL SEQUENCES

by

William A. Berggren Department of Geology and Geophysics, Woods Hole Oceanographic Insti tution, Woods Hole, Mass., U.S.A., and Department of Geology, Brown University, Providence, R.I., U.S.A.

and

John A. van Couvering University of Colorado Museum, Boulder, Colo., U.S.A.

Reprinted from Palaeogeography, Palaeoclimatology, Palaeoecology, Vol. 16 No. 1 / 2

ELSEVIER SCIENTIFIC PUBLISHING COMPANY

Amsterdam Oxford New York 1974

ELSEVIER SCIENTIFIC PUBLISHING COMPANY 335 JAN VAN GALENSTRAAT, P.O. BOX 211, AMSTERDAM, THE NETHERLANDS

AMERICAN ELSEVIER PUBLISHING COMPANY, INC. 52 VANDERBILT AVENUE, NEW YORK, NEW YORK 10017

LIBRARY OF CONGRESS CARD NUMBER: 74-10257

ISBN 0-444-41246-8

WITH 16 ILLUSTRATIONS AND 12 TABLES

COPYRIGHT @ 1974 BY ELSEVIER SCIENTIFIC PUBLISHING COMPANY, AMSTERDAM

ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRO-

OR BY ANY MEANS, ELECTRONIC, MECHANICAL, PHOTOCOPYING, RECORD- DUCED, STORED IN A RETRIEVAL SYSTEM, OR TRANSMITTED IN ANY FORM

ING, OR OTHERWISE, WITHOUT THE PRIOR WRITTEN PERMISSION O F THE PUBLISHER, ELSEVIER SCIENTIFIC PUBLISHING COMPANY, JAN VAN GALENSTRAAT 335, AMSTERDAM

PRINTED IN THE NETHERLANDS

To the memory of Professor Orville Lee Bandy (1 91 6-1 973) friend, colleague, antagonist,

but above all gentleman

Hipparion (a 3-toed horse) with Globigerina nepenthes Todd (a planktonic foraminifer) in his jaws. The initial appearance of these two forms has been radiometrically dated at about 12-12.5 m.y. ago and symbolizes the correlation between marine and continental biostratigraphic events and their calibration to a biochronologic time-scale discussed in this study.

FOREWORD

The attainment of ever more precise correlation is the main goal of stratigraphical geology. Subdivision of geological time, as represented in the sedimentary rock record, has occupied the talents and tested the ingenuity of geologists for over two centuries, if we take the work of William Smith as the first significant contribution in this respect. The ability to recognize and distinguish smaller and smaller units of time by biostratigraphical methods is a tribute to an increased understanding of evolutionary patterns, improved instrumentation, and a growing body of informative, descriptive and inter- pretative literature.

of a more refined zonation of Cenozoic marine biostratigraphy using the planktonic Foraminifera. This work, origiiially begun in the Soviet Union in the 1930’s, was later expanded to the Caribbean region, primarily by oil company paleontologists. The information obtained by specialists in various regions lay fallow for many years, but with the recent increase in communication between specialists in all parts of the world, we have seen in the past ten years greater advances in intercontinental zonation and correlation of the Cenozoic biostratigraphy of planktonic Foraminifera than in the preceding thirty. These more recent advances have gone foreward at the same time with the formulation of zonation schemes based on calcareous nannoplankton, radiolarians and diatoms, three other groups of microplankton which are increasingly useful in regional biostratigraphic correlation. More or less simultaneously, studies of “micro-”mammals in the Cenozoic of Europe have brought about remarkable advances in continental biostratigraphy.

Recent advances in geophysics have also had a strong influence on paleontology. Paleomagnetism has aided in the reconstruction of past paleo- geographies (and attendant paleolatitudes) and thus provided a framework for the interpretation of distribution patterns of fossil life forms as well as understanding past climatic history of the earth. Of particular importance has been the influence of the Late Neogene paleomagnetic polarity reversal time-scale whereby biostratigraphic horizons and/or zones can be calibrated to the ordinal time-scale with a high degree of accuracy. Furthermore, the Neogene approximates that part of geological history in which the biological record itself can be calibrated by radiometric dates having a confidence interval less than the life span of an average species (0.1-1.0 m.y.). Such calibrations appear to justify the use of biochronological “datum” events - changes in the fossil record with extraordinary geographical limits - in correlating both land mammal and marine microplanktonic successions in different species assemblage contexts, and in evaluating the synchroneity of successions which have evolved in parallel with occasional exchanges.

Within the space of the last thirty years we have witnessed the development

X

Thus, the last decade has witnessed a veritable explosion of activity in the interrelated fields of Late Neogene marine and continental biostratigraphy, chronostratigraphy and paleoclimatology . We believe that with these refinements to geochronology it is appropriate to pause for a moment and summarize the available information in these areas, discuss some of the conflicts in matters of interpretation and present to the best of our ability a general, uni- fying synthesis of Late Neogene earth history. If our discussions clarify some problems of interpretation we shall be gratified. If it eventually raises more questions than it answers, we may claim some measure of success in our original purpose, for we would agree with Gertrude Stein who summed up the life of the mind so succinctly on her deathbed:

Alice B. Toklas: “What are the answers?’’ Gertrude Stein: “What are the questions?”

Woods Hole, Massachusetts Boulder, Colorado

November, 1973

ACKNOWLEDGEMENTS

This review has profited from discussions with several colleagues on various aspects of the problems discussed here. We should particularly like to thank the late O.L. Bandy (Los Angeles), D.D. Bayliss (Llandudno), W.W. Bishop (London), L.H. Burckle (New York), the late W.H. Blow (Sunbury-on-Thames), M.B. Cita (Milan), E. Delson (New York/Pittsburgh), S. Gartner (Miami), C. Geitzenauer (New York), J. Imbrie (Providence), J. Kennett (Kingston, Rhode Island), K. Perch-Nielsen (Copenhagen), R.Z. Poore (Providence), P. Robinson (Boulder), W.B.F. Ryan (New York), T. Saito (New York), F. Steininger (Vienna/Los Angeles), and L. Thaler (Montpellier). In particular, the senior author thanks Dr. Isabella Premoli-Silva (Milan/Woods Hole) for lively and informative discussions on Late Neogene stratigraphy and for critically reviewing this manuscript.

The senior author would also like to acknowledge a debt of gratitude to the participants at Symposium 109: Late Neogene Epoch Boundaries at the 24th International Geological Congress in Montreal ( August-September, 1972) and the editors of the proceedings of this symposium, Dr. T. Saito and Dr. L.H. Burckle (New York) for allowing him to incorporate unpublished data from the manuscripts in this work. We should like fi,ially to acknowledge the help we have received from the veritable multitude of colleagues who at various time have offered their advice, criticism and information. In a review of this scope and magnitude, which draws on so many published (and, in some instances, unpublished) sources, it would be easy to misrepresent the views of others. In an area which is developing as rapidly as that surveyed here, we have found it necessary to involve a liberal amount of our own interpretation and

XI

evaluation of information. Thus, we request the indulgence of our colleagues if we have misrepresented their views in any way and we assume full respon- sibility for the opinions expressed herein. This investigation has been supported by Grant GA-30723X from the Submarine Geology and Geophysics Program, Oceanography Section, National Science Foundation, to the senior author, and by the Office of Naval Research, contract number N00014-66-C-0241; NR083-004. The junior author received generous support in 1970-1971 from grant no. 2710-1834 of the Wenner-Gren Foundation for Anthropological Research, Inc., of New York, to assist research on paleoecology and intercontinental correlation of African and European Neogene mammal faunas. This is Contribution No, 3152 of the Woods Hole Oceanographic Institution.

CONTENTS

FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . IX

CHAPTER 1 . INTRODUCTION . . . . . . . . . . . . . . . . . 3 Marine biostratigraphic zones . . . . . . . . . . . . . . . . . 4 Geochronology and biochronological ages . . . . . . . . . . . . . 5

CHAPTER 2 . A LATE NEOGENE MICROPLANKTONIC TIME-SCALE . . . . 9

CHAPTER 3 . LATE NEOGENE PLANKTONIC BIOSTRATIGRAPHY . . . . 29

CHAPTER 4 . LATE NEOGENE CHRONOSTRATIGRAPHY . . . . . . . . 37 Late Miocene . . . . . . . . . . . . . . . . . . . . . . . 37 Pliocene . . . . . . . . . . . . . . . . . . . . . . . . 45 The Pleistocene . . . . . . . . . . . . . . . . . . . . . . 50

Base of the Pleistocene . . . . . . . . . . . . . . . . . . . 52

CHAPTER 5 . LATE NEOGENE MARINE BIOSTRATIGRAPHY AND EPOCH BOUNDARIES . . . . . . . . . . . . . . . . . . . . . . 57

The Late Miocene of the Mediterranean Basin . . . . . . . . . . . . 57 ThePlioceneEpochand theMiocene/Plioceneboundary . . . . . . . . 59

Introduction . . . . . . . . . . . . . . . . . . . . . . 59 Discussion . . . . . . . . . . . . . . . . . . . . . . . 60 Paleomagnetic calibration of Pliocene zones . . . . . . . . . . . . 70 Mediterranean D.S.D.P. results . . 1 . . . . . . . . . . . . . . 73 Nannofossil criteria for MioceneIPliocene boundary . . . . . . . . . 74 Pliocene geochronology of the Western Pacific . . . . . . . . . . . 77

The Pleistocene Epoch and the Pliocene/Pleistocene boundary . . . . . . . 79 Beginning of the Pleistocene in the Calabrian stratotype . . . . . . . . 80 Climatic definition of the Pleistocene . . . . . . . . . . . . . . 86 Age of the base of the Pleistocene . . . . . . . . . . . . . . . 87

_.

CHAPTER 6 . LATE NEOGENE MAMMALIAN BIOCHRONOLOGY AND K-Ar TIME-SCALE . . . . . . . . . . . . . . . . . . . . . . . .

Biochronology vs biostratigraphy . . . . . . . . . . . . . . . Europe and Africa . . . . . . . . . . . . . . . . . . . . .

Oeningian (Heer. 1858) . . . . . . . . . . . . . . . . . . Vallesian (Crusafont. 1950) . . . . . . . . . . . . . . . . . Turolian (Crusafont. 1965) . . . . . . . . . . . . . . . . Ruscinian (Kretzoi. 1962) . . . . . . . . . . . . . . . . . . Villafranchian (Pareto. 1865) . . . . . . . . . . . . . . . . Pliocene mammals of Africa . . . . . . . . . . . . . . . . . Biharian (Kretzoi. 1941) . . . . . . . . . . . . . . . . . . Oldenburgian (Kretzoi) . . . . . . . . . . . . . . . . . .

Late Neogene in Asia . . . . . . . . . . . . . . . . . . .

91 91 97 97

101 104 107 109 112 114 118 118

CHAPTER 7 . LATE NEOGENE MARINE PALEOCLIMATOLOGY . . . . . . 123 Atlantic Ocean . . . . . . . . . . . . . . . . . . . . . . 125 Arctic Ocean . . . . . . . . . . . . . . . . . . . . . . . 132 Equatorial Pacific . . . . . . . . . . . . . . . . . . . . . 134 Antarct icrsubantarct ic . . . . . . . . . . . . . . . . . . . 135

CHAPTER 8 . A CHRONOLOGY OF LATE NEOGENE CLIMATE EVENTS . . . 143

CHAPTER 9 . PLIOCENE-PLEISTOCENE INTERCONTINENTAL CORRELATIONS . . . . . . . . . . . . . . . . . . . . . . 159

CHAPTER 10 .SUMMARY . . . . . . . . . . . . . . . . . . 1 6 5

APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . 167

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . 175 Additional references . . . . . . . . . . . . . . . . . . . . 215

CHAPTER I

INTRODUCTION

The term Neogene was established by Hoernes (1853, 1856) to emphasize the fact that Miocene and Pliocene faunas exhibit a close relationship in contrast to the Eocene, sensu Lye11 (1833, 1839, et seq.). He included in his term Neogene essentially all strata from the Aquitanian to the Pleistocene and Holocene (Denizot, 1957), although subsequent usage has generally limited application of the term Neogene to Miocene and Pliocene (Gignoux, 1950, 1955). In the Soviet Union, in particular, the upper limit of the Neogene is drawn at the base of the Quaternary (although there is considerable disagreement where this boundary lies) and, in fact, within the past year the “International Subcommission on the Neogene/Quaternary Boundary” met in the Soviet Union to discuss various aspects of what we would refer to as the Pliocene/Pleistocene boundary. The Neogene, as used in this study, represents the interval from the base of the Miocene to the present time. The Late Neogene encompasses the interval since the Middle/Late Miocene boundary, that is the Tortonian/Serravallian boundary, about eleven million years ago (Berggren, 1972a).

Although there has been a general lowering of temperature of the earth during the Neogene, climatic deterioration began to accelerate during the Late Neogene (about 10 m.y. ago). The evidence from several areas, such as paleontology and oxygen isotope studies (Gill, 1961; Dorman, 1966; Kennett, 1967b, 1968a), sea-level changes (Tanner, 1968), changes in pelagic sedimentation in the equatorial Pacific (Heath, 1969a,b), and radiometrically or paleomagnetically controlled dates on glaciation or ice-rafting (Doumani, 1964; Goodell et al., 1968; Rutford et al., 1968, 1970; Hamilton and Arm- strong, 1969; Denton et al., 1969, 1970), indicate that by late Miocene time an extensive ice-sheet existed in West Antarctica. By middle Pliocene time the ice-sheet in East Antarctica had grown to, or surpassed, its present size (Denton et al., 1971). In the Northern Hemisphere major continental glaciation was initiated about 3 m.y. ago, within the middle part of the Pliocene (Berggren, 1972b).

occurs between the Pliocene and Pleistocene as they are defined at present. Indeed, the conspicuous climatic fluctuations such as those long used as criteria in distinguishing between the two epochs, probably extend well down into the Pliocene and even into the Late Miocene. The onset of continental glaciation which has long been thought to be a major environmental

On a worldwide basis no significant stratigraphical or faunal discontinuity

4

change of sufficient magnitude to separate the Quaternary from the Tertiary Period has no value in worldwide correlation. The world climate has apparently deteriorated gradually with continental glaciation beginning in the high latitudes of the Southern Hemisphere somewhat earlier than in the Northern Hemisphere.

The distinction between a Tertiary and Quaternary Period is invalid and rather artificial (Flint, 1965; Hays and Berggren, 1971). The Quaternary Period has been generally used to encompass the Pleistocene and Holocene epochs. However, the hierarchy of criteria used in distinguishing the various Cenozoic epochs are of the same order. Thus, the Pleistocene may be validly distinguished from the Pliocene as an epoch of the Cenozoic. On the other hand, the Holocene is but an interglacial interval within the Pleistocene Epoch when viewed objectively against the background of the geologic history of the past 2 m.y. It might be more appropriate to recognize simply a Cenozoic Period co-extensive with the Cenozoic Era.

The late Neogene history of the earth should be viewed, then, within the special framework of a gradual, but accelerated, climatic deterioration. This climatic deterioration affected the globe in a subtle manner in tropical regions and more directly and pronouncedly in temperate regions.

MARINE BIOSTRATIGRAPHIC ZONES

The evolution of organisms through time has provided the framework for a system of zonations by which discrete units of time represented by material accumulation of sediments can be recognized. Biozones may generally be grouped into three categories depending on their characteristic features: (I) assemblage zones, those in which strata are grouped together because they are characterized by a distinctive natural assemblage of an entirety of forms (or forms of a certain kind) which are present; (2) range zones, those in which strata are grouped together because they represent the stratigraphic range of some selected element of the total assemblage of fossil forms present; and (3 ) acme zones, those in which strata are grouped together because of the quantitative presence (abundance) of certain forms, regardless of association or range. The latter are, qualitatively, of lesser importance than the first two. Most planktonic zones used in current biostratigraphic work are range zones and are of the following types:

(a) Taxon Range zone - a body of strata representing the total range of occurrence (horizontal and vertical) of specimens of a taxon (species, genus, etc.).

(b) Concurrent Range Zone - a range zone defined by those parts of the ranges of two or more taxons which are concurrent or coincident. It is based on a careful selection (and rejection) of faunal elements which have a concurrent, though not necessarily, identical stratigraphic range with a view to achieving a biostratigraphic unit of maximum time-discrimination and extensibility.

5

(c) Oppel Zone - a zone characterized by a distinctive association or aggregation of taxa selected because of their restrictive and largely concurrent range, with the zone being defined by the interval of common occurrence of all or a specified portion of the taxa. This is a less precise, and more restricted, relative of the Concurrent Range Zone described above. It is little used in planktonic biostratigraphy. As a biochronological concept, it is exemplified by the Land Mammal Age.

(d) Lineage Zone or Phylozone - the body of strata containing specimens representing the evolutionary or developmental line or phylogenetic trend of a taxon or biologc group defined above and below by features of the line or trend. It has been commonly referred to as a phylogenetic zone.

The scope of a lineage zone (phylozone) may extend from the first (evolutionary) appearance of some form in an evolutionary bioseries to the termination of the lineage, thus including the whole bioseries or lineage, or it may include only a segment of the lineage (lineage-segment z o n e ) . For a further discussion on the nature and application of the lineage-zone concept to biostratigraphic studies the reader is referred to Van Hinte (1969) and Berggren (1971~).

opment of some species, genus or other taxon, but not its total range. It is little used, except in local biostratigraphy in planktonic studies.

(f) Interval Zone - the interval between two distinctive biostratigraphic horizons but not in itself representing any distinctive biostratigraphic range, assemblage or feature. The so-called Partial-Range Zone actually corresponds to an interval zone and is commonly used in current planktonic biostratigraphic studies. For a more comprehensive discussion on the nature of biostratigraphic zones the reader is referred to the Preliminary Report on Biostratigraphic Units published by the International Subcommission on Stratigraphic Classification, Report No. 5 at the 24th International Geologic Congress, Montreal, 1972.

(e) Acme Zone - a body of strata representing the acme or maximum devel-

GEOCHRONOLOGY AND BIOCHRONOLOGICAL AGES

Geochronology is the measurement of geological time, and in actual practice consists of placing geological events in a chronological framework. There are fundamentally two ways of determining the age of a prehistoric event: by counting repetitive superimposed features of the geological record such as tree rings, varves, and sedimentary cycles, or by locating events on an ordinal time-scale. An ordinal time-scale is an incremental, continuous, and irreversible modification of some system with each part recognizably unique to its place in the sequence. Measuring time by a non-ordinal scale is like counting library books from the end of the row; the absence of any book need not present an anomaly and the position and nature of a missing book is unknown unless the sequence is compared with other libraries or a perfect model (e.g., a card catalogue) which may or may not exist. The ordinal

6

scales, on the other hand, are like an alphabetized row of encyclopedias which show gaps in their sequence clearly and predict the missing elements by internal comparison. Only two ordinal scales are widely used today, that of radiochronology (based on isotope-decay rates) and that of biochronofogy (based on organic evolution). Geomagnetic polarity reversals are non-ordinal repetitions but because of their wide applicability have been closely correlated to the ordinal time-scales for the Late Neogene; as a result the reversal sequence has taken on a secondary, shadowy ordinality of its own where the paleomagnetic record is so complete that its more distinctive variations can be securely identified.

Obviously, the radiochronological time-scale can be quantified since it is based on well-documented assumptions of the invariancy of isotope-decay rates through geological time, and since radiometric ages are expressed in numbers they conform to what we are conditioned to accept as measurement itself. There is a persistent movement among earth scientists, under- standably among specialists in physical geology and mineralogy but also among paleontologists who work with the more “datable” parts of the fossil record, to do away with biochronology as a primary means of dividing up geological time and to rely entirely on arbitrary and unquestionably convenient time-lines, expressed in rounded-off and exact numbers of years, which are to be identified in the geological record by radiochronology. Biochrono- logical age measurements, even though they are not as easily quantified, are nevertheless implicit in the fossil record itself and do not systematically lose resolution in the older parts of that record as radiometric ages do. Argu- ments as to the real meaning of radiometric numbers, and as to their practical utility everywhere, will eventually divide on the degree to which (numerical) certainty can be demanded in an uncertain universe, but it is our opinion that “biological time” - that is the history of the biosphere and its physical environment - is best measured biochronologically with a critically evaluated calibration assist wherever possible from radiochronology. To do otherwise is to let the radiometric tail wag the paleontological dog.

The relationship between the biostratigraphical age, which is part of the time-stratigraphic face of the lithostratigraphic stage, and the biochronological age is very poorly defined and tends in the minds of all but the most localized workers to merge into one. Our working definition is that when the paleontological expression of a Stage/Age is more or less completely divor- ced from its lithostratigraphical context - either because of distance from the stratotype area so that none of the lithological features associated with the stratotype fossils have meaning in the referred fossiliferous section, or because the interests of the problem do not include lithostratigraphy - it becomes, de facto, a biochronological age or a “biochron”. Problems arise from this which are not wholly conceptual since very often a Stage/Age stratotype is chosen for the convenience it offers to geological mapping and not for the delineation of significant paleontological events. In some well-known instances, and probably others yet to be recognized, the truncation of a

7

fossil sequence by a local lithological accident in a stratotype is mistakenly correlated with some notable paleontological change in more complete sequences elsewhere; the often-quoted but apparently mistaken correlation of the first appearance of Globigerinoides in the stratotype Aquitanian (above an unconformity) and the Globigerinoides global datum itself (Anglada, 1971a) is a case in point.

we see that these originate as paleontological concepts even though they are also tied to a reference section of strata - a “type section” rather than a “lithostratotype” (cf. Blow, 1969, for numerous examples) - and thus are more readily (if unspokenly and commonly without realization) converted to useful biochrons, following our definition above. Mammalian biochrons (Land Mammal Ages, etc.) also originate as local zones tied to reference sections and “type faunas”, even though they are commonly liberated from such earthly bondage almost at birth, and in many instances are created with inferred or abstract limits not observed in the type section itself. Because microplanktonic and mammalian biochronological time units, some under borrowed identities, are becoming increasingly widely used and better characterized and correspondingly more popular with non-specialists, we cannot refrain from quoting the admonition of D.E. Savage (written communication, 1973) that there are observational limits to paleontology beyond which categories (and categorists) should not run: “...when the typify- ing association of fossils is reduced to much less than a majority of the typi- fying taxa, or when the characterizing association is absent, the recognition of the paleontostratigraphic zone (or biochron) ... becomes imaginary”. In other words, tropical planktonic foraminifera zones/biochrons can be correlated with, but not extended into, high-latitude zonation schemes, to give one example, and no matter how perfect the radiometric dating becomes it will never be correct to apply European land mammal biochronological age names to the East African Neogene succession. There will, unfortunately, be just such lapses in the discussions to follow as well as other offenses against the purity we preach, but this can be blamed at least in part on our attempt to be both as current and as comprehensive as possible in this review, so that some of the information dealt with remains incompletely digested. For such unintentional irregularities we beg the reader’s pardon.

Turning to the marine biostratigraphical zone, discussed more fully above,

C HAPTER 2

A LATE NEOGENE MICROPLANKTONIC TIME-SCALE

A Cenozoic radiometric time-scale, to which various zonation schemes based upon planktonic microfossils and time-stratigraphic boundaries have been calibrated, has been formulated over the past several years (Berggren, l968,1969b, 1971a, 1972a). A revised and updated version is presented here (Fig. 1). This time-scale has been the result of the evaluation of a large number of radiometric dates summarized and/or discussed in various papers by Curtis et al. (1961), Evernden et al. (1964), Funnel1 (1964), Evernden and Curtis (1965), Bodelle et al. (1969), Odin et al. (1969a,b, 1970), Bandy and Ingle (1970), Evernden and Evernden (1970), Page and McDougall (1970), Turner (1970), Van Couvering and Miller (1971), Ikebe et al. (1972), Van Couvering (1972), Gill and McDougall(1973), and McDougall and Page (1974).

A review of the development of the geomagnetic time-scale has been presented in recent papers by Dalrymple (1972), Opdyke (1972) and Watkins (1972). In the last decade there have been over fifteen refinements of the original K--Ar-based geomagnetic reversal time-scale established by Cox et al. (1963a). At the present time this time-scale has been calibrated to K-Ar dates back to about 4.5 m.y. (Cox, 1969), although it uses some interpretations based on sea-floor magnetic anomalies and paleomagnetic stratigraphy in deep-sea cores. In a recent development Foster and Opdyke (1970) have extended the paleomagnetic record beyond 5 m.y. (Gilbert/Epoch 5 boundary) and have erected a late Miocene sequence of seven new paleomagnetic epochs numbered from Epoch 5 to 11, the bottom of which is correlated with sea-floor spreading Anomaly 5 at 9.5 m.y.

The relationships between various faunal and floral datum levels to the geomagnetic polarity time-scale have been demonstrated by Opdyke et al. (1966), Berggren et al. (1967), Hays and Opdyke (1967), Hays et al. (1969), Hays and Berggren (1971), Berggren (1972a), Burckle (1972), among others.

A time-scale for the past 15 m.y. is presented in this study. The calibration of land mammal events t o this scale is discussed elsewhere in Chapter 6 on mammalian biostratigraphy and biochronology. The methodology and reasoning by which the marine time-scale has been developed is as follows:

(1) From 0 to 5 m.y. the time-scale is based upon the radiometrically controlled paleomagnetic time-scale of Cox (1969) and the calibrations to it of biostratigraphic events which have been achieved by investigations on deep-sea cores (i.a., Berggren et al., 1967; Hays et al., 1969; Gartner, 1969,

10

1973; Berggren, 1973). The calibration within this part of the time-scale is believed to have a high degree of accuracy and reliability.

graphic events to potassium-argon (K-Ar) dates and to suggested correlation between the sea-floor magnetic anomalies (Heirtzler et al., 1968) and paleomagnetic epochs (below Epoch 5 ) by M.M. Dreyfus and W.B.F. Ryan (24th International Geologic Congress, Montreal, September, 1972, and personal communication). The calibrations within this part of the time-scale must be viewed as moderately tentative.

In correlating the paleomagnetic epochs 6-11 to the magnetic anomaly patterns, Foster and Opdyke (1970, fig. 3) and Opdyke (1972, fig. 9) relied heavily on a correlation between Paleomagnetic Epoch 11 and Magnetic Anomaly 5, the limits of which are, according to the revised reversal chronology of Talwani et al. (1971, fig. 11) 8.71 and 9.94 m.y.

A somewhat different interpretation and calibration of magnetic anomalies and paleomagnetic epochs was provided by Dreyfus and Ryan in a verbal presentation at the 24th International Geologic Congress in Montreal (Sep- tember, 1972). The paleomagnetic epochs were, in general, expanded and displaced downwards relative to the magnetic anomaly pattern such that Epoch 9, rather than Epoch 11, was shown to correspond with Anomaly 5 so that below Magnetic Epoch 9 there is approximately a 2-m.y. difference in the age estimates for the magnetic epoch boundaries. A comparison of the estimated ages of the paleomagnetic epochs according to Dreyfus and Ryan (personal communication), Foster and Opdyke (1970), and Opdyke (1972) is presented in Table I (see Appendix, note 1).

(2) From 5 to 15 m.y. the time-scale is based upon calibration of biostrati-

TABLE I

Age in m.y. of Paleomagnetic Epoch boundaries 6-11

Paleomagnetic Paleomagnetic Epoch boundaries dates Epoch

Dreyfus and Ryan Foster and Opdyke

(personal communication, 1972) (1970)

6

7 t--- 6.6 a 6.6

8.1 t--- 7.2 8

+----- 8.1 a 7.1 9

1 0 - 10.0

a 11.6

t--- 8.1 - 8.1 11 a 12.4 t--- 10.0

12

11

(3) The correlations between the various zones based upon calcareous plankton (planktonic foraminifera and calcareous nannofossils) and siliceous plankton (diatoms and radiolaria) are based upon the senior author’s interpretation of various zonations and correlations proposed by specialists (i.a., Blow, 1969; Martini, 1971; Riedel and Sanfilippo, 1971; Burckle, 1972). (4) In the interval between 5 and 15 m.y. there are few radiometric dates

which can serve as base points. The following are noted here as points of departure:

(a) 15 m.y. estimate for Orbulina Datum (= base Zone N9, Ikebe et al., 1972).

(b) Glass shard dates of 11.4 f 0.6 m.y. and 12.3 ? 0.4 m.y. (Dymond, 1966) on levels within the Mohole dated, respectively as Mohnian and Luisian. A level between the two dates was correlated by Martini and Bram- lette (1963) with the Globorotalia foshi robusta Zone of Bolli (1957), which is the equivalent of upper N12-lower N 1 3 of Blow (1969).

Turner (1970) has determined that the Luisian/Mohnian boundary is younger than 13 m.y. based on a plagioclase date of 13.2 f 0.4 m.y. from a rhyolite flow overlain concordantly by Luisian (possibly lower Luisian) on San Clemente Island, California. This date is consistent with the dates of Dymond (1966). The Relizian/Luisian boundary has been estimated by Turner (1970) to lie within the interval of 13.7-14.5 m.y. This boundary was correlated with the Sphenolithus heteromorphus (NN5)/Discoaster exilis (“6) boundary by Bandy and Ingle (1970). According to Bramlette and Wilcoxon (1967) the upper limit of S. heteromorphus is within the Globoro- talia barisanensis Zone of Bolli (1957, 1966) which is equivalent to Zone N9 (Banner and Blow, 196513; Blow, 1969). However, results of various legs of the Deep Sea Drilling Project have shown that the upper range of S. heteromorphus probably extends to within the Globorotalia peripheroacuta Zone (N10).

A.D. Warren (1972, and written communication to W.A.B.) has made a biostratigraphic study of the coccoliths of the Newport Lagoon section in California (see also Berggren, 1972a, p. 205). His results are summarized below:

(a) Discoaster kugleri was not found in the Luisian Stage although Wil- coxon (1969) reported it together with Sphenolithus heteromorphus in the Luisian at this locality.

(b) Sphenolithus heteromorphus occurs to the top of the Luisian. (c) Discoaster kugleri S.S. was found in a sample of the late Lower

Mohnian. Warren concludes that, “I have good reason to equate the top of the Lo-

wer Mohnian with or close to the top of the D. kugleri Zone and the top of the Luisian with the top of the S. heteromorphus Zone. Additional data supporting this interpretation are the highest occurrence of Cyclococcoli- thina neogammation within the Lower Mohnian and the lowest occurrence of orbulinids in the Luisian.”

12

These results are also confirmed by Lipps and Kalisky (1972) who note that: (a) they have never observed Discoaster kugleri in the Luisian; and (b) this absence is consistent with the presence in the Luisian of Sphenolithus heteromorphus, which in the tropical regions last occurs below the first occurrence of D. kugleri. Thus, the Luisian best correlates with NN5 (see Lipps and Kalisky, 1972, fig. 7).

These data are not in conflict with the dates by Dymond (1966) of levels bracketing the N12/N13 boundary. They merely conflict with the biostratigraphic (zonal) age determination of the position of the Luisian/Mohnian boundary as determined by Parker (1964), Wilcoxon (1969) and Bandy and Ingle (1970) and suggest that the oldest levels penetrated in the Mohole drilling were within the Mohnian rather than the Luisian.

At the same time the correlation of the Luisian/Mohnian boundary with the NN5/NN6 boundary by Warren (1972) and Lipps and Kalisky (1972) provides additional support for a 14--15-m.y. estimate for the Orbulina Da- tum. The initial appearance of Orbulina is within the Lower Luisian Stage in California and thus within the S. heterornorphus ("5) Zone (Warren, 1972; Lipps and Kalisky, 1972). The evolutionary appearance of Orbulina in the tropics occurs approximately at the Helicopontosphaera ampliaperta/Sphe- nolithus heteromorphus (NN4/NN5) boundary. This suggests that the (delayed ) appearance of orbulinids in the Californian Luisian (Bandy et al., 1969c) is minimal (see also Lipps and Kalisky, 1972, who reach the same conclusion). Turner (1970) has estimated a date of 13.7-14.5 m.y. for the base of the Luisian. Lipps and Kalisky (1972) correlate the Orbulina Datum with the base of the Luisian and estimate an age of 14 m.y. for this level. The base of the Relizian has been dated at about 15.3 m.y. by Turner (1970) and correlated with a level within Zone N7 (Bandy et al., 1969a; Lipps and Kalisky, 1972). Thus the Relizian would encompass zones N7 (part at least) and N8 and span the interval of 15.3-14 m.y. approximately if the radiometric dates are accepted at their face values (Lipps and Kalisky, 1972).

Acceptance of a 14-m.y. date for the Orbulina Datum (= approximately base Luisian) would be consistent with Turner's (1970) suggestion that the Luisian/Mohnian boundary is younger than 1 3 m.y. Evidence from deep-sea cores indicates that Zones N10 and N 1 1 are extremely short (probably having a combined duration of no more than 0.3 m.y., whereas Zones N9 and N12 are relatively longer. Thus if Zone N9 is estimated to span the interval of 14-13 m.y., the NN5/NN6 boundary, which corresponds to the Lui- sianfMohnian boundary and to a level within Zone N10, could be placed slightly above 13 m.y. This would still allow for a relatively long Zone N12, of the order of 0.5 m.y. At this point, however, we have a problem in that it is difficult to reconcile the data above with recent information concerning the age of the base of the Tortonian.

The age of the Tortonian/Serravallian boundary is estimated in the following manner. JOIDES Site 16, on the western side of the Mid-Atlantic Ridge, is located over positive Magnetic Anomaly 5. This Anomaly has been

5 5 4 I w i; TORREJONIA

PUERCANl DRAGONIAN

u V

EST COAS ;ALIFORNI A

MARINE STAGES

HALLIAN

IHEELERlAl

)ELMONTI AN

MOHNIAN

LUISIAN

RELlZlAN

SAUCESIAN

!EMORRIAL

-- - ?---

REFUGIAN

NARIZIAN

ULATISIAN

PENUTIAN

BULITIAN

YNEZIAN

"DANIAN"

0'' - PALEOTEMPERATLWE URVE - NEW ZEALANO

(DEVEREUX, 19671

CENOZOIC PLANKTONIC

Pg- ( ~ ~ ~ ~ ~ ~ N ~ E & 'Yd, TAGES BLOW. (369; BLOW and D A T U M PLANES

CENOZOIC PLANKTONIC FORAMINIFERAL

BERGGREN. UnPW -

S)<Globorotolia acostaenais Datum

D d G I o b i g a r i m napcnlhes Datum . 8 1 ~ G l o b o r o I o l i a fohsi Dalum

Fig. 1. Cenozoic radiometric time-scale, chronostratigraphy, planktonic foraminifera1 zonation and datum levels.

17

dated at 9.5 m.y. (Talwani et al., 1971). The oldest sediments recovered were approximately 13.4 m above the basement basalt and were correlated to the Tortonian by calcareous plankton and planktonic foraminifera. Thus, the base of the Tortonian is apparently older than 9.5 m.y.

In a study of an equatorial Pacific core (RC12-65) Burckle (1972, p. 223) reports Discoaster hamatus in the upper part of Magnetic Epoch 11 (as well as the lower part of Epoch 10, L.H. Burckle, personal communication, 1973), whereas in core RC12-64 Saito (in Burckle, 1972) records Globorota- lia mayeri in the lower part of Magnetic Epoch 11 but not Discoaster hamatus. The significance of these observations lies in the following facts:

(a) The Tortonian/Serravallian boundary lies within Zone N15 (Cita and Blow, 1969).

(b) Elements of the Discoaster hamatus Zone ("9) are present in the lower part of the Tortonian type section (Martini, 1971) and also appear to be restricted to the Globorotalia menardii Zone of Bolli (1957, 1966) (which is equivalent to Zone N14) and N15 (Bronnimann et al., 1971, p. 1732- 1742). Thus the Tortonian/Serravallian boundary lies within Zone "9.

(c) The record of Globorotalia mayeri by Saito (in Burckle, 1972) could indicate the presence of Zone N13 in the lower part of Magnetic Epoch 11.

This means that the base of the Tortonian is probably situated within the interval of the upper part of Magnetic Epoch 11 to the lower part of Epoch 10. On the scale of Opdyke (1972) this would be within the time interval of 8.5-9 m.y. This is quite inconsistent with marine data summarized above as well as with comparative data available from continental stratigraphy. On the calibration of Dreyfus and Ryan, however, this would lie within the interval of -11-12 m.y. (see Table I). The significant point is that the correlation of Dreyfus and Ryan of paleomagnetic epochs to the magnetic anomaly pattern is more consistent with the available date on the ages of the planktonic zones between N9 and N13. Adoption of, and calibration to, the scheme of Foster and Opdyke (1970) and Opdyke (1972) would necessitate the expansion of zones N13-Nl5 to cover the interval of 12-12.5 m.y. up to 9 m.y. which is entirely inconsistent with the evidence in the deep-sea record.

ary, and an age of 11-11.5 m.y. is estimated for the N15/N16 boundary (which is approximately correlative with the base of the Tortonian), then zones N13 and N14 must be placed within the time span of about 1 m.y. In the context of our knowledge of the relative length of these zones in deep-sea cores this is not unrealistic. On the basis of presently available data we are faced with the following dilemma, however: if we directly calibrate biostratigraphic (zone) ages of California stage boundaries to the radiometrically determined ages (Turner, 1970) and accept an ll-11.5-m.y. age for base Tortonian we are faced with a very short span of time for zones N13 and N14. On the other hand, when we estimate an age of 15 m.y. for the Orbulina Datum at the base of Zone N9 we are faced with the necessity of

On the other hand, if an age of 12 m.y. is assigned to the N13/N14 bound-

18

extending the top of this zone up to at least 13 m.y. to be consistent with Turner's (1970) data which indicates that the base of the Mohnian is younger than 13 m.y. It would also require a downward revision of the 15.3-m.y. &mate for the base Relizian (within Zone N7) in order to allow room for the N7-N8 interval. It would appear on the basis of the principle of minimum-data manipulation that it would be more satisfactory to estimate a date for the Orbulina Datum (and the base of the Luisian) around 14 m.y. rather than 15 m.y.

Estimates on the ages of various Middle-Late Miocene planktonic zones (Tables 111, V, VII, and IX) are based on extrapolated rates of sedimentation between reliably dated datum levels (Tables 11, IV, VI, and VIII). Data from four sites of the Deep Sea Drilling Project (62, 63, 77 and 214) have been examined in detail. Once a rate of sedimentation is calculated, an interval check is run on the biostratigraphic distribution of some of the taxa within and adjacent to the calculated interval. For instance, a sedimentation rate of 2.8 cm/1000 years was calculated for the Pliocene at Site 77 (Table VI). Applying this rate to the Pliocene the disappearance of Globigerinu nepenthes at 69 m (Hays et al., 1972a, p. 80) occurs at a level estimated to be about 3.9 m.y. which agrees well with its known extinction in the Cochiti Event of the Gilbert Reversal Epoch at 3.7 m.y. The base of the Reticulofe- nestru pseudoum bilica Zone ("15) whose paleomagnetic age base is at about 3.5 m.y. occurs at 65 m at Site 77 (Hays et al., 1972a, p. 80) which is also in close agreement.

Of the four sites examined Site 62 has shown itself to be the most reliable for estimating ages of zonal boundaries. It would appear that this site had a rather constant rate of sedimentation throughout the Late Miocene and Plio-

TABLE I1

Sedimentation rates between datum intervals used in calculating ages of Late Miocene zonal boundaries in D.S.D.P. Site 6 2

Datum Interval Depth Datum Sedimentation rate (cm) (age in m.y.) (cm/1000 years)

2.5 I. a. extinction discoasters 42 1.7

d. N11/12 146 5.7

11. c. NN12/13 126 4.5 e. NN8/9 314 12.0 2.5

111. b. NN15/16 c. NN12/13

2.7 85 3.0 126 4.5

2.67 IV. a. extinction discoasters 42 1.7 e. NN8/9 314 12.0

Comparison of the figures shows that rates of sedimentation were relatively constant throughout the Late Miocene to Pliocene at Site 62.

19

TABLE 111

Estimated age of Middle-Late Miocene planktonic zonal boundaries in D.S.D.P. Site 62 based on rate of sedimentation in Table 11, Datum Interval IV

Planktonic foraminifera Calcareous nannoplankton ~

Zone Depth Age Zone Depth Age (m) (m.y.) (m) (m.y.)

N18117 140 5.4 NN 1311 2 126 4.8

N17/16 21 5 8.2 NN 12/11 146 5.6

N16/15 280 10.6 N N l l / l O 263 10.0

N15/14 30 1 11.4 NN 10/9 291 11.0

N14/13 309 11.7 NN 9/8 314 12.0

NN8/7 316 12.1

NN7/6 335 12.6

"615 360 13.6 __ -

Radiolarian

Zone Depth (m) Age (m.y.) ~~ ~~~

0. penultimus/O. antepenultimus 226 8.6

0. antepenuftimus/C. petterssoni 295 11.2

C. petterssoni/D. alata 330 12.5

cene of about 2.67 cm/1000 years (Table 11). The use of this rate yields the data in Table 111.

Rates of sedimentation apparently vaned at Site 63 from approximately 1 to 2 cmflOO0 years (Table IV) which accounts for the spread in estimated ages shown in Table V. Below 113 m (NN8/9) the estimates based on a rate of 2 cm appear to be more reliable because an estimate of 14.3 m.y. for the N8/9 (Orbulina Datum) is more consistent with other data (Turner, 1970). At Site 77 markedly different rates of sedimentation occurred in the Late Miocene and the Pliocene (Table VI) and it is not possible to determine where the change occurred or whether it fluctuated significantly. This accounts for the spread in age within the later part of the Late Miocene (Table VII). The dates in the earlier part of the Late Miocene and in the

TABLE IV

Sedimentation rates in datum intervals used in estimating ages of Middle-Late Miocene planktonic zonal boundaries in D.S.D.P. Site 63

Datum Interval Depth Datum Sedimentation rate (m) (age in m.y.) (cm/1000 years)

1 .o I. a. S. pentas/S. peregrina 31 4.2

11. b. 0. antepenultimus/

c. NN9/8 113 12.0

C. petterssoni 103 11.5 c. NN9/8 113 12.0 2.0

Comparison of the figures shows that rates of sedimentation here are twice as high in the Middle Miocene as in the Late Miocene-Pliocene. (Age of datum levels comes from primary dating and not from sedimentation-rate ages in other sites.)

TABLE V

Estimated ages of Middle and Late Miocene planktonic zonal boundaries in D.S.D.P. Site 63 based on rates* of sedimentation in Table IV

Planktonic foraminifera Calcareous nannofossils

Zone Depth Datum Age Zone Depth Datum Age ~-

(m) Interval (m.y.) (m) Interval (m.y.) _________.__

within I 11.5 within I 7.6 N15 lo5 I1 11.6 N N l l 65 I1 9.5

N13/12 132 I1 13.0 NN8/7 119 I1 12.3

N 10/9 143 I1 13.5 NN7/6 128 I1 12.7

N9/8 160 I1 14.3 NN6/5 135 I1 13.1 -~ - - ______-- ____._____

Radiolarian

Zone Depth Datum Age (m) Interval (m.y.1

within 0. penultimus

0. penultimus/O. antepenultimus

65 I 7.6 I1 9.5

68 I

I1 7.9 9.8

I 11.4 - I1 0. antepenultimus/C. petterssoni 103

C. petterssoni/D. alata 132 I1 13.0

D. alata/C. costata 150 I1 13.8

* Rates of sedimentation are twice as high in Middle Miocene as in Pliocene. Below 113 m (NN9/8) the ages derived from Datum Interval I1 are consistent and more accurate, whereas above this level those derived from Datum Interval I are consistent and more accurate.

__-__ ~-

21

TABLE VI

Sedimentation rates in datum levels used in calculating ages of Late Miocene zonal boundaries at D.S.D.P. Site 77

Datum Interval Depth Datum Sedimentation rate (m) (age in m.y.) (cm/1000 year)

2.8 I. a. top Coccolithus pelagicus 27 1.7

b. NN12/13 85 4.5

11. b. NN12/13 c. NN8/9 1.1

85 4.5 171 12.0

The rate of sedimentation changes considerably between the Late Miocene and the Plio- cene; this is consistent with the calculated ages of datum levels within these two intervals.

TABLE VII

Estimated ages of Middle and Late Miocene planktonic zonal boundaries in D.S.D.P. Site 77 based on sedimentation rates in Table VI

Planktonic foraminifera Calcareous nannofossils

Zone Depth Datum Age (m) Interval (m.y.)


N18/17 82 I I1

I 154 I1

N17/16

4.5 4.2

8.1 10.8

Radiolarian

I 8.4 11.3

NN9/10 160 II

NN8/7 180 I1 13.1


S. p e n t a d s . peregrina

S. peregrina/O. penultimus

0. penultimus/O. antepenultimus

I 4.7 I1 4.7

87

100-110 I I1

151 I I1

5.1-5.9 5.9-6.8

7.9 10.5

0. an tepen u ltim us/C. p e t tersso ni 165 I1 11.5

C. petterssoni/C. laticonus (D. alata) 187 I1 13.6

Rates of sedimentation are twice as high in Middle Miocene as in Pliocene. Below 113 m (NN9/8) the ages derived from Datum Interval I1 are consistent and more accurate, whereas above this level those derived from Datum Interval I are consistent and more accurate.

22

TABLE VIII

Sedimentation rates in datum levels used in calculating ages of Middle-Late Miocene planktonic faunal boundaries in D.S.D.P. Site 214

Datum Interval Depth Datum Sedimentation rate (m) (age in m.y.) (cm/1000 year)

I. a. NN12/13 b. NN9/8 1.0

86 4.5 161 12.0

TABLE IX

Estimated ages of Middle-Late Miocene planktonic faunal boundaries at D.S.D.P. Site 214 based on sedimentation rates in Table VIII

Planktonic foraminifera Calcareous nannoplankton

Zone Depth Age Zone Depth Age (m) (m.y.1 (m) (m.y.1

N 18/17 95 5.4 "13 + 14/15 76 3.5

N 16/15 153 11.2 NN11/12 9 3 5.4

N 15/14 161 12.0 NN 10/9 154 11.3

N 13/14 166 12.5 NN 716 171 13.0

N 12/13 171 13.0 NN 6/5 180 14.4

Radiolarian

Zone Depth (m) Age (m.y.1

S. peregrina / 0. penultimus 119 8.0

~ _ _ _ _ _ ~ _ ___ _-___

0. penul t inus / 0. antepenultimus 134 96.6

0. antepenultimus / C. petterssoni 152 11.1

C. petterssoni/ D. alata 165 12.4 __ __ . -~ - _______---

Middle Miocene are more reliable when the lower rate (1.1 cm/1000 years) is used. At Site 214 (Indian Ocean, Leg 22, unpublished) a relatively constant rate of sedimentation of about lcm/1000 years occurred (Table VIII). The estimated ages for middle and late Miocene zones appear to be consistent with the data at the other sites in most instances.

A synthesis of the data in Tables 11-IX yields the following information:

23

(1) Sedimentation-rate estimates on the age of the Ommatartus antepenultimus/Cannartuspetterssoni boundary vary from 11.1 to 11.5 m.y. (11.1 at Site 214, 11.2 at Site 62, 11.4 at Site 63, and 11.5 at Site 77).

(2) The Discoaster hamatus Zone ("9) spans the interval between -12 and 11 m.y. (12-11 at Site 62; 12-11.3 a t sites 77 and 214).

(3) The estimated duration of the Globorotalia continuosa Zone (N15) is approximately 0.8 m.y., during the period from roughly 12 to 10.5 m.y. (12-11.2 in Site 214; 11.4-10.6 in Site 63; a date of 11.5 is also estimated within N15 at Site 63). An estimate of -11.5 to 10.7 m.y. appears reasonable for Zone N15.

(4) The Globigerina nepenthes Datum (N13/N14) occurs between 11.7 and 12.5 m.y. (11.7 in Site 62; 12.5 in Site 214). It is shown to occur slightly above the base of the Discoaster hamatus Zone ("9) in Bronnimann et al. (1971, p. 1732, 1742) and by the paleontologists on Leg 22 (Site 214, unpublished: verified by a personal examination of this material by W.A.B.), whereas it is shown to occur below the base of the D. hamatus Zone in Hays et al. (1972a, p. 61, 81). This apparent diachronism could account for the spread in ages which brackets the estimated age of 12.0 m.y. for the base of the D. hamatus Zone.

The weight of evidence indicates that the Globigerina nepenthes Datum (N13/14) is equivalent to, or only slightly younger than, the NN9/8 boundary at 12.0 m.y. If the N14/15 boundary is at about 11.5 m.y. the N13/14 boundary occurs between 11.5 and 12 m.y. It is apparent that we are dealing here with a very short zone. The Globigerina nepenthes Datum (N13/14) is correlated here with the NN9/8 boundary at 12.0 m.y.

in the interval of 12.4-13.0 m.y. (12.4 in Site 62; 13.0 in Site 63). This boundary has been consistently correlated with a level equivalent to the Discoaster kugleri Zone ("7) and the Sphaeroidinellopsis subdehiscens- Globigerina druryi Zone (N13) (Bronnimann et al., 1971, p. 1471, 1472; Hays et al., 1972a, p. 48, 87-88). Thus the D. alata Zone is equivalent to calcareous nannofossil zones older than N N 8 and planktonic foraminifera1 zones older than N14. If the G. nepenthes Datum (N13/14) is approximately equivalent to the NN9/8 boundary, which in thrn occurs within the mid-part of Magnetic Epoch 11, then the record of Globorotalia mayeri in the lower part of Magnetic Epoch 11 (Burckle, 1972) may well refer to the presence of Zone N13, not necessarily N14. The C. petterssoni/D. alata boundary, which occurs within N13 may thus be correlated to the base of Magnetic Epoch 11 at 12.4 m.y. (Burckle, 1972, fig. 10, correlated the C. petterssoni/D. alata boundary with the mid-part of Magnetic Epoch 10. This is quite impossible on the basis of Burckle's own data and the biostratigraphic data presented above.)

(6) The estimated age of the Ommatartus antepenultimus/O. penultimus boundary ranges from 7.9 to 10.5 m.y. (7.9 or 9.8 in Site 63, 8.6 in Site 62, 7.9 or 10.5 in Site 77, and 9.6 in Site 214). The older age (10.5 m.y.) may

(5) The Cannartus petterssoni/Dorcadospyris alata boundary occurs with-

24

be dismissed as probably too old in view of the fact that the N15/16 boundary is estimated to occur at about 10.7 m.y. and the 0. antepenultimus/O. penultimus boundary is rather younger. The 0. antepenultimus/O. penultimus boundary corresponds approximately with the N16/17 and the NN10/11 boundaries, although on the other hand there is evidence that the NN10/11 boundary may lie within the 0. antepenultimus Zone. The 0. penultimus/O. peregrina boundary may be reliably correlated with Paleomagne- tic Epoch boundary 6/7 (L.H. Burckle, 1972, fig. 4 and personal communication). The 0. antepenultimus/C. petterssoni boundary has been shown above to lie at roughly 11.1-11.5 m.y., that is within the lower part of Mag- netic Epoch 10. The N16/15 boundary is at about 10.5 m.y. (within the upper part of Magnetic Epoch 10). Thus the 0. penultimus/O. antepenultimus boundary occurs within the interval of magnetic epochs 7-9 (> 6.3 - < 8.7 m.y.). A tentative estimate of 8.6 m.y. is made here which corresponds approximately with the Magnetic Epoch 8/9 boundary.

(7) The N17/16 boundary presents some problems. Age estimates vary from 8.1 to 10.8 m.y. (8.1 or 10.8 at Site 77, 8.2 at Site 62). The older age of 10.8 m.y. may be dismissed as too old since the N16/15 boundary occurs at about 10.7 m.y. The N16/17 boundary was placed within the 0. penulti- rnus Zone by Bronnimann et al. (1971, p. 1741), whereas it was placed within the 0. antepenultimus Zone by Hays et al. (1972a, p. 48). This discrepancy is due, no doubt, to differing concepts on the part of the planktonic foraminiferal specialists of Globorotalia plesiotumida, nominate taxon of Zone N17. The personal experience of the senior author indicates that the base of Zone N17 coincides approximately with the 0. penultimus/O. antepenultimus boundary which has been estimated above at 8.6 m.y., although this age is uncertain within large limits.

(8) The Discoaster quinqueramus Zone ("11) spans the interval between 5.7 m.y. and approximately 10.0 m.y. The NN11/10 boundary occurs between the N16/15 boundary (dated at 10.7 'm.y. b.ere) and the 0. penultimus/O. antepenultimus boundary (dated at 8.6 m.y. here). The NN11/10 boundary should coincide approximately with the Paleomagnetic Epoch 9/10 boundary, at about 10.0 m.y.

regarding the age of several important Middle-Late Miocene boundaries*.

ary in the Middle East. This level was dated at about 12 m.y. by Dymond (1966) and estimated to be about 12-12.5 m.y. old by Page and McDougall (1970). The estimate was revised to 14 m.y. by Berggren (1972a) in order to agree with the estimate by Van Couvering and Miller (1971) of 13.5 m.y. for the Globigerina nepenthes Datum which was then believed to be older than the Hipparion Datum at 12.5 m.y. Recent evidence (see above, and Chap- ter 6 on mammalian biochronology) has shown that the G. nepenthes Datum is approximately equivalent to, or slightly younger than, the Hipparion Da-

* See Appendix, note 1.

On the basis of the data presented above we may summarize the evidence

(1) The N13/12 boundary and the age of the Tf,-,/Tf, letter stage bound-

25

tum, thus obviating the need for reevaluating the estimated age of the Tf, -2 /Tf, boundary. This boundary is drawn here at about 12.5 m.y. to coincide with the N12/13 and the C. petterssoni/D. d a t a boundaries at the Magnetic Epoch 11/12 boundary.

(2) The Serravallian/Tortonian boundary occurs within Zone N15 and Zone "9. The 0. antepenultimudC. petterssoni boundary occurs within the same interval and only slightly above the base of Zone N15. These zones are estimated to have the following duration:

N15: 11.5-10.7 m.y.; "9: 12-11 m.y. Thus the Serravallian/Tortonian boundary probably coincides approximately with the 0. antepenultimus/C. petterssoni boundary; an age of 11-11.2 m.y. is estimated for this level.

(3) The age of the Messinian/Tortonian boundary may be estimated in the following manner:

The N16/17 boundary would appear to lie within the upper part of the Tortonian. The upper part of the Tortonian (uppermost part of Gino's Unit 4, and Units 5-7) overlaps, at least in part, the biostratigraphic extent of the neostratotypic Messinian (Cita and Blow, 1969, fig. 9 on p. 583, 590, 591). According to the principle that "base defines boundary" (George et al., 1969) this section should be included in the Messinian.

The total range of Globorotalia plesiotumida (nominate taxon of Zone N17) and Globorotalia tumida (nominate form of Zone N18) were recorded at Sites 62 and 77 in the western equatorial Pacific (Bronnimann et al., 1971; Hays et al., 1972a). The range of G. plesiotumida at the two sites differs in terms of the radiolarian and calcareous nannofossil biostratigraphy , but this difference is probably due to differing taxonomic concepts of the paleontologists involved. Our work suggests that the base of the G. plesiotumida Zone (N17) is approximately correlative with, or only slightly older than, the 0. antepenultimus/O. penultimus Zone and is within Zone N N l l (the base of which is dated here at about 10 m.y. and correlated with Paleo- magnetic Epoch 9/10 boundary).

A preliminary estimate of 8.6 m.y. is made here for the 0. penultimus/O. antepenultimus boundary, and the N17/16 boundary is tentatively correlated with this level (Figs. 2, 3).

The Tortonian/Messinian boundary lies within Zones N17 and "11. Apparently no significant planktonic foraminiferal or calcareous nannoplankton event occurs at the Tortonian/Messinian boundary. However, current studies on the diatom floras of the type Messinian may ultimately play a major role in determining the age of the base of the Messinian. L.H. Burckle (personal communication, 1973) finds a diatom flora in the basal Messinian characteristic of Magnetic Epoch 6. W.W. Wornardt Jr. (personal communication, 1973) has found diatom floras characteristic of the Delmontian in the type Messinian. The base of the Delmontian Stage of California (base Sti- chocorys peregrina Zone) corresponds to the base of Magnetic Epoch 6 which is at about 6.6 m.y. (see below). Thus present evidence suggests that

26

STATIGRA- in

5

6

7 -70 -

6.0 - W

- 8 . 0 - z 8

9 w - 9.0 -

- iO.0 - 0

lo -11.0 - 11 -12.0- 0 4

-130 - - -14.0-

2 - 1 5 0 -

16

ClPLlFORNi

N i 9

a 0. penultirnur

0 ontepsnulti -

0 C. peiterrsoni

DATUM LEVELS

G tvmido - S dehiscenr G dehiscenr (extinction)

G ocortaensis

G neDenth0.S

G fohsi

Orbulino

Te I Fig. 2. Correlation and calibration of Middle and Late Miocene planktonic zones to the paleomagnetic time-scale.

the Tortonian/Messinian boundary may correspond approximately with the Magnetic Epoch 6/7 boundary at about 6.6 m.y. This estimate may be compared with the radiometric dates of 8-6.6 m.y. on the lower Messinian Melilla tuffs in Morocco, which unconformably overlie folded “marnes bleues” with an N15 fauna (Choubert et al., 1968, p. 197).

(4) Age estimates of Late Neogene radiolarian zones and California stages can also be made in the following manner:

(a) The radiolarian biostratigraphy of Late Neogene sequences in southern California and the experimental Mohole has been delineated by Casey (1972).

(b) Late Neogene radiolarian biostratigraphy of two southwest Pacific deep-sea cores has been discussed by Opdyke (1972). The ranges of several taxa are plotted with reference to the magnetic epoch sequence (see Figs. 2,

(c) Calibration of the magnetic epoch sequence to the paleomagnetic ti- 3).

me-scale based on extrapolated rates of sea-floor spreading can be made (M.M. Dreyfus and W.B.F. Ryan, personal communication, 1973; see Ta- ble I). In this manner it is possible to estimate the relationship of various stratigra-

27

GlLBER

0

60 6

70

4' 8 0

5 i.." 100

N19 "13

- 80

- 9 0

-100 NN 10

Fig. 3. Correlation and calibration of Middle and Late Miocene planktonic zones, datum levels, and chronostratigraphic units.

phic boundaries in the California succession to those described elsewhere. Some interesting results are listed below:

(1) The Ommatartus penultimuslStichocorys peregrina boundary occurs near the Mohnian/Delmontian boundary (Casey, 1972). The base of the S. peregrina Zone occurs near the base of Magnetic Epoch 6, which is at about 6.6 m.y. This means that the Mohnian Stage of California spans a 6-7-m.y. time interval and probably corresponds to zones Nll--16, and at least the lower half of N17 (see Fig. 3). In its extent it is comparable to the Saucesian Stage (Early Miocene).

de with the extinction level of Prunopyle titan and Lychnocanium grande and the initial occurrence of Lamprocyclas heteroporos near the Delmont- ian/Repettian boundary (Casey, 1971). These events correspond to a level slightly above the Kaena Reversed Event (2.8 m.y.) of the Gauss Magnetic Epoch. However, this is in the Upper Pliocene in terms of Mediterranean stratigraphy. The Miocene/Pliocene boundary can be dated at about 5 m.y. (see Berggren, 1973, and below). The Miocene/Pliocene boundary in Califor- nia is situated within the Delmontian Stage, the limits of which span the interval of approximately 6-2.8 m.y. (see Fig. 3).

(2) The Miocene/Pliocene boundary in California is interpreted to coinci-

CHAPTER 3

LATE NEOGENE PLANKTONIC BIOSTRATIGRAPHY

Planktonic foraminiferal zonations of late Neogene equatorial sediments have been developed by Bolli (1957,1966,1970; modified by Bolli and Premoli-Silva, 1973), Banner and Blow (1965b), Bolli and Bermudez (1965), Cati et al. (1968), Blow (1969), Berggren (1971b), Lamb and Beard (1972), Cita (1973), among others. A sixfold zonation of the Pliocene has been recently suggested (Berggren, 1973; see Fig. 4) to supplement the standard threefold subdivision by Blow (1969). Within the Pleistocene a number of paleoecologically controlled “zones” have been developed by Ericson et al. (1963,1964b) and Ericson and Wollin (1968). Aspects of these various zonations have been used by numerous authors, and variants of one or another of these various schemes are now routinely applied in stratigraphic studies. Calcareous nannoplankton zones have been developed for the Late Neogene of the equatorial regions by Gartner (1969), Martini and Worsley (1970), and Martini (1971).

Biostratigraphic zonation of late Neogene sediments based upon siliceous organisms has been developed only relatively recently and almost entirely as a consequence of the concentrated investigations in deep-sea areas. The pioneering investigations of Riedel (1957,1959), Riedel et al. (1963), Riedel and Funnel1 (1964), and Friend and Riedel(l967) may be said to have laid the groundwork for a radiolarian zonation of equatorial Pacific Neogene sediments (Riedel and Sanfilippo, 1970, 1971). The correlation and calibration of Neogene equatorial planktonic zones is shown in Fig. 5.

High-latitude zonation schemes based on planktonic foraminifera have been formulated for the Pleistocene of the Subantarctic (Kennett, 1970), the mid-Pliocene to Pleistocene of New Zealand (Kennett et al., 1971), and for the Late Neogene of New Zealand (Jenkins, 1971). Lack of data has prevented similar subdivision in the North Atlantic, although a preliminary attempt was made by Berggren (197213) to subdivide the Late Neogene of the North Atlantic based on assemblage zones of planktonic foraminifera.

In the Antarctic radiolarian zonations developed by Hays (1965), Opdyke et al. (1966) and Hays and Opdyke (1967) each zone was designated by a Greek letter with the youngest zone being designated by the last letter of the Greek alphabet, C2 (omega), and each older zone by the next earlier letter backwards of the Greek alphabet. The relationship of this zonation to a paleomagnetic time-scale was shown by Opdyke et d. (1966) and Hays and Opdyke (1967). The zonation has been modified by Bandy et al. (1971a).

I - A;fEAG-, NETIC '

SCALE

W

Z

W

0

0

-

-I

a ~

l

+

a

eudoumbilica

Discoaster

losymmetricus 1 1 ' 1 TABlANlAN "14

Ceratolithus

ZANCLIAN

MESSlNlAN

Discoaster

I I 1

1 PLi

G l o b o r o f o l i o pler iolumida

I A 2 IN IT IAL APPEARANCE

E = E X T I N C T I O N ' I

W 0

Fig. 4. Correlation and calibration of Pliocene planktonic foraminifera1 and calcareous nannoplankton zones.

31

-

IME IN

my

-

9 9

10-

15-

20 -

9 -

EPOCH

/ SERIES L - E

L

E

-

L

M

E

- LEIS1 CENE - Y z W

P a -

W

z

W

0

0

-

I

-

4GE /1 PLAN KTONl C

FORAMIN I FERAL DATUM PLANES / STAGE

G. lruncalulinoides Dolum

G. oltispira. G. mu1ticameroto.- S. seminulino (edinctionl Datum

B G. nepenthes (ealinclionl Dalum Z S

CALAERIAN

'IACENZIAN - ZANCLIAN

I

OLIGOCENE

PLAN KTONlC FORAM I N IFERAL

ZONES BLOW (1969)

p z i r NANNO-

PLANKTON ZONES

ERAMLETTE e WILCOXON (1967). HAY el oi (1967). GARTNER (1969),

EAUMANN a R O T H ( I ~ ~ S MARTINI 8 WORSLEY 119701

~

RADIO- LARIAN ZONES

REDEL e SANFl L I PPO 1970) (197

MOORE (197

'p r i rmal ium

S. pentas

S. peregrina

~

0,penultimus

0. onlepenul-

C. (? )

limus

pellerssoni

D. a la to

C. caalala

C.virginis

_ _ _ _ _ L. biper _ _ - - -

D. papi I io

Fig. 5. Correlation and estimated biochronologic relationship of Neogene planktonic zones and planktonic foraminifera1 datum levels (see text for further explanation).

The zonation and its chronologic relations are shown in Table X which has been modified from Bandy et al. (1971a, table 1, p. 2). The modifications made by Bandy et al. (1971a, p. 3) were said to have been based upon "warm water polycistines that are still living today, a group of Miocene polycistines that with one exception have been reported almost exclusively from Miocene strata and several orosphaerids that include some very important Miocene index species." Bandy (1967a) had previously suggested that the widespread occurrence of Prunopyle titan in the lower part of many Antarctic cores might be used to denote the top of the Miocene in view of its restriction to upper Miocene beds (Campbell and Clark, 1944; Ingle, 1967). How- ever, it should be pointed out that this argument is specious, inasmuch as the

32

TABLE X

Antarctic - Subantarctic radiolarian zonation (modified from Bandy et al., 197 l a , according to most recent calibration of paleomagnetic time-scale)

Radiolarian zone Age according to Letter subdivision by Hays (1965) paleomagnetic Bandy et al. (1971a) Opdyke et al. (1966) time-scale Hays and Opdyke (1967) (m.y. B.P.)

(omega)

X (chi)

T (upsilon)

0.4

0.7

1.8

2.4

2.1

3.4

3.6

4.2

T (tau) 4.5

a. modern Spongoplema antarc- ticum complex with warm- water forms such as Theoconus zancleus and Saturnulus planetes

b. modern Spongoplema antarc- ' ticum complex without a

warm-water influence

Acanthosphaera sp. group (middle Brunhes)

Saturnulus planetes and Pterocanium trilobum (upper Matuyama boundary)

Eucyrtidium calvertense (Gilsa Event)

_ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - -

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - -

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - -

_ _ _ _ _ _ _ _ _ _ _ _ _

a. Desmospyris spongiosa and Helotholus uema (lower Matuyama)

_ _ _ _ _ _ _ _ _ - _ _ _ _ - - _ _ _ b. Prunopyle titan and

Lychnocanium grande (upper Gauss)

_ _ _ _ _ _ _ _ _ - _ _ _ - _ - - - - c. Oroscena (digitate) and

Oroscena carolae (upper Gilbert)

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _

d. Cyrtocapsella tetrapera and Theocyrtis redondoensis (upper Gilbert)

_ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - a. Triceraspyris sp.

(Gilbert b ) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - b. Ommatocampe hughesi and

Cannartiscus marylandicus (Gilbert c)

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

33

distribution of these radiolarian species has been recorded from the “Mio- cene” of California (among other places), the upper limits of which bear no relationship to the Miocene/Pliocene boundary as typified in the Mediterra- nean (see above and discussion in section on Pliocene below).

A radiolarian zonation has been recently established in the North Pacific for approximately the same interval ( -last 3 m.y.) by Hays (1970). A diatom zonation for the Pleistocene of the Antarctic-Subantarctic region was developed by Donahue (1967), the upper Pliocene and Pleistocene sediments of the North Pacific by Donahue (1970), and for the equatorial Pacific by Burckle (1972).

Gartner (1969) suggested the following calcareous nannofossil zonal subdivision of the Pleistocene (from the bottom) : Pseudoemilicnia Zone, Gephyrocapsa Zone, and the Emiliania Zone. On the basis of the presence of Coccolithus doronicoides in the sediments of early Pleistocene age in one subtropical (E21-5) and one subantarctic (E21-17) core as well as elsewhere in tropical and subtropical areas (Bukry, 1970; Bukry and Bramlette, 1970). Geitzenauer (1972, p. 53) proposed a new zone - the Coccolithus doronicoides Zone - which extends from the last occurrence of Discoaster brouweri to the first occurrence of Gephyrocapsa and is equivalent to the lower part of the Pseudoemiliania Zone of Gartner (1969). However, a problem arises in view of the stratigraphic overlap between Gephyrocapsa and Dis- coaster brouweri in late Pleistocene sediments at Le Castella and in deep-sea cores from the Philippine Sea (Takayama, 1970). It may be that the appearance of Gephyrocapsa in the core studied by Geitzenauer was somewhat subsequent to its earlier occurrence elsewhere. Alternatively, there may be a taxonomic problem involved here inasmuch as Gephyrocapsa caribbeanica and Cyclococcolithus doronicoides “are quite similar in size and form, save for the distinguishing bridge of the former species” (Geitzenauer, 1972, p. 48). Coccolithus doronicoides, Cyclococcolithus leptoporus, and Pseudo- emiliania lacunosa (nom. invalid) are all dominant forms in lower Pleistocene sediments so that the zonal relationships appear to require some clarifica- tion. A chronologic framework for the coccolith zonation in the paleoclimatic interpretations is provided by the excess thorium method using gamma- ray spectroscopy (Osmond and Pollard, 1967). A 300,000 (+25%) year datum can thus be recognized and extrapolations are then based upon this level in different cores (Geitzenauer, 1969, 1972; Kennett, 1970; Huddle- stun, 1971). The following data are relevant:

(1) The 300,000-year isochron occurs within the lower part of coccolith “stage” V in three cores and within the upper part of “stage” VI in the other three (Geitzenauer, 1972) indicating an approximate age of 300,000 years B.P. for the coccolith “stage” VI/V boundary.

(2) Coccolith and planktonic foraminifera1 data suggest that coccolith “stage” V can be correlated with the upper part of the V Zone of Ericson and Wollin (1968). This correlation is supported by recent estimates on the age of the V Zone of from slightly greater than 150,000 years to about

-0

-0

LA

TE

E

AR

L

Y L

AT

E

MIO

CE

NE

P

L

I 0

C

E

N

E

P L

E IS

TO

CE

N E

PE

Fig. 6. Interregional correlation and calibration of Pliocene and Pleistocene planktonic zones (se6 text for further explanation).

35

375,000 years (Ericson and Wollin, 1968), and of about 177,000--427,000 years (Briskin and Berggren, 1974) based upon interpolations on the paleomagnetic record. (See also in this respect Broecker and Van Donk, 1970, who have estimated the U/V boundary at about 380,000 years based on extrapolated sedimentation rates.)

boundary (0.7 m.y.) and the mated 0.52-0.58 m.y. B.P. for the Pseudoemiliania lacunosa/Gephyrocapsa boundary. It should be noted that the Gephyrocapsa Zone is defined as the last occurrence of Pseudoemiliania lacunosa to the first occurrence of Emilia- nia huxleyi. The initial appearance of Gephyrocapsa is said to define the Pseudoemiliania lacunosa/Coccolithus doronicoides boundary in the Lower Pleistocene (Geitzenauer, 1972, p. 53, 54). Gartner (1973) has estimated an age of 350,000 years for the Pseudoemiliania lacunosa Datum in tropical areas and suggested that, although P. lacunosa disappears earlier in high latitudes than in low latitudes, the estimate of 0.52--0.58 m.y. may be too old by 100,000--150,000 years. Pseudoemiliania lacunosa disappears just before the temperature minimum of Stage 12 (Emiliani, 1966a,b; 1972) which Emiliani (loc. cit.) estimates is about 290,000 years old. This datum level coincides closely with the disappearance of Stylatrachtus universus (= Styla- trachtus sp. of Hays in Hays et al., 1969). However, more refined dating and correlation of Emiliani’s stages and the zones of Ericson and Wollin based on integrated quantitative paleoclimatologic investigations (oxygen-isotope studies, micropaleontology ) and paleomagnetism, have shown that this level (i.e. P. lacunosa Datum) is closer to 0.45 m.y. (Broecker and Van Donk, 1970; Shackleton and Opdyke, 1973; Briskin and Berggren, 1974). In a similar manner an extrapolated age for the base of the Emiliania huxleyi Zone ranges from 110,000 to 120,000 years B.P, with a mean of 150,000 years. According to McIntyre (1969) the evolution of E. huxleyi from Gephyro- capsa protohuxleyi in the Atlantic Ocean occurred between 250,000 and 220,000 years B.P.

siliceous planktonic zones and their relationship to the paleomagnetic time- scale is shown in Fig. 6. These zones have provided the chronologic framework within which Late Neogene biostratigraphy and paleoclimatic interpretations are discussed below. The boundaries between some of the zones are drawn with a thick rather than a thin line to emphasize the fact that although in nature some boundaries may indeed be relatively instantaneous our analytic methods are incapable of perceiving them as such. It would appear that we now have available a relatively accurately calibrated set of planktonic zones and datum levels for the past 15 m.y.

(3) Extrapolation between the estimated ages of the radiolarian X/+ boundary (0.4-0.5 m.y.) yields an esti-

The inter-regional correlation of these various late Neogene calcareous and

CHAPTER 4

LATE NEOGENE CHRONOSTRATIGRAPHY

LATE MIOCENE

Haug (1911, p. 1607) observed that a great disproportion existed between Miocene and Pliocene, the Miocene representing a considerably larger interval than the Pliocene. He thus distinguished a lower, middle and upper Neogene, the latter corresponding t o the Pliocene. He was of the opinion that these three units were more or less equivalent to the first (lower) three Mediterra- nean stages of Suess.

Late Miocene and Pliocene chronostratigraphical terminology has been the subject of vigorous debate for many years. It is doubtful whether any comparable time-interval in geological history has as many stage names in current use vying for superiority. A detailed analysis of each of these stage names and their affinities is beyond the scope of this study. However, a general discussion is presented below which is aimed at acquainting the reader with the complexities of the problem, and at offering a possible workable and usable scheme for stratigraphical studies in the Late Cenozoic.

The seemingly impossibly complex problem of Late Miocene /Pliocene Age-Stage terminology can be traced t o a single main factor: towards the end of the Miocene a widespread regression occurred throughout the Medi- terranean region. Its effects were felt earliest in the Caspian Sea and Vierna Basin of the Paratethys, north of the Alpine-Dinaride-Tauride orogenic belt in the east and several million years later in the oceanic basin of the western Tethys (proto-Mediterranean Sea) (Van Couvering and Miller, 1971). Ultimately the western Tethys Sea was cut off from the Atlantic Ocean t o the west, and a series of gypsum-anhydrite beds, alternating with clays with marine fossils, was deposited over large areas in the central basin. This major regression resulted in a widespread alteration in the existing geographical relationships between marine, non-marine and intermediate areas. The resulting changes were reflected in highly complex stratigraphical sections, which, when studied in various areas which once formed a part of the great Tethys Sea, proved difficult t o correlate with other (supposedly equivalent) sections in the Tethyan region. In the western Mediterranean, marine sections extend (with a late Miocene interruption) up t o the Pleisto- cene, whereas in the Paratethyan basins of southeastern Europe and southwestern Soviet Union dominantly brackish-limnic sequences characterize the entire Late Neogene. Thus various local stage names have been formulated by

38

46‘

44‘

4 2‘

40‘

38‘

6’ 10. 12. 1 4. 16’ 18.

F R A N C E YUGOSLAVIA

44.

42.

40.

TYRRHENIAN SEA MEDITERRANEAN SEA

38.

4- 6’ 10. 14. 16.

Fig. 7 . Late Cenozoic type localities in Italy.

different workers in attempts to delineate the local stratigraphical succession in the Late Cenozoic of the Mediterranean region. The location of the various Late Neogene stage stratotypes in the Mediterranean is shown in Fig. 7.

The name Tortonian is accepted by nearly all stratigraphers as a useful stage in the Miocene, although it is doubtful whether many could delineate its exact chronostratigraphical limits in regards to paleontological criteria. Mayer-Eymar is intimately associated with the Tortonian, for it was at the Trogen Congress in 1857 that he first defined and used the term (Mayer-Eymar, 1858). His original definition of Tortonian was quite extensive and included:

(1) Marine strata of Tortona (Italy), Baden (Vienna Basin) and Cabrieres d‘ Aigues (Vaucluse, France).

(2) Continental beds with Hipparion at Eppelsheim; beds at Cucurou (Mont-Leberon, Vaucluse, France) and Orignac (Hautes PyrGnGes, France) with the same fauna. (It should be pointed out that Hipparion appears also in mxine Toytsiiian at a k0.rizon cQntxining Nassa mickaudi at hu@,xid (Iskre, France) .)

The biostratigraphy and planktonic foraminiferal fauna of the type (or “syntype”) Tortonian beds in Italy have been described by Gianotti (1953), and more recently by Cita et al. (1965). The foraminiferal mark of the Ita- lian Tortonian are justly famous for their variety and abundance of forms, and have since come to be generally accepted (in lieu of legal machinery) as the primary reference section or stratotype.

39

Cita and Blow (1969) have recognized the upper part of Zone N15 and all of Zone N16 in the (Italian) type Tortonian. The Tortonian and Messinian were said to overlap within Zone N17. Zone N17, and at least part of Zone N18, were said to occur in the stratotype Messinian, whereas N18 and N19 were recognized in the Trubi Marl (basal Pliocene) of Sicily. The base of the Tortonian is placed only slightly earlier than the first evolutionary appearance of Globorotalia acostaensis Blow (= Globigerina globorotaloides Colom) which defines the base of Zone N16.

Mayer-Eymar was in the habit of modifying his concepts of the limits of his Tertiary stage names over the years. The Tortonian was no exception. In 1868 he essentially decapitated his Tortonian Stage and applied the term Messinian for the regressive, brackish-water strata above the marine Torto- nian. (Actually three stage names were created for the Late Miocene-Plio- cene within two to three years of each other: Messinian (Mayer-Eymar, 1858, 1868); Zanclian (Seguenza, 1868); Sarmatian (Suess, 1866).) How- ever, in dividing his Messinian into three part‘s (from the bottom up): Billo- vitz strata, Inzersdorf strata, and Eppelsheim strata (all part of the Paratethys regressive facies in Austria), he considerably weakened the usefulness of this term in regional stratigraphical correlation and can be said to have contrib- uted the ultimate seal of approval to the miscorrelation of the Paratethys regression to that of the central basin. However, he did refer a series of marine units in southern Italy (Messina-Calabria) to his Messinian, which allows a more objective comparison and evaluation.

Messina, near Gesso, southern Italy, Selli (1960) selected and described a neostratotype section for the Messinian Stage of Mayer-Eymar, the Pas- quasfa-Capodarso series between Enna and Caltanissetta, in central Sicily. The sequence is bounded below by marls of Tortonian age and above by the “trubi” generally accepted as representing the basal Pliocene (= Zanclian) in this area. In its original concept the Messinian of the Messina area included the diatomaceous marls (“tripoli”), evaporites, and overlying “trubi” (deep- sea lutites with rich planktonic faunas). Seguenza (1868, 1879) restricted the Messinian to the evaporitic series and the “tripoli”. He considered the overlying unit, the “trubi” to be Pliocene, and introduced the term Zanclian for the calcareous-marly sediments with uniform fossil content and lithology up to, but excluding, the Amphistegina-bearing sands. According to Seguenza the Zanclian represents a distinct interval between the Tortonian and the Pliocene (Astian-Piacenzian), and he placed it at the base of the Pliocene, which he believed could be subdivided into three parts: Zanclian, Piacenzian, Astian. (It should be remembered that at the time Seguenza introduced the term Zanclian, the Tortonian of Mayer-Eymar extended up to the base of the Piacenzian, the latter of which Mayer-Eymar considered to be basal Plio- cene.) Mayer-Eymar was of the opinion that the strata attributed by Seguenza (1868) to his Zanclian were of the same age as his own Messinian; indeed, finding that the name Zanclian (from Zancla, the name of the Greek colony

Because of tectonic and sedimentological complications in the vicinity of

40

of Messina before the Roman conquest) was too classic for his taste, he essentially replaced it with his own Messinian. As a result the use of the term had led to confusion as it has been applied variously to the gypsum- anhydrite series (“gessoso-solfifera”) of the central basin which marine biostratigraphers unanimously consider late Miocene in age and. to the deposits of “Pontian” = Sarmatian age of the Paratethys; which most continental biostratigraphers consider early Pliocene in age.

The main shortcoming of the Messinian Stage is its general lack of fossils, these being restricted primarily to the marine marly intercalations between the gypsum-anhydrite beds. Thus a distinct paleontological definition of the stage appears difficult. However, as Selli (1960) points out, the chronostratigraphical limits of the stage can probably best be set by the paleontological character of the adjacent Tortonian and Piacenzian s.1. (i.e. Tabianian or Zanclian), as in the neostratotype.

The planktonic foraminifera1 fauna from the type section of the Messinian has been studied by D’Onofrio (1964) and Colalongo (1970). The presence of the Globorotalia plesiotumida Zone (N17) in the Messinian of Italy was pointed out by Blow (1969) and documented more fully by Catalan0 and Sprovieri (1969,1971), and Colalongo (1970) (see also Cita, 1974).

Depbret (1895) proposed that the combined Helvetian and Tortonian stages should be collectively referred to as Vindobonian, which he equated with Suess’s second Mediterranean Stage, below the Sarmatian. He was of the opinion that although in northern Italy, southern France and the Vienna Basin it was possible to distinguish the Helvetian and Tortonian, they were actually facies, partially superimposed upon each other, rather than distinct time-sequential units. DepQet excluded the Sarmatian from his Vindobonian and Mayer-Eymar consistently interpreted his Messinian Stage as equivalent to the Sarmatian-Pontonian, equating it with the Late Miocene. Thus the Helvetian-Tortonian interval has generally been regarded as synonymous with Middle Miocene.

The terms Lower, Middle, Upper, or Early, Middle, Late are however merely conventions, based on the propensity of scientists to categorize things and events into threefold subdivisions on a linear scale. This accession to convention has actually little t o do with the geological and faunal evidence upon which subdivisions of geological time should be more properly based. The Vindobonian has, to this day, not been stratotypified, so that its use as a stage term in marine sequences is to be discouraged. (It has a completely independent definition in mammalian biostratigraphy .) However, the Torto- nian and Messinian are closely linked with one another both in terms of sedimentology and fauna, so that inclusion of these two stages in the Late Miocene would be more appropriate than a distinction between Middle and Late Miocene being made between the two stages.

to Messinian, the Pontian, Sahelian, Sarmatian, and Andalusian. Pareto (1865) also introduced the Serravallian as a Late Miocene Stage because he consid-

Names commonly used for supposedly Late Miocene strata are, in addition

41

ered the Tortonian to be Pliocene. We discuss the Andalusian first because of its supposed direct relationship with the Messinian.

In an attempt to find a more fully marine section of late Miocene age, equivalent to the Messinian, Perconig (1964) proposed the Andalusian Stage (see review by Verdenius, 1970). Planktonic foraminifera of the Andalusian Stage have been described by Perconig (1968) and Tjalsma (1971).

scale has been reviewed by Verdenius (1970). He concluded that there is no sound basis for correlating its base with the Tortonian and the remainder with the Messinian (Perconig, 1968). Although it is difficult to ascertain exact stratigraphic relationships from the literature, it would appear that some conclusions can be drawn on the basis of Perconig’s (1968) data. The data are of two types - presence and absence.

( 1 ) Presence. The presence of Globorotalia margaritae throughout the Andalusian (Perconig, 1968, table 8) and the suggestion that G. margaritae is present in sediments of Tortonian age (Perconig, 1968, p. 207, table 5) are inconsistent with known facts. G. margaritae appears within Late Miocene Zone N17 between 5.5 and 6 m.y. ago according to our data in the North Atlantic (see also Saito et al., 1974). It has not been recorded, to our knowledge, in upper Miocene strata in the Mediterranean Basin. It is, essentially, an early Pliocene species. Its presence throughout the Andalusian would suggest that the latter is of late Miocene and early Pliocene age.

The appearance of Globorotalia margaritae in the uppermost part of the stratigraphic section (Porcuna Formation) studied in the eastern Guadalquivir Basin (Tjalsma, 1971) and in the middle and upper parts of the Ecija Forma- tion, Carmona section, in the western Guadalquivir Basin (Verdenius, 1970) allow direct correlation of these beds and substantiate the view that the Andalusian Stage is latest Miocene and/or early Pliocene in age.

(2) Absence. Although less definitive, the conspicuous absence of Globo- quadrina dehiscens and Glo borotalia crassaformis also suggests the Andalusian is of early Pliocene age. G. dehiscens became extinct at about 5 m.y. ago, near the Miocene / Pliocene boundary, and G. crassaformis appeared at about 3.5 m.y. ago, i.e. within the Early Pliocene (Hays et al., 1969; Berggren, 1972b). From Perconig’s (1968) data it would seem that the Andalusian probably lies within these limits.

More precise limits may be placed on the Andalusian Stage on the basis of investigations of Martini (1971, and in Perconig and Granados, 1973). The lowest part (base of the gray marls below the “Caliza Tosca”) of the Andalu- sian belongs to the upper part of the Discoaster calcaris (NN 10) Zone. The bulk of the marls below the “Caliza Tosca” belong to the Discoaster quinqueramus (“11) Zone and both the “Caliza Tosca” itself and the basal Pliocene marls immediately above it are correlated with the Cera tolithus tricorniculatus (“12) Zone. The Andalusian thus spans the interval of late Zone N N l O to within Zone “12.

As we have seen above an estimate of 10.0 m.y. is made for the N N 1 0 / 1 1

The position of the Andalusian Stage in the late Cenozoic time-stratigraphic

42

boundary, 5.7 m.y. for the NN11/12 boundary and 4.4 m.y. for the NN 12 / 13 boundary. The Andalusian thus apparently spans the time interval of >10 m.y. to >4 m.y. In its lower part it is therefore the time as well as the paleontological equivalent of the Tortonian Stage of Italy, because as we have seen above, the Tortonian/Messinian boundary is at about 6.5-7 m.y. Furthermore, the Andalusian-equivalent strata of Murcia contain Turolian and Ruscinian mammal faunas (Montenat and Crusafont, 1970) which are independently calibrated to the same time-interval (see below). Its upper limit is probably the direct equivalent of the Messinian/Zanclian (i.e. Miocene/ Pliocene) boundary in the Mediterranean for which an age of 5 m.y. has been estimated (Berggren, 1973, 1974). In short the Andalusian corresponds to a part of the Tortonian and all of the Messinian (see Appendix, note 2).

The Sahelian (Pomel, 1858) was defined in Algeria where it was believed to be the only area in the western Mediterranean where the Late Miocene was represented by marine strata. Haug (1911) championed its use in his Treatise, but the term has been little used in regional stratigraphy. Ruggieri et al. (1969) have resurrected the name Sahelian as a lower substage of the Messinian, and Catalan0 and Sprovieri (1969) report a marked faunistic change at the base of the “Sahelian” near Enna (Sicily) which they attribute to Zone N17. However, the planktonic foraminifera of the classic reference sections of the Sahelian at Carnot, northern Algeria (Brives, 1894,1895, 1896,1897) have recently been studied by Tjalsma and Wonders (1972) who assign them to the upper part of Zone N16 (= Tortonian). The term Sahelian is inadequate and indeed quite worthless as a chronostratigraphic unit.

The term Sarmatian was introduced by Suess (1866, p. 232) for the “Ceri- thien-Schichten” and “Hernalser-Tegel” of the Vienna Basin, although the name was derived from southeastern Russia and the brackish-marine molluscan faunas of that area were cited as typical. The Vienna Basin Sarmatian is younger than the Vindobonian or Suess’s Mediterranean Stage I1 (but see Haug, 1911, p.1608, who included the Sarmatian at the top of his threefold subdivision of the Vindobonian), which it overlies, and older than the succeeding beds assigned to the Pliocene. However, the equally typical Sarmatian sections in eastern Europe cannot in all cases be satisfactorily correlated with central European (Vienna Basin) sequences. Thus, various interpretations have developed regarding the identity and affinities of the Sarmatian Stage. Renevier (1897) interpreted the Sarmatian as a brackish-water facies of the Tortonian, and in this he appears to have been followed by Gignoux (1950). Munier-Chalmas and De Lapparent (1893), Depkret (1893,1895), Janoschek (1951), and Glaessner (1953) considered the Sarmatian as definitely younger than the marine Tortonian. The stratotypes of the two stages are indeed of different ages. However, recent investigations suggest that the Sarmatian-age mammalian faunas and volcanic rocks are older, at least in part, than the type Torto- nian and are thus of middle Miocene age (Van Couvering and Miller, 1971).

Andrusov (1899) divided the Sarmatian into a lower, middle and upper part; these were in turn, subsequently named Volhynian, Bessarabian and Chersonian by Simionescu (1903).

43

As early as 1913 Winkler-Hermaden expressed the opinion that the Lower Sarmatian (Volhynian) of Russia corresponded to the Lower and Middle Sarmatian of Steiermark (Austria), and that the Middle and Upper Sarmatian of Russia (Nubeculuriu horizon of Bessarabia) corresponded to the Upper Sarmatian of Steiermark. This interpretation was modified by comparative studies of various investigators (cited in Winkler-Hermaden, 1960, p.227) in which the Sarmatian of the Vienna Basin area is shown to correspond only to the Lower and Middle Sarmatian of the euxinic basins in the eastern Paratethys region, and the Upper Sarmatian of the Vienna Basin corresponds to the Maetoian of the eastern Paratethys and the Lower Pannonian in the Pannonian Basin of Hungary.

The Pontian Stage was introduced by Le Play (1842) for limestone strata near Odessa, Novocherkassk and Taganrog. Barbot de Marny (in Suess, 1866) restricted the term Pontian to include only the former two, since the out- crops at the latter locality had been shown to be of Sarmatian age. The Pon- tian was thought to contain beds transitional between the Miocene and Plio- cene in southern Russia, on the basis of the percentage of living species in the highly endemic brackish-limnic molluscan fauna (Lyell, 1865), but accurate correlation has not been possible. In the last century it was common practice to correlate the Pontian with the marine Piacenzian of Italy which lies above the Messinian gypsums and shales. Variation in its age assignment ranges from its almost unanimous inclusion in the Pliocene in eastern Europe (Soviet Union) to upper Miocene in Western Europe. Whereas Gignoux (1950) placed the Pontian above the Tortonian, Munier-Chalmas and De Lapparent (1893) earlier placed it above the Sarmatian; Laffitte (1948), on the contrary, even correlated it with the Tortonian.

Winkler-Hermaden (1924) suggested that the marine Piacenzian of the Mediterranean region corresponds to the thinner and less widely developed Dacian Stage (= Dazische stufe) of the Vienna Basin, and that the Pontian Stage (sensu lato) and the Piacenzian are broadly correlatable (see also Winkler-Hermaden, 1943, 1952,1960). In a recent study of the Late Neo- gene deposits of the Ponto-Caspian Basin the Maetoian (Lower Pontian s.1.) is assigned to the Miocene, and the Miocene/Pliocene boundary is drawn between the Maetoian/Pontian (sensu stricto) stages (Nevesskaja and Iljina, 1969). Papp (1969) has correlated the Pontian Stage ( s . s . ) with the Upper Pannonian and Upper Messinian and thus placed the Miocene/Pliocene boundary at the top of the Pontian S . S .

It should be pointed out here, perhaps, that the Pontian has been used in two different ways by authors: Pontian sensu lato and Pontian sensu stricto. The Pontian sensu lato includes the original Lower Pontian, which is approximately equivalent to the Vallesian land mammal age of European stratigraphers (Crusafont, 1950); it is referred to as Upper Sarmatian and Maetoian in the eastern Mediterranean region by Soviet geologists and is middle Miocene (mainly pre-Tortonian) in age. The Pontian sensu stricto is the rough equivalent of the Turolian land mammal stage of European verte-

44

brate paleontologists (Crusafont, 1964) and to the Pannonian of the Hun- garian Plain and southeastern Europe which is of late Miocene age. It was this latter part of the Pontian that was stratotypified by Stevanovic (1964). The occurrence of Hipparion was cited by Matthew (1929) from the lower part of the “Pontian Stage” in its original sense; this level probably represents a time earlier than 9 m.y. (Samos and Pikermi faunas), and as Van Couvering and Miller (1971, p. 562) conclude, “Matthew’s proposition, made for the convenience of workers in the Siwaliks, that the appearance of Hipparion should mark the beginning of the Pliocene in Eurasian land mammal biochronology is invalid and should be abandoned.”.

Europe and the Mediterranean area culminated in drastic sea-level lowering and evaporation in the western remnant of the once extensive Tethys Sea. A t the Miocene Colloquium (1958) in France it was suggested that the base of the Pliocene be drawn at the first appearance of the Pliocene marine transgression. The Miocene would thus include all continental and brackish-limnic beds which characterize the end of the Late Miocene sedimentary cycle in the Mediterranean Basin itself. Winkler-Hermaden (1960) has pointed out that if the Piacenzian is considered equivalent to the Pontian Stage ( s . s . ) of Russia (= Upper Pannonian of the Pannonian Basin), which begins with a regional transgression of shallow brackish marine seas, then correlation can be carried across the Mediterranean area into the Paratethys. However, as we have seen above, the Pontian Stage S . S . is primarily, if not entirely, of Miocene age and much older than Piacenzian.

to designate time-stratigraphical units represent little more than rock-stratigraphical entities in local areas. From the point of view of chronostratigraphy most of the terms suggested as stage names in the Late Miocene-Early Pliocene are unsatisfactory. They are represented by brackish- or fresh-water sequences with restricted faunas, which do not allow satisfactory correlations outside the immediate area of their development.

Thus Denizot (1957) suggested a return to the original concept of Torto- nian of Mayer-Eymar (1858), i.e. extending Tortonian up to the base of the Piacenzian s.1. (or Tabianian-Zanclian). Selli (1960, footnote p. 313) object- ed to this on the grounds that the Tortonian would then represent the time interval from Helvetian to Piacenzian s.l., and would include perhaps the major part of the Miocene, particularly since Drooger, Papp and others tend to place the base of the Tortonian approximately at the first appearance of Orbulina. This argument has been rendered superfluous to some extent by the recent realization that Helvetian and Tortonian are not closely related from the point of view of time. The Helvetian would appear to be pre- Orbulina in age (pre-Zone N9) and broadly correlatable with (at least a part of) the Burdigalian, whereas the base of the Tortonian is within Zone N15. Thus the Late Miocene (including the Tortonian) would include part of Zone N15, and Zones N16 and N17 - a relatively short interval of time (about

The widespread regression which characterizes the Late Miocene in

It is more than likely that most of the various terms which have been used

45

5 m.y.) with respect to the remainder of the Miocene, and slightly more than the duration of the Pliocene. Banner and Blow (1965a, p. 1165) and Cita and Blow (1969) report that both the Upper Tortonian and at least a part of the Messinian belong in Zone N17, and thus may overlap in time. The Mes- sinian, or latest Miocene, is then a relatively short interval of time (7-5 m.y.), approximately equivalent to less than a single planktonic foraminiferal zone, N17. The Messinian is a very distinct facies everywhere in the Mediterranean, and if it is actually associated with a major Antarctic glacial, advance (see below) should be distinct (i.e. recognizable) worldwide. The suggestion by Denizot (1957) that we return to Mayer-Eymar’s original concept of Tortonian (which included strata now placed in Tortonian and Messinian) is therefore judged unwise.

PLIOCENE

Lyell (1833) proposed the term Pliocene for the youngest Tertiary faunas which he recognized at the time, all of which in the Mediterranean Basin lay above the regional unconformity at the top of the Miocene sequence. He divided his Pliocene into an “Older” Pliocene (based on more than 50% species still living), which would correspond in general to the Astian- Piacenzian as it is now defined in Italy, and a “Newer” Pliocene (with 90-95% of species still living), for which he subsequently (Lyell, 1839) introduced the term Pleistocene. Although he later suggested abandonment of the term Pleistocene in favour of post-Tertiary, or post-Pliocene, the Pleistocene has become firmly rooted in stratigraphical literature and is used as a virtual synonym of Quaternary.

An illustration of the lack of understanding of basic stratigraphic principles in defining chronostratigraphic boundaries is furnished by Alt (1968, p. 90-91) in an interesting discussion on Late Neogene marine terraces on the eastern coast of the United States. Alt (1968) suggested that the Pliocene Epoch be redefined,

“as the period during which the emergent shoreline at an elevation of 90-100 ft . (the Surrey scarp) was occupied by the sea.”

The abandonment of the Surrey scarp is considered

“almost certainly a glacial-eustatic event probably marking the onset of Northern Hemisphere glaciation. Chief among these is the fact that both the upper and lower boundaries of the Pliocene so defined would coincide with eustatic fluctuations of sea level, the effects of which have already been independently recognized on several continents. Local marine sections could be identified on each continent as the deposits underlying the marine terrace associated with the 90-100 ft. shoreline. The boundaries of the Pliocene so defined could thus be unequivocably correlated from continent to continent with considerable confidence in their absolute time equivalence. This would

46

eliminate complstely the problems involved in attempting to carry boundaries based‘on fossil correlations from one continent to another.”

The geological history of our planet has been classically subdivided into discrete chronostratigraphic units based on the irreversibility of evolutionary processes in marine organisms. Because of the greater continuity of marine sedimentary processes, and the ubiquity and abundance of marine organisms, the classical subdivisions of geological history (including those of Lyell) have been based upon changes in the marine fossil record. Thus, Alt’s suggestion to redefine the Pliocene Epoch as the period of time during which an eastern North American emergent shoreline was occupied by the sea would bear no relationship to the way the word Pliocene was used by Lyell, even if it should prove to represent the same timeinterval. As a matter of fact, if the upper limit of the Pliocene were to be defined by the age of the abandonment of the Surrey scarp and if this corresponded to the major refrigeration increment at about 1.5 m.y. (see Chapter 7 on paleoclimatology below), it would be within the Early Pleistocene as presently defined by paleontological and radiometric techniques (Berggren, 1972a, b). If nothing else, the identification of a marine terrace surface uniquely associated with this 90-100 f t . sea-level stand would be an insuperably difficult problem in some areas of the world and would inevitably end in mainly paleontological arguments, thus negating whatever advantages the scheme might be said to have. Furthermore, the initiation (if not the end) of sea-level stand at this elevation would be extremely difficult to determine as a chronologic event. Alt (1968) points out that a further advantage of defining the boundaries of the Pliocene on the basis of eustatic fluctuations of sea-level would be that the boundaries defined in this manner would probably coincide with climatic changes, but as we have stressed above and elsewhere (Hays and Berggren, 1971), chronostratigraphic boundaries cannot be based upon climatic changes, for these may be local or regional or diachronous in different hemispheres, and often are not reflected in the rock record. Furthermore, they depend fundamentally upon observed changes in the fossil fauna ultimately in order to be placed within an ordinal time-scale.

The term “deposit0 subappenino” was used by Huot (1837) to characterize the late Tertiary clays and mark in the hilly regions in the vicinity of Castell’Arquato. This concept was based on earlier work in this area by Brocchi, who in 1814 described the calcareous sands and mark and their molluscan faunas in this region. D’Orbigny (1852) elevated the term “Sub- apennin” to be the youngest stage of the Tertiary, essentially substituting it for the Older Pliocene of Lyell.

created the Astian Stage for the yellow, calcareous sands of Asti, in northern Italy. Mayer-Eymar (1858) proposed the name Piacenzian for the Pliocene - originally as a substage of the Astian. Seguenza introduced the term Zanclian for Lower Pliocene in 1868, and distinguished three stages in the Pliocene:

The subdivision of the Pliocene began with De Rouville (1853), who

41

Zanclian, Astian and Sicilian. The Tabianian was introduced by Mayer- Eymar (1868) for the Lower Pliocene.

fold subdivision of their Pliocene succession on the basis of the micro- and macrofauna, Di Stefani (1893) demonstrated in the latter part of the last century that the faunal differences between Astian and Piacenzian are due to facies differences. Gignoux (1913, p. 13) discussed the concept of the sedimentary cycle and the relationships of successive cycles to geological time. This was subsequently to form an integral part of his later stratigraphical thinking and was formulated clearly in his textbook on Stratigraphic Geolo- gy (English edition, 1955), where he characterized the stage as “a stratigraphic synthesis”. For Gignoux the Pliocene was just a sedimentary cycle, which he believed possessed a stratigraphical unity and paleontological identity. He suggested that the Pliocene be divided on the basis of its paleontological characters into a lower part, Plaisancian and Astian, and an upper part, for which he had earlier (1910) proposed the name Calabrian (see also Migliorini, 1950, who interpreted the Astian--Plaisancian as facies rather than “true stages”).

These then are the units which stratigraphers have utilized in their attempts to subdivide the Pliocene of the western Mediterranean region (Italy, southern France). In the section below we shall consider in rather more detail the relationships which these units bear to one another.

Mayer-Eymar (1868) introduced the Tabianian for the blue marls in the locality of Chiesa Nuova at Tabiano Bagni, near Salsomaggiore, in northern Italy as the lowest stage of the Pliocene. The Tabianian was designated as “Tabbianer Schichten” in 1874, as “Tabbianin” in 1884, and as “Tabbianon” in 1888. This stage has been studied by Iaccarino (1967) who defined and discussed the stratotype section. Stratigraphic continuity was proposed between the Messinian and the Tabianian with a basal conglomerate denoting the base of the Tabianian (and the Pliocene transgression). The blue and gray clays of the type Tabianian which are about 130 m thick pass gradually into the overlying Piacenzian blue marls. The initial appearance of Globorotalia puncticulata was noted about 50 m above the base of the Tabianian. In deep-sea cores the initial appearance of G. puncticulata is recorded at about 4.5 m.y. (within the Early Pliocene).

stratotype locality has been proposed by Cita (1974) for the Zanclian Stage at Capo Rosello near Agrigento on the south coast of Sicily. She indicates that the lithostratigraphic unit of the Trubi marl extends downwards beyond the chronologic limits of the Tabianian Stage (see Blow, 1969, who indicated that the Tabianian was older). The base of the stratotype locality corresponds to the beginning of permanent open marine conditions in the Medi- terranean following the late Miocene conditions of restricted (evaporite) deposition. I t is aptly referred to by Cita (1974) as the “birth of the Neo-

Whereas most Italian geologists have recognized either a twofold or three-

As we have seen the Zanclian Stage was introduced by Seguenza (1868). A

48

Mediterranean”. This level occurs within the Ceratolithus tricorniculatus Zone and at the base of the Sphaeroidinellopsis Acme Zone as used by Medi- terranean geologists.

The microfauna of the type level of the Zanclian - the Trubi marl and its equivalents - is, in Silicy, rich and diversified, and consists to a large extent of planktonic foraminifera and other pelagic organisms (nannofossils). Corre- lation with tropical microplanktonic faunas can be achieved by comparison with the Zanclian, whereas the shallower-water deposits of the Tabianian do not contain the diversity of faunal elements essential for close intercontinental stratigraphical Correlation. Indeed, it was on the basis of Zanclian faunas that Banner and Blow (1965b) were able to recognize their Zone N18 in the Mediterranean area and were able to demonstrate a partial overlap between the Piacenzian and Zanclian stages. The development of Sphaeroidinella dehiscens from Sphaeroidinellopsis subdehiscens, as well as the overlap of Globigerina nepenthes and S. dehiscens in the lower part of the “Trubi” of Sicily, has been considered characteristic of the Lower Pliocene (W.H. Blow, personal communication, 1968; Parker, 1967; cf. Blow, 1969, p. 196). For this reason we prefer the use of Zanclian for the Lower Pliocene rather than Tabianian.

Ruggieri and Selli (1950) have suggested a threefold subdivision of the Pliocene with the Lower Pliocene said to correspond with the Tabianian of Doderlein (1872) and Cocconi (1873). The Middle Pliocene was said to be typically developed in the argillaceous facies at Santa Maria Maddalena (several kilometers west of Castell’ Arquato on the road to Lugagnano), and the Upper Pliocene was said to be typically developed in the calcareous-- arenaceous facies along the Rio Rorzo (near Castell’ Arquato). A general threefold subdivision of the Pliocene is common to the works of most Italian paleontologists (Barbieri, 1952; Conato, 1952; Di Napoli, 1952; Perconig, 1952, 1955; Gianotti, 1953; Martinis, 1954; Papani and Pelosio, 1962; Pezzani, 1963).

The Astian was erected by De Rouville (1853, p. 185) to replace the Subapennin of Huot (1829,1837) and D’Orbigny (1852), which corresponded exactly to the Older Pliocene of Lyell. The Astian was created for the sandy facies which interfingers with the Piacenzian clays and marls in northern and central Italy. The type locality is in the Valle Andona, about 6 km west of Asti. The Astian is developed here as a sequence of calcareous sands, shell conglomerates, oyster beds and soft marls, about 30-40 m thick. Shallow-water foraminifera characterize the lower part of the type section, with marine benthonic forms common at higher levels. Planktonic forms are generally a minor constituent in these faunas and consist primarily of species of Globigerina and Globigerinoides.

The Piacenzian was introduced by Mayer-Eymar (1858) (as the “Piacenzi- sche Stufe”) for the argillaceous facies of the Lower Pliocene with Nassa semistriata in northern Italy. It was originally distinguished as a substage of the Astian (of De Rouville, 1853) and subsequently erected as a stage by

49

Renevier (1897). Pareto (1865) adopted the term, utilizing the French equivalent Plaisancian, and specified that the typical development of the stage was to be seen in the hills around Castell’ Arquato.

In an otherwise informative discussion of the Pliocene stratigraphy in southern and southwestern Europe, Movius (1949, footnote pp. 381, 382) erroneously attributed the type locality of the Piacenzian Stage t o “Plaisance ... a small village on the road between Castell’ Arquato and Lugagnano”. On the contrary, the term Piacenzian was taken by Mayer-Eymar from the city (not village) of Piacenza (which lies approximately midway between Parma and Milan, and only a few miles from Salsomaggiore, near which the type Tabianian can be seen). Contrary to Movius’ statement that the Piacenzian, erroneously attributed to Gignoux (1913), “never came inio general use”, Piacenzian is widely used in Italy. I t is only correct that this name be used, since the type locality and most typical development of the sediments of this age are to be seen in north-central Italy. As pointed out above, the term Plaisancian is merely the French translation of the word Piacenza, and Pareto (1865) appears to have been the first t o use this term.

The foraminifera of the Castell’ Arquato region were described by D’Or- bigny in 1826, and Malagoli (1892) and more recently by Di Napoli (1952), Barbieri and Medioli (1964a,b) and Barbieri (1967).

The type Piacenzian lies conformably upon the Vernasca Sand (Tabianian) and consists of marl and claystones (the typical blue clays of the Piacenzian), which grade upward into sandy and silty claystones, heralding the Astian facies. Marine benthos and benthonic foraminiferal faunas are abundant in these blue mark; faunal lists from localities in the Lower, Middle and Upper Pliocene in the Castell’ Arquato- Lugagnano- Vernasca area were given by Barbieri in the guidebook to the 4th International Micropaleontological Colloquium in Italy (1958; see also Barbieri, 1967). The Piacenzian/Tabian- ian boundary was shown to correspond at the type locality of the Piacenzian to the disappearance of Globorotalia hirsuta auct. (= G. margaritae Bolli and Bermudez) and can be extended into the type locality of the Tabianian Stage (Iaccarino, 1967). This is an important datum level within the Pliocene and corresponds approximately to the Gauss/Gilbert Magnetic boundary (- 3.3 m.y.). Thus, a twofold subdivision of the Pliocene into an early and late part may be conveniently placed at about 3.3 m.y. (see Figs. 1,2,4-6). Jenkins (1964) presented a list of planktonic foraminiferal species observed in samples from the “Upper Pliocene” of Castell’ Arquato. Globorotalia cf. G. hirsuta, G. scitula and G. inflata were among the forms cited but G. menardii was not found, nor were G. fistulosus or Discoaster spp.

Both Depbret and Gignoux have suggested that the Astian-Piacenzian be grouped into a single “Older Pliocene” - a sandy and an argillaceous facies - the term Astian being suitable for this combination. On the other hand De Lapparent and Haug accepted the two stages as being distinguished by Rene- vier: Astien (De Rouville, 1853); Piacenzian (Mayer-Eymar, 1858).

Ruggieri (1961) proposed the following biostratigrsphical zonation of the

50

Pliocene and Pleistocene in Italy: a lower zone (without Cyprina islandica) with two subzones - A. Globorotalia hirsuta, B. absence of both forms, which corresponds to Gignoux’s (1913) “Pliocene antico”; and an upper zone (with Cyprina islandica) with two subzones - C. (without Anomalina balthica), D. (with A. balthica). The base of D was considered to be the base of the Pleistocene (see also Ruggieri and Selli, 1950). This type of zonation, though of interest in terms of local stratigraphical studies, can hardly be applied outside the immediate area of investigation as already pointed out by Banner and Eames (1966). Banner and Blow (1965b) subdivided the Plio- cene into three planktonic foraminifera1 zones: N18-N21; and Berggren (1973) has recently proposed a sixfold zonation of the Pliocene: zones P1-P6 (see discussion below in section on Pliocene Epoch and the Miocene/ Pliocene boundary). The Pliocene is actually a relatively short period of time (- 3 m.y.).

THE PLEISTOCENE

In 1833 Sir Charles Lye11 introduced the term Pliocene Period as the youngest division of the Tertiary and subdivided it, on the basis of the percentages of molluscan species still living, into two parts:

Newer Pliocene : 90-95 9% still living. Older Pliocene : more than 50 9% still living.

He subsequently introduced the term Pleistocene as a substitute for Newer Pliocene, although he later recommended that the term Pleistocene be abandoned.

In 1824 Marcel de Serres (in Ckeuze de Lesser, 1824, p. 174) discussed the presence in France of relatively recent calcareous sediments which he attributed to his “quatrieme formation d’eau douce”. He later (1855) claimed, somewhat unjustly it would appear, priority for the creation of the term Quaternary. He had used the term “Quaternaire” in 1830 (a year after its introduction by Desnoyers) in considering it a synonym of Diluvium.

of Touraine and Languedoc than in the Seine Basin. He proposed the term Quaternaire or “Tertiaire r6cent” for marine deposits younger than those in the Seine Basin and divided it into three parts (from the top):

Desnoyers (1829) suggested that the seas had remained longer in the region

(3) Rbcent. (2) Diluvium. (1) Faluns de Touraine, molasse Suisse, Pliocene marin de Languedoc.

We see, thus, that the Quaternary originally included a heterogeneous assemblage of rocks (even within a given subdivision such as the lower one) and corresponded, essentially, to the Neogene of Hoernes and later authors in its general application.

language and literature and subsequently (1858) modified it to “Quart’~en”, a translation of the term “Quartaire” which he had proposed that same year.

Morlot (1854) introduced the word “Quaternaren” into the German

51

Marcel de Serres (1830) had affirmed that early man was contemporaneous with these deposits of the Quaternary. Thus within a year of its creation by Desnoyers, De Serres had considerably modified the concept of Quater- nary to that form in which it was subsequently used by most stratigraphers. The classic treatises of Reboul (1833) and D’Archaic (1849), in which the “Terrain Quaternaire ou diluvien” is discussed (in 439 pages) gave added weight to the general acceptance of the Quaternary in geological literature. The term “Pliosthe” was introduced by Buteux (1843) as a brief epoch of approximately post-Pliocene time to denote the Quaternary, but it is a superfluous term.

geologists working on continental and non-marine section, whereas Pleisto- cene has generally been applied to corresponding marine deposits. More accurately the Quaternary designates a period in the chronostratigraphical scale of the Cenozoic, which includes the Pleistocene and Holocene Epoch/ Series. Distinction between the latter two is rather artificial, as the Holocene, is but an interglacial event in which we are still living. It might be more appropriate to speak of the Pleistocene Epoch as the youngest (and current) Epoch of the Cenozoic Period and Cenozoic Era.

Modifications of the term Quaternary were suggested from the beginning. Anthropological investigations yielded recurring evidence of the relationship of early hominids in the Quaternary and among the various terms proposed we can recall the following: “Pbriode anthropbenne” (Reboul, 1833), “Pbriode Lamozoi’que” (Vezian, 1865). “Terrain humain” (De Mercey, 1874-1877), “temps anthropiques primitifs” (Piette, 1880), “Psychozoi- que” (Le Conte, 1887, in Meunier, 1908). At the first International Geologi- cal Congress in London (1888) Gaudry, with the approval of Prestwich and De Lapparent, made the proposal that Man (represented by his artifacts in particular) was the characteristic element of the Quaternary, which justified the creation of a geological period distinct from the Tertiary period. Thus, some geologists (particularly those working with anthropological sources) would define the base of the Quaternary at the level of first human industry. In Europe, however, these earliest levels ( Abbevillian) are considerably younger (ca. 0.5 m.y.) than the earliest artifact levels discovered in East Africa (Tanzania; ca. 2 m.y. or older) (Howell, 1972; Isaac, 1972), and essentially man-like fossils appear to go back even further in time (R. Leakey, 1971).

At the Warsaw Congress of 1961 (Grichuk et al., 1965) strong emphasis was placed upon the initial thermal decrease which led to an irreversible change in the flora as a distinctive criterion in determining the base of the Pleistocene. However, as pointed out elsewhere in this article, such a criterion must be used with caution.

Among vertebrate paleontologists it has been common to consider the base of the Quaternary at the base of the Villafranchian Stage on the basis of the occurrence in this Stage of Equus, Bos (Leptobos) and Elephas. The

In general, it appears that the term Quaternary has been used primarily by

52

inclusion of the Villafranchian within the Quaternary was first made by Julien in 1869, ad‘opted by Haug (1900) in his Treatise, and more recently by the Temporary Commission on the Pliocene/Pleistocene Boundary at the 18th International Geological Congress in London in 1948. However, recent data have shown that the lower (type) Villafranchian at Villafranca d’Asti (see Fig. 7) is older than most deposits ascribed t o it elsewhere and that none of the characterizing genera listed above, save perhaps Leptobos, have been identified with certainty from the type Villafranchian (see Chapter 6 on mammalian biostratigraphy).

The original concept of a glacial epoch came from Switzerland where Agassiz had boldly suggested in 1837 that the European continent had been recently covered by ice. In 1846 Forbes redefined the Pleistocene of Lyell (1839) as equivalent t o this “glacial epoch”, i.e. “the time distinguished by severe climatal conditions through a great part of the northern hemisphere” (p. 402). Much of the evidence for this supposed equivalence is physical in nature (glacial drift, etc.) and some if faunal (first appearance of boreal marine and terrestrial taxa in the lower Pleistocene of England and southern Europe). However, definitive correlations between these various occurrences have not been possible to date and the evidence cited elsewhere in this text should help to dispel the prevalent misconception that onset of glaciation and the base of the Pleistocene are synchronous.

Most early glaciologists refused to believe in more than a single glaciation, but the pioneering work of Penck and Bruckner (1909) in the Swiss Alps led to the recognition in the Danube Basin of the four classic glaciations: Gunz, Mindel, Riss, and Wurm. It is still generally accepted among glaciologists and many stratigraphers that the base of the Quaternary coincides approximately with the earliest glacial deposits, although it is becoming more widely recognized that pre-Gunz glacial deposits occur in Europe and elsewhere.

Base of the Pleistocene

Determination of the base of the Quaternary (Pleistocene) in marine sedimentary sections using paleontological criteria has been a goal which has eluded stratigraphers for many years. The criteria suggested are numerous, but correlation with the stratotype section in southern Italy has proved difficult.

the Calabrian Stage of Gignoux (1910). Gignoux (1910, p. 841) introduced his Calabrian Stage for the post-Pliocene formations in Italy (characterized by the appearance of Cyprina islandica), which precede the Sicilian, overlie the Astian, and are equivalent to the Newer Pliocene of Lyell. As localities where the stratigraphical succession was said to be typical of the Calabrian he cited the following (1913, p. 2 3 ) : “collines toscanes h Vallebiaja, celle du Monte Mario, pr& de Rome, et surtout celles de Calabre (Gravina di Puglia, de‘troit de Catanzaro, environs de Messine et de Reggio)”. He included his

The base of the Quaternary (Pleistocene) is recognized here as the base of

53

Calabrian in the Pliocene, believing that is represented the upper, concluding part of the Pliocene sedimentary cycle in Italy.

At the 19th International Geological Congress in Algeria (1952) the Cala- brian was designated the oldest stage of the marine Pleistocene (Quaternary). This was based upon a recommendation of a special temporary commission of the 18th International Geological Congress in Great Britain that the “Lower Pleistocene” should include as its basal member in the type area the Calabrian (formation marine) together with its (supposed) continental equivalent the Villafranchian (Int. Geol. Congr. 19th, Algeria, 1952, sect. H,

Migliorini (1950) presented a review of studies on the Pliocene/Pleistocene P. 6).

boundary in Italy. He reached several important conclusions including those enumerated below.

(1) In Italy the Pliocene/Pleistocene boundary may be placed either between Plaisancian-Astian and the Calabrian, or between the Calabrian and the Sicilian.

(2) The Calabrian can be distinguished satisfactorily from the Plaisancian- Astian by means of both molluscan and foraminifera1 faunas.

(3) The Calabrian rests unconformably upon the Astian-Plaisancian in some areas, and in areas where deposition is continuous there is evidence that the Calabrian is transgressive. Significant orogeny occurred in the Apen- nine Region between the Late Pliocene and Early Pleistocene. (4) Fossil faunas and floras indicate that pronounced climatic cooling

occurred between the Astian-Plaisancian and the Calabrian. Migliorini was of the opinion that this cooling coincided with the onset of the “Ice Age”, but this seems unlikely (see Zeuner, 1959, who reviewed the evidence which suggests that the Calabrian is wholly pre-Gunzian).

In a report prepared by the commission nominated by the Italian Geologi- cal Society t o study the Pliocene/Pleistocene boundary in Italy, Gignoux (1954) expressed his general agreement with the placement of the boundary at the base of his Calabrian Stage and with the discussion presented by Migliorini (1950). Although Gignoux (1910,1913) did not designate a stratotype locality for his Calabrian Stage, the generally accepted (Bayliss, 1969; Blow, 1969) type section is at Santa Maria di Catanzaro in southern Italy where the base of the Pleistocene (“Pliockne supe/rieur” of Gignoux) was placed at the base of a calcarenite unit, G-G’ , in which Cyprina islandica was recorded. Selli (1971, p.57) has recently designated the Santa Maria di Catanzaro section as the provisional stratotype of the Calabrian.

The section at Le Castella, described by Emiliani et al. (1961) was later proposed as the definitive section for the Pliocene/Pleistocene boundary. As thus defined, the Pleistocene represents the time which has elapsed since the first appearance of Hyalinea balthica (Schroeter) in the section at Le Castella (above sample 50). Correlation between the two areas is generally good. The most apparent weakness of this type of criterion is that i t utilizes an event which may be strongly controlled by facies influences to define a chrono-

54

stratigraphical boundary. In fact, Ruggieri (1971) has emphasized the point that the Pliocene/Pleistocene boundary corresponds with the initial appearance in the Mediterranean of the pelecypod Arctica islandica (in bed G-G‘ of Gignoux at the base of the Calabrian at Santa Maria di Catanzaro) and not with the first appearance of Hyalinea balthica, which has been recorded in pre-Calabrian levels at the Santa Maria di Catanzaro stratotype by Bayliss (1969). Chronostratigraphical boundaries are better founded upon clearly defined phylogenetic changes in fossil populations.

have developed over the years: earliest evidence of Man and his culture (anthropologcal), earliest Equus-Lepto bos-Elephas faunas (vertebrate paleontologists), earliest glaciation (glaciologists), and earliest appearance of cool-water molluscan and foraminifera1 species in southern Europe (inverte- brate paleontologists).

The last three of these were recommended as criteria by the 18th Interna- tional Geological Congress in Great Britain (Section H, 1950), and several more, including anthropological, were listed in a report of the Subcommis- sion on the Pliocene/Pleistocene Boundary at the VI International Congress on the Quaternary, Warsaw (1961) (Grichuk et al., 1965, p. 324). Since that time a considerable body of information has accumulated which has added to our knowledge of Late Cenozoic earth history and, in some cases, drasti- cally modified earlier conceptions. For example:

(1) Radiometric and paleontologic information is in general agreement that the lower part if not all of the Villafranchian S.S. is of Pliocene age, and is equivalent, for the most part, to the Astian/Piacenzian Stage (see below, chapter 6 on mammalian biostratigraphy).

(2) Oxygen-isotope (paleotemperature) studies by Emiliani (Emiliani et al., 1961) have revealed a general cooling trend across the Pliocene/Pleisto- cene boundary at Le Castella in Calabrian, southern Italy from an average summer surface-water temperature of 23-25°C in the Late Pliocene to about 15” C in the Calabrian.

(3) Paleobotanical investigations of lacustrine lignites in the upper Val d’Arno area, near Bastardo, and at Leffe in northern Italy and in the Rh6ne Valley have revealed a warm flora in the Lower Villafranchian (Venzo, 1965, p. 313), and West African molluscan genera are abundant in the marine equivalent of Villafranchian levels (Ballesio, 1971). On the other hand, Lona (1963) has discovered a pre-Calabrian cold phase designated the “Arquatian” within the Astian (in strata below the Amphistegina limestone) which may represent an earlier climatic fluctuation recorded in the Mediterranean region.

ly developed. One considers that Postglacial, or Recent, or Holocene (Ger- vais, 1867-1869) constitutes a fifth era, the “Quinquennaire” (= Quin- quenary) of Parandier (1891), which would appear completely unnecessary. A second school of thought considers the Holocene as synonymous with

Thus, four independent criteria for recognizing the base of the Quaternary

With regard to the upper limits of the Quaternary, two ideas have general-

55

Postglacial and as a subdivision of the Quaternary. Indeed, in terms of the glacial and chronostratigraphical scale of the Quaternary developed in recent years, we are living in an interglacial interval within the Pleistocene, and the recognition of the Holocene as a distinct Epoch/Series does not appear warranted.

Zagwijn (1974) points out that subdivisions of the Pleistocene have commonly been based upon paleoclimatic criteria (the alternation of warm and cold phases). He suggests a more precise wording of the recommendations of the 18th International Geologic Congress (1948) that the Pliocene/Pleisto- cene boundary be drawn at “the first indications of climatic deterioration”, inasmuch as indications of climatic deterioration known to predate the base of the Pliocene/Pleistocene boundary as typified in Italy are now known to have occurred. He suggests a “model theory” for defining the Pliocene/ Pleistocene boundary in which this boundary would be the earliest of a series of boundaries between glacial and interglacial periods characterized by similar climatic changes. Furthermore glacial/interglacial (and interglacial/ glacial) boundaries in a given area should be defined by comparison with a model paleoclimatic boundary based on the Pleistocene/Holocene boundary. The Pliocene/Pleistocene boundary is then the first of a sequence of boundaries based on similar paleoclimatic changes in late Cenozoic time. A stable geographic area is a paramount requisite for the successful application of this model.

Unfortunately, this approach ignores the fact that other subdivisions of the geologic record are based on clearly defined lithologic units in the marine biostratigraphic record and that the boundaries between these units are very much limited by original designation. Having once been fixed their age is then determined by paleontological or radiometric or other means.

of about 10,000 radiocarbon years, at which time an open tundra-like vegetation was replaced by forest vegetation, indicating that the 10” C mean summer isotherm was surpassed. In fact, three levels for the Pleistocene/ Holocene boundary have been suggested in this region:

The Pleistocene/Holocene boundary in The Netherlands is placed at a level

Time Climatic change Mean summer isotherm surpassed in “C

(1) 10,000 years tundra forest 10”

(2) 13,000 years barren soils (polar desert); 5” open tundra and steppe vegetation

(3) 8,000 years appearance in Subarctic and Boreal forests of deciduous forest elements

14”

56

Level-1 climatic changes are found at the base of the Praetiglian (Zagwijn, 1960,196313) which has been recently estimated to be older than 2.5 m.y. and perhaps as great as 3 m.y. (Van Montfrans, 1971a) and suggested by Berggren (197213) to have been an expression of the initiation of glaciation in the Northern Hemisphere and equivalent to the oldest European Alpine glaciation, the Donau, and the Nebraskan in North America.

mid-Pleistocene (Zagwijn et al., 1971).

Upper Miocene at about the base of the Susterian Stage between 7-10 m.y. ago (Van der Hammen et al., 1971). None of these particular climatic conditions occurred at the base of the Pleistocene as defined in the Calabrian of southern Italy and dated at about 1.8 m.y. Zagwijn (1974) suggests that if we apply the criterion of the time at which the inland icecap of Iceland, which covered Iceland during the late glacial time, had dwindled t o such an extent that the icecap just barely covered the whole area within the present coast line (Einarsson et al., 1967), the first occurrence of an icecap of this dimension would correspond to level 1 of the model boundary above about 3 m.y. ago. This actually corresponds t o the first extension of the inland ice-sheet in Iceland, which occurred just after the formation of the marine Tjornes beds which have been dated at about 3 m.y.

as follows:

Level-2 climatic change (5" isotherm) first occurred 0.7 m.y. ago in the

Level-3 (14" isotherm) climatic conditions occurred as early as in the

The commonly accepted subdivision of the marine Pleistocene in Italy is

Tyrrhenian (Issel, 1914) Mil; izzian (Dep&et, 1918) Sicilian (Doderlein, 1872) Em .lian Cahbrian (Gignoux, 1910)

(Ruggieri and Selli, 1950)

In actual fact only the Calabrian, and perhaps the Sicilian, have the requisite characteristics of a time-stratigraphic unit. Pleistocene marine stratigraphy is in need of a more suitable chronostratigraphic subdivision.

CHAPTER 5

LATE NEOGENE MARINE BIOSTRATIGRAPHY AND EPOCH BOUNDARIES

THE LATE MIOCENE O F THE MEDITERRANEAN BASIN

The investigations of Benson (1972a,b, 1974), Benson and Ruggieri (1974), and Benson and Sylvester-Bradley (1971) on the Late Neogene ostracodal fauna of the Mediterranean Basin has thrown considerable light on the geologic evolution of this region. Middle and Late Miocene ostracod faunas indicate progressive warming of the bottom waters of the Tethys culminating in the “lago mari” of the Messinian “salinity crisis”. Deep-sea (psychrospheric) ostracods appear in the basal Pliocene marls of the Tyrrhe- nian (D.S.D.P. Site 132) and Balearic (D.S.D.P. Site 134) basins and also occur in the lowermost part of the stratotype section of the Tabianian (Lower Pliocene) near Bologna as well as in the Upper Pliocene and Lower Pleistocene section at Le Castella and in the Pliocene near Santa Maria di Catanzaro (Squillace) in Calabria. Psychrospheric ostracods are also found along the south coast of Sicily, in the “Trubi” marls (Lower Pliocene) from Agrigento eastward. The psychrospheric fauna became extinct within the Mediterranean during early Pleistocene time (Calabrian) apparently because of the uplift of the present Gibraltar Sill and the increased isolation of the Mediterranean Basin. There is as yet no evidence of the penetration of the Pliocene psychrospheric fauna as far east as the Mediterranean Ridge (D.S.D.P. Site 125). However, the presence of Pliocene psychrospheric forms in South Calabria and in the eastern Po Valley near the Adriatic - localities which are now east of the barrier which separates the eastern and western basins of the Mediterranean - remains an unresolved question of paleo- bathymetry .

The presence of a psychrospheric ostracod fauna in the Serravallian of Sicily similar to that which reappears in the Zanclian, and the progressively thermospheric fauna of the Tortonian and Messinian suggests an iterative faunal evolution under local ecologic control.

Whereas most authors have suggested deposition of the Messinian evaporitic sediments in shallow water in either shallow or deep basins, Benson (1974) suggests that deep water must have been present in at least some parts of the Mediterranean during the Messinian. His argument focuses on: (1) the indigenous phyletic continuity of some ostracod lineages, within the evaporite- dominated sequence, which are adapted to live beneath the photic zone

58

together with the scattered probable occurrences of shallow-shelf assemblages in the same sequence, and (2) the demonstrated occurrence of ostracod faunas from upper cold Atlantic water masses in the Mediterranean fossil record before and immediately after the Messinian “salinity crisis” without evidence of the invasion of an Atlantic shallow-water fauna that one would expect to accompany the (slow) refilling of such a desiccated basin. Benson (1974) prefers to explain the rather extreme contrasts in terms of water mass evolution under threshold control rather than in terms of widespread changes in basin physiography and/or vast oscillations of sea level. In this interpretation the following sequence of events is outlined :

(1) Serrauallian - psychrospheric ostracod fauna indicates connection of at least some of the minor basins of the Tethys with the deeper world ocean fauna.

( 2 ) Tortonian- progressive transition of the psychrospheric fauna to a thermospheric fauna through the gradual isolation of the Tethys from the Atlantic by uplift of the western threshold (i.e. Gibraltar Sill, Iberian Portal Sill).

(3) Messinian - restricted, thermospheric ostracod (Cyprideis) fauna developed in localized basins, some of which were deep (i.e. > 500 m). Pro- gressive warming of basin waters above psychrospheric-thermospheric interface (10” C), caused by increasing isolation of the basins, would have destroyed the original oceanic two-layered stratification and thereby the depth control of the faunas. These so-called “lago mari”, or thermally uniform sea lakes, with local variations in water chemistry mainly in response to dontinental influences were solely thermospheric and contained lagoonal faunas or shallow marine refugees. (4) Zanclian - abrupt invasion of deep, oceanic waters and reinvasion of

new psychrospheric fauna from Atlantic water masses at least 500-1000 m deep.

( 5 ) Piacenzian - continued development of psychrospheric fauna in deeper parts of the Mediterranean with contemporaneous transition to thermospheric fauna in progressively shallower areas, culminating in a repeti- tion of the Aurila-Loxoconcha faunas of the Late Tortonian in the Astian (Late Pliocene).

(6) Calabrian - extinction of the psychrospheric fauna in the Mediterra- nean by the continued uplift of the Gibraltar Sill and the isolation of the Mediterranean from Atlantic deep water.

The paradox of the Mediterranean Late Miocene remains (Hsu, Cita et al., 1973; Benson and Ruggieri, 1974). Late Miocene sedimentation occurred under what appears to be shallow-water, restricted conditions. Yet the subsequent (Pliocene) sedimentation was of deep-sea nature. No transitional shallow-water benthic faunal elements have been found in between. Appar- ently catastrdphic flooding must have (aken place at the beginning of the Pliocene, 5 m.y. ago. One is constrained by geophysical common sense from suggesting that the basins in the Mediterranean were suddenly dowpwarped

I

59

to depths t o which the psychrospheric fauna was adapted (500-1000 m and more). Some of them must have been topographically deep during the Mes- sinian (having been already deep in the preceding Serravallian and Torto- nian), and the observed faunal changes were evidently caused by local evolution in isolated and variable water masses rather than by abrupt tectonic movements. Suffice it to say that the terminal Miocene events in the Medi- terranean were of extraordinary magnitude and they led to the rather rapid (in geologic terms) transformation of the Tethys Ocean Sea into the Mediter- ranean Sea *.

THE PLIOCENE EPOCH AND THE MIOCENE/PLIOCENE BOUNDARY

In trod uc tion

The age of the Miocene/Pliocene boundary has been estimated t o be as young as 2.7-3.0 m.y. (Bandy and Casey, 196913; Bandy, l971,1972b,c; Bandy e t al., 1971a, 1972) and as old as 1 5 m.y. (Trevisan, 1958). Estimates by most vertebrate paleontologists have ranged between 10 and 1 2 m.y. because of the supposed correlation between the initial appearance of Hippa- rion in the Pontian of the eastern Mediterranean and the Miocene/Pliocene boundary (e.g., Matthew, 1929). More recent estimates based on K-Ar dating (7 m.y. for the Mt. Capanne granodiorite of Elba, 6.6- 8.0 m.y. for the volcanic tuffs intercalated with Messinian beds in Morocco and 6.0 m.y. for pre-Pliocene tuffs of Ischia in the Bay of Naples; cf. Van Couvering, 1972) have indicated that the Miocene/Pliocene boundary as defined in the marine sedimentary succession in the Mediterranean region, is younger than 6 m.y. The date of 4.7 m.y. on the Orciatico trachyte which metamorphoses lower Pliocene clays in Tuscany and 3.4 m.y. on glass shards in middle Pliocene beds in Crotona (Selli, 1970), together with integrated biostratigraphic paleomagnetic studies on deep-sea cores (Hays et al., 1969) have indicated that the boundary is older than 4.5 m.y. An age of 5 m.y. was suggested by Berggren (1969c, 1972a), and recent investigations have added support t o this estimate (Berggren, 1973; Cita, 1973,1974; Gill and McDougall, 1973; Saito e t al., 1974).

In a similar manner age estimates of the Pliocene/Pleistocene boundary have varied over a wide range from about 0.6 m.y. (Emiliani, 1961,1964, 1966a,b) t o over 3.5-4 m.y. (the dates of the earliest Villafranchian faunas, attributed by many vertebrate paleontologists in America, Europe and the Soviet Union to the Pleistocene). Recent work has shown that the type Villafranchian fauna is essentially late Pliocene (pre-Calabrian) in age, that the Villafranchian s.1. is as young as 1.0 m.y. and that the base of the stratotype Calabrian at Santa Maria di Catanzaro, i.e., the Pliocene/Pleistocene boundary, is about 1.8 m.y. old (Berggren et al., 1967; Glass e t al., 1967; Berggren, 1968; Briskin and Berggren, 1974).


60

The discussion below deals primarily with the Miocene/Pliocene boundary and Pliocene biostratigraphy. Reference is made in appropriate places to the Pliocene/Pleistocene boundary, but a separate section is reserved for a discussion of the problems associated with that boundary.

Discussion

The problems associated with the determination of the position of the Miocene/Pliocene and Pliocene/Pleistocene boundaries based on planktonic foraminifera have had a long and complex history spanning the last decade. In the interest of providing a historical perspective for these aspects of Late Neogene biostratigraphy, a summary is presented below of the major studies dealing with these topics. In this way some of the conflicts which have arisen in the course of this debate will become apparent. In the discussion which follows the reader is referred to Fig. 8.

In 1963 two important papers appeared (Ericson et al., 1963; Riedel et al., 1963) in which several criteria for recognizing the Pliocene/Pleistocene boundary in deep-sea sediments are suggested. These included : (1) the extinction of discoasters; (2) change in the coiling direction in the planktonic foraminiferal group of Globorotalia menardii from 95% dextral below the boundary to 95% sinistral above it; (3) a reduction in the Globorotalia menardii group to a single fairly uniform race above the boundary; (4) an increase in the diameter of the test of Globorotalia menardii above the boundary; (5) the appearance of Globorotalia truncatulinoides in abundance above the boundary; (6) the disappearance of Globigerinoides triloba sacculi- fera (= Globigerinoides fistulosa) at the boundary.

After studying the planktonic foraminiferal biostratigraphy in the Philip- pines, Bandy (1963a) suggested that sections in eight deep-sea cores, reported b y the Lamont group to have been o€Pliocene age, were in actuality latest Miocene and that a gap of about 10 m.y. of Pliocene time occurred in them. On the basis of his work in the Philippines Bandy (1963b) considered the lower Pliocene to be characterized by an abundance of Sphaeroidinella dehiscens dehiscens. In the basal part of his “lower Pliocene”, Pulleniatina obliquiloculata was observed to coil sinistrally, whereas in the upper part of the “lower Pliocene” this species coiled dextrally . The appearance of Sphae- roidinella dehiscens dehiscens was thought to mark the Miocene/Pliocene boundary (Bandy, 1963a,b) and this was defined more fully by Bandy (1964) as the Sphaeroidinella dehiscens dehiscens Datum :

“The Sphaeroidinella dehiscens dehiscens Datum is selected as the Miocene- Pliocene boundary (Text Fig. 6) on the basis of the deep water conditions of the basin in southern Iloilo (Bandy, 1962 and paper in press). This level marks the upper limit of the Globoquadrina altispira globosa-Globoqua- drina dehiscens dehiscens assemblage and the beginning or basal occurrence of Sphaeroidinella dehiscens dehiscens and Globorotalia truncatulinoides.

61

Fig. 8. Comparative Pliocene and Pleistocene planktonic biostratigraphy and epoch boundary determinations (see text for explanation).

62

Pulleniatina obliquiloculata is relatively rare in the upper part of the Mio- cene. It becomes abundant and dominantly left coiling in the basal Pliocene and then reverses abruptly to right coiling for the remainder of its range. Sphaeroidinella dehiscens immatura also occurs in the uppermost Miocene although it is most characteristic of the Pliocene and Quaternary interval.”

(Bandy, 1964, p. 10)

The Sphaeroidinella dehiscens Datum was also thought by Bandy (1964, p. 11) to mark the top of the Pontian Stage as used in the eastern Mediterra- nean region and the Pontian and Sarmatian were placed together within the Sahelian and in the latest Miocene (Bandy, 1964, p. 11). The Sahelian, if it has any stratotypic meaning at all (Sissingh, 1972; Tjalsma and Wonders, 1972) is now considered to be essentially correlative with the upper part of the Tortonian Stage whereas the Pontian land mammal association is essentially middle-late Miocene in age (Van Couvering and Miller, 1971). It should be remembered, however, that a decade ago the planktonic foraminiferal biostratigraphy of European/Mediterranean time-stratigraphic units was only vaguely understood.

Two years later the Sphaeroidinella dehiscens dehiscens Datum was used to define the lower boundary of Zone N19 at a level only slightly above (- 40 f t ) the base of the Trubi marl in Sicily and the Miocene/Pliocene boundary (Banner and Blow, 1965b). This use of the Sphaeroidinella Datum was subsequently spelled out more clearly by Blow (1969, p. 253-256,338, 386-387, 418, see in this connection his fig. 20 on p. 289). Blow (1969, p. 254) defined the base of his Zone N19 (and his Sphaeroidinella Datum) in the following manner: “The base of this zone is placed at the horizon of the first evolutionary appearance of Sp haero id ine lla deh iscens de h iscens (including forma immatura) from its immediate ancestor Sphaeroidinellopsis subdehiscens paenedehiscens.” He noted (1969, p. 418), “that the development of Sphaeroidinella dehiscens (s.1.) from Sphaeroidinellopsis su bdehiscens paenedehiscens is about 40 ft. above the base of the exposed Trubi marl in Sicily. Therefore the upper part of the Trubi marl formation is referable to Zone N19 and the lower part referable to Zone N18. The whole of the Trubi was included in the Zanclian Stage by Seguenza and accordingly the Plio- cene/Miocene boundary is geostratigraphically within Zone N18 and thus prior to the evolution of Sphaeroidinella dehiscens (s.l.).”

The Sphaeroidinella dehiscens Datum, as originally defined by Bandy (1963a,b, 1964), refers to the initial appearance of large robust individuals of Sphaeroidinella dehiscens with a broad scalloped everted flange. This level, associated with the extinction of the Sphaeroidinellopsis group, is slightly older than the extinction level of Globoquadrina altispira and Globo- rotalia multicamerata, is coincidental with the planktonic foraminiferal N20/N21 boundary and has been shown by Hays et al. (1969) to be about 3 m.y. old on the paleomagnetic time-scale. I t does not correspond to the Miocene/Pliocene boundary as defined in the marine sedimentary succession in the Mediterranean.

63

The Sphaeroidinella dehiscens Datum, as used by Blow (1969) (and followed by Berggren, 1969a,b, 1971a,b, 1972a, 1973,1974; Saito, in Hays et al., 1969; and Saito and Funnell, 1971) on the other hand is based upon the initial appearance of forms referable to the genus Sphaeroidinella with a small but distinct sutural opening on the spiral side. Blow (1969) considers that the taxon immatura refers to very phylogenetically primitive forms of Sphaeroidinella dehiscens dehiscens. He observed the fact that forms with the small aperture typical of immatura are soon replaced by the typical Sphaeroidinella dehiscens dehiscens with larger dorsal apertures and therefore united them under a single taxon distinguishing the morphological feature of the smaller aperture by reference to the nomenclature “forma immatura”. The gradual enlargement of the supplementary aperture and the extension of the periumbilical flange in Sphaeroidinella dehiscens to the morphologic expression which it exhibits in present-day seas was observed by this writer (Berggren, 1969a) in lower Pliocene cores from the Gulf of Mexico obtained on Leg 1 of the Deep Sea Drilling Project.

The initial occurrence of Sphaeroidinella dehiscens has been shown by Berggren (1973) and Saito et al. (1974) to occur at about 4.8 m.y. and an age of 5 m.y. has been estimated for the Miocene/Pliocene boundary (Berg- gren, 1973; see also discussion later in this section). Thus, we see that the definition of the Sphaeroidinella dehiscens Datum by both Bandy (1963a,b, 1964) and Blow (1969) is clearly and unequivocally stated but refers to biostratigraphic events which are significantly different in space and time. More recently Lamb (1969, p. 571; pl. 1 , figs. 1-5; pl. 2, figs. 1-3) described a new species, Sphaeroidinella sphaeroides, which appears to overlap Blow’s (1959,1969) concept of both Sphaeroidinellopsis subdehiscens paenedehiscens and Sphaeroidinella dehiscens forma immatura. Lamb (1969, p. 528) observes that “individual specimens may show a break in the cortex, as solution pits, along the sutures on the spiral side, but these should not be mistaken for true secondary apertures as seen in Sphaeroidinella dehiscens (Parker and Jones)”. However, this would appear to be an incorrect assess- ment of the true situation. A small supplementary aperture in forma immatura is indeed real. It can be seen to have a rim-like lip around the edge indicating that it is a genetic-growth feature and not a product of solution. It can be seen to gradually increase in size during the Early Pliocene in the interval between 5 and 3 m.y. (Berggren, 1969a). This form was recorded by Lamb and Beard (1972) from a level within the Late Miocene to just about the top of the Pulleniatina obliquiloculata Zone, that is at about and slightly above the first appearance of Sphaeroidinella dehiscens (with flange) and approximately at the level of the extinction of Globoquadrina altispira. It is significant that all forms showing the supplementary aperture illustrated by Lamb (1969, pl. 1, figs. 1-5; pl. 2, figs. 1-3) and Lamb and Beard (1972, pl. 1, figs. 3,4; pl. 34, figs, 3-8; pl. 35, figs. 1-7) are from Pliocene levels (that is within their margaritae Zone or younger). Those from the Sphaeroi- dinella sphaeroides Zone (Lamb and Beard, 1972, pl. 34, figs. 4,5 and 7) are

64

shown only in side or frontal view so that no evaluation is possible from their illustrations. I t is quite possible that in some instances however the supplementary apertures on the spiral side are indeed due to solution or breaks in the cortex and not true supplementary apertures as seen in forma immatura. The need to distinguish between the two is obviously important in view of the current controversy.

Blow (1969, p. 387) distinguished paenedehiscens from subdehiscens in that the former possessed “a relatively much larger primary aperture, a more tightly coiled test in which the chambers are more embracing and in lacking clearly defined, externally visible sutural commissures between any of the three final chambers” and in being consistently much larger than subdehiscens S.S. (Most current authors do not distinguish paenedehiscens and subdehiscens in their work so that the extinction level of subdehiscens of most authors coincides with that of paenedehiscens of Blow (1969, p. 272, fig. 8).) That the taxon sphaeroides Lamb is probably identical to paenedehiscens Blow is further suggested by the fact that the initial appearance of both is shown within the Late Miocene (Zone N17 of Blow, 1969, p. 272, fig. 8; S. sphaeroides Zone of Lamb and Beard, 1972, p. 41, table 2).

It is important to bear two facts in mind in reading the section which follows regarding the vicissitudes of the Miocene/Pliocene boundary namely : (1) the appearance of Sphaeroidinella dehiscens forma immatura occurs at about 5 m.y., that of Sphaeroidinella dehiscens dehiscens (with flange) occurs at about 3 m.y. (documentation of this will be presented below); (2) in their studies on Late Neogene biostratigraphy Saito, Berggren and Cita have consistently used the criterion of Blow in defining the position of the Miocene/ Pliocene boundary - the same may be said to be true of the various initial reports of the Deep Sea Drilling Project.

In 1967 Bandy and Wade assumed that the extinction of Sphaeroidinel- lopsis spp. (seminulina, subdehiscens and others), Globoquadrina altispira altispira, and Globorotalia tumida miocenica occurred at the Miocene/Plio- cene boundary and that Globorotalia truncatulinoides appeared above this boundary, that Sphaeroidinella dehiscens “becomes well developed above the boundary” and that Pulleniatina obliquiloculata and an abundance of Globorotalia puncticulata appear at the boundary, whereas very rare specimens occur below the boundary. The Globigerina nepenthes extinction datum was thought to correlate to the Pontian/Sarmatian boundary in the brackish-marine sequence of the Paratethys (Bandy, 1964). Globorotalia truncatulinoides and Globorotalia inflata were said to make their appearance at or near the Sphaeroidinella dehiscens Datum (the Miocene/Pliocene boundary).

The Miocene/Pliocene boundary in California, which was correlated t o the (Philippine) Sphaeroidinella dehiscens Datum sensu Bandy (1964), was dated at 9 m.y. (Ingle, 1966; Bandy, 1967b, p. 101) based on a vitric ash date in the upper part of the Malaga mudstone (Delmontian age) of Malaga Cove. Southern California. Biostratigraphic control is poor in this part of the

65

section, however (lngle, 1967), and i t is not clear whether this level is indeed correlative with the Sphaeroidinella deiiiscens Datum as defined in the Philippines. The Pliocene/Pleistocene boundary of California is traditionally defined by a definitive shift from dextral (late Pliocene) to sinistral (early Pleistocene) populations of Globigerina pachyderma (Bandy, 1967b). This was based on earlier studies (Bandy, 1959, 1960a) in which coiling changes in Globigerina puchyderma were used t o define the limits of the Pleistocene in the general area of Southern California as well as climatic changes within the Pleistocene. An age of 3 m.y. based on K-Ar date (glauconite) on the Lomita Marl of the Palos Verdes Hills (Obradovich, 1965) was suggested for this level. This time framework (9-3 m.y.) for the Pliocene was reiterated in several subsequent publications (Bandy, 1968a,b, 1969, 1970a; Bandy and Ingle, 1970).

The advent of paleomagnetism and its application to deep-sea biostratigraphy caused Bandy to reinterpret the radiometric dates assigned to the biostratigraphic datum levels (Bandy and Casey, 1969b). However, he indicated that the Miocene/Pliocene boundary as determined by the initial appearance of Sphaeroidinella dehiscens in California corresponded to the boundary between Blow’s N l 8 j N l 9 and pointed to the conflict betweell the K-Ar date of 9 m.y. and the paleomagnetic date of 3 m.y. for this ltwel. A source of error here lies in the assurnption lhat the aypearancx of Lp:.aeroidineila dehiscens (with flange) was equivalent to the boundary between Blow’s Zone N18 and N19. As we have seen above the initial apprarance of Sphaeroidinel- la dehiscens forma immutura corresponds to the boundary hetween Zone N l 8 and N19 and is significantly lower and older than the appearance of Sphueroidinella dehiscens (with flange). The latter datum ievel corresponds approximately with the boundary between Blow’s Zone N20 and N21 (see below). Bandy and Casey (196913) observed that in California their Sphaeroi- dinella dehiscens Datum occurred just below the extinction of Prunopyle titan and that this occurred within the upper part of the Gauss Normal Magnetic Epoch for which a radiometric age of about 3 m.y. had been determined. On this basis he concluded that if the radiometric age of this paleomagnetic event is correct then the Miocene/Pliocene boundary occurs within the Gauss Epoch. They concluded (196913, p. 252) that, “either the date from land based marine sections are in error by a factor of 3 or those dates associated with the paleomagnetic scale are incorrect”.

tion in recognizing the beginning of the Pleistocene (Berggren et al., 1967; Phillips e t al., 1968; Berggren, 1968; Glass e t al., 1967) led Bandy to a revision of earlier determinations of this boundary in California. This paleontologic event occurred in California near the base of the Wheelerian Stage and in deep-sea cores near the base of the Gilsa (= Olduvai) Event at a level approximately equivalent to the upper limit of Eucyrtidium caluertense. Again Bandy was faced with the dilemma that either the radiometric date or the paleomagnetic time-scale was incorrect. In adjusting his biostratigraphy

The use of the Globorotalia tosaensisltruncatulinoides evolutionary transi-

66

to the paleomagnetic time-scale Bandy now had a Pliocene which spanned the interval from 3 to 1.8 m.y.

between high and low latitudes and their relation to a paleomagnetic time- scale was discussed by Bandy and Casey (1969a,b) and Bandy et al. (1971a). In warm, temperate, and tropical regions the Sphaeroidinella dehiscens Datum (sensu Bandy, 1964) was shown to occur in the Gauss at approximately the same position as the Prunopyle titan Datum level from the Ant- arctic. This level at 2.7 m.y. was correlated with the base of the Zanclian Stage of the Mediterranean and with the N18/N19 boundary of Blow (1969). The Pliocene/Pleistocene boundary was placed at the base of the Gilsa (= Olduvai) at about 1.8 m.y. The extinction of Sphaeroidinellopsis subdehiscens and Globoquadrina altispira near the appearance of Sphaeroi- dinella dehiscens was reiterated. Pulleniatina obliquiloculata and Globiger- inoides fistulosa were said to first appear at this level as well. The assumption that the N18/N19 boundary (approximately equal to the base of the Zancl- ian and base Pliocene) occurs at about 2.7 m.y. leads to a significantly younger age assignment to the late Miocene Zone N17 than had been suggested by other workers. Bandy observed that the upper Neogene N17 fauna in the tropics has a temperate Globigerina bulloides fauna, whereas in temperate regions it is characterized by a Globigerina pachyderma left-coiling population and in Antarctic seas by levels in which the first glacially rafted deposits occur (approximately 4.5 m.y. ago) at or above Gilbert c. This “upper Miocene” cold cycle occurs between the Gilbert a and c according to Bandy et al. (1971a). However, these correlations are quite incorrect inasmuch as the extinctions of the forms cited above are characteristic of the mid-Pliocene (about 3 m.y. ago) and not of the Late Miocene as typified in the Mediterranean region. The correlation of the Gilbert a through c with Zone N17 is incorrect.

the lower Matuyama Epoch. There is ample documentation of this upper Pliocene cooling (at about 2.4 m.y.) but this cannot be placed within the framework of Bandy’s “late Pliocene”. If the Pliocene as typified in the Mediterranean region is represented by the time interval of 2.7 to 1.8 m.y., then the entire Pliocene is seen to be approximately 1 m.y. in length. Within this interval of time it would be impossible to recognize a three-fold subdivision based upon biostratigraphy and/or paleoclimatology . Furthermore, the base of Zone N21 (defined by the evolutionary appearance of Globorotalia tosaensis from Globorotalia crassaformis) occurs at about 3 m.y. (Saito, in Hays et al., 1969) which is below the level at which Bandy has defined the Miocene/Pliocene boundary; yet this is supposed to correspond to the N18/N19 boundary (Bandy, 1969; Bandy and Casey, 1969a,b; Bandy et al., 1971a). A further indication that the Miocene/Pliocene boundary (2.7-3.0 m.y.) of Bandy is in need of revision are the well-documented dates of 2.5-3.5 m.y. and as low as 4.4 ? 0.3 m.y. from middle to early Villa-

The relationship of planktonic foraminifera1 and radiolarian datum levels

An “upper Pliocene” cold interval was said to correspond to Zone N21 in

67

franchian strata in the Mediterranean region (Savage and Curtis, 1970; Am- brosetti et al., 1972) which can be correlated with the marine Pliocene in Italy and the Rh6ne Estuary and which indicates that the marine Pliocene as typified in the Mediterranean extends back to at least 4.5 m.y. This is also corroborated by the date of 4.7 m.y. on the Orciatico trachyte (Pisa, Tusca- ny) which metamorphoses lower Pliocene clays (Tongiorgi; in Selli, 1970).

Bandy et al. (1971a) discuss several points pertinent to the position of the Miocene/Pliocene boundary.

(1) The N18/N19 boundary according to Bandy et al. (1971a) is a datum plane which separates Sphaeroidinellopsis su bdehiscens (below) from Sphae- roidinella dehiscens (above) with only questionable forms of the former overlapping the lower range of the latter. According to these authors (1971a, p. lo ) , the upper significant limit of Sphaeroidinellopsis subdehiscens has been clearly established in the Gauss Normal Magnetic Epoch (about 3 m.y.) just above the Mammoth Event, and in deep-sea cores this would correlate with the N18/N19 boundary of Blow (1969) and Berggren (1969b,c). How- ever, this is incorrect as was pointed out above. The upper limit of Sphaeroi- dinellopsis subdehiscens paenedehiscens was said by Blow (1969) to occur within the lower part of Zone N20. The upper limit of this taxon as well as Sphaeroidinellopsis seminulina is now known to occur at about 3 m.y. which is approximately the boundary between Zone N20/N21 (Saito, in Hays et al., 1969). Neither Saito nor Berggren distinguish between paenedehiscens or subdehiscens and thus the upper limit of their taxon subdehiscens is the same as Bandy’s Sphaeroidinella dehiscens Datum but this is not the N18/N19 boundary of Blow (1969). Bandy et al. (1971a, p. 10) indicate that the rare occurrences of Sphaeroidinella dehiscens below the upper limit of Sphaeroidinellopsis subdehiscens (cited in Hays et al., 1969) are “likely rare contaminants in the piston cores in the same manner as rare occurrences far down section of G. truncatulinoides that we have noted in other cores. If the zone of overlap is actual then part of N19 spans 1.5 m.y. or more extending from the middle Gauss all the way down to below Gilbert c, a relationship that is highly unlikely”. Actually, this is exactly the relationship which does exist. The occurrences of Sphaeroidinella dehiscens below the upper limit of Sphaeroidinellopsis subdehiscens which Saito (in Hays et al., 1969) records refer to forma immatura with a small supplementary aperture and are not rare contaminants. Thus the zone of overlap of Sphaeroidinellop- sis su bdehiscens and Sphaeroidinella dehiscens forma immatura is approximately 2 m.y. and this is the interval spanned by zones N19 and N20.

(2) Globigerinoides fistulosa commences near the upper limit of Sphae- roidinellopsis subdehiscens and its range is largely in the lower part of the range of Sphaeroidinella dehiscens, extending from within Zone N18 up to the upper part of Zone N21. In actual fact Globigerinoides fistulosa appears only about 3 m.y. ago a t about the level of mid-Gauss (Mammoth Event) (Saito, in Hays et al., 1969). It does not appear within Zone N18. Thus its appearance coincides approximately with the Sphaeroidinella dehiscens Datum of Bandy.

68

(3) hlleniat ina obliquiloculata was said to originate from Pulleniatina primalis in the lower part of Zone N19 and the two species were said to continue upward into Zone N20. In actual fact, in deep-sea cores the initial occurrence of Pulleniatina obliquiloculata occurs consistently in the middle or upper part of the Gauss, just above the extinction level of Sphaeroidinel- lopsis su bdehiscens. This level is at about 3 m.y. and near the N20/N21 boundary, not in the lower part of N19.

(4) The extinction level of Prunopyle titan and Lychnocanium grande occurs within the upper Gauss in Antarctic cores. In California Prunopyle titan disappears approximately with Sphaeroidinellopsis su bdehiscens near the upper boundary of the Miocene as recognized there (Ingle, 1967; Bandy and Ingle, 1970). This level, however, bears no relationship to the Miocene/ Pliocene boundary as defined in the Mediterranean region. This level is within the Middle Pliocene as shown by dates within the Pliocene summarized above.

(5) The authors draw attention to problems associated with the distribution of Globigerina nepenthes. In the Pacific it disappears at the Gilbert a (3.7 m.y.) according to Saito (in Hays et al., 1969), well below the level of extinction of Sphaeroidinellopsis subdehiscens. Parker (1967), however, shows it extending as high and even higher than Sphaeroidinellopsis subdehiscens. Bandy et al. (1971a) suggest that differences in paleooceanograph- ic conditions may have accounted for this difference in stratigraphic range, but it would appear more likely that inasmuch as Globigerina nepenthes was recorded as high as within Zone N19 by Parker (1967), where it is known to become extinct, that the real upper limit of Sphaeroidinellopsis su bdehiscens was not recorded. The distinction between subdehiscens and dehiscens forma immatura is extremely difficult to determine in some instances.

is Globoro talia margaritae and the anomalously high stratigraphic range attributed to it (above the extinction of Sphaeroidinellopsis subdehiscens) by Parker (1967). They conclude that “as a probable temperate species, its most typical and complete range would be in temperate areas, whereas its range in tropical areas would be most abbreviated and restricted to the upper Miocene cool cycle that centered in N17”. Thus, Bandy et al. (1971a) again assume that the upper range of margaritae (within the Gilbert and lower Gauss) is equal to Zone N17 but this is shown above to be quite impossible. Globorotalia margaritae disappears in most cored sequences at the Gauss/ Gilbert boundary (3.3-3.4 m.y.) (Datum VI of Hays et al., 1969; Cita, 1973) and this level is within the upper part of the Discoaster asymmetricus Zone, “14 (Berggren, 1973; Cita, 1973) which is within the upper part of Zone N19 of Blow (1969).

The authors summarize their position on the Miocene/Pliocene boundary in the following manner (Bandy et al., 1971a, p. 21):

(6) Another species to which Bandy et al. (1971a, p. 12) devote attention

“The Sphaeroidinella dehiscens Datum plane in this study is essentially

69

correlative with the Sphaeroidinellopsis su bdehiscens Extinction Datum plane and it is only slightly below the level that could be referred t o as the Pulleniatina obliquiloculata Datum plane in the tropics. In temperate areas the Globorotalia puncticulata Datum plane or the Prunopyle titan Extinc- tion Datum plane is approximately correlative with the S. dehiscens Datum plane. In the Antarctic the Prunopyle titan Extinction Datum plane or the Lychnocaniurn grande Extinction Datum plane would represent approximately the same horizon as the S. dehiscens Datum plane.

ate and Antarctic cores are correlative with a level in the upper Gauss Nor- mal Magnetic Epoch above the Mammoth Event, perhaps above the Kaena Event.”

The S. dehiscens Datum plane and its equivalent datum planes in temper-

In their paper the authors assign an age of approximately 2.7 m.y. t o the Sphaeroidinella dehiscens Datum plane.

The Pliocene/Pleistocene boundary in California is placed near the base of the Wheelerian Stage based on the appearance of Globorotalia truncatulinoides and the extinction of discoasters in the lower part of the Wheelerian in Balcom Canyon near Bakersfield, California. Radiometric dates on the Bailey Ash bed in Balcom Canyon (which occurs within the zone where G. tosaensis gives rise t o G. truncatulinoides) range from 1 to almost 10 m.y. (Yeats 1965; Yeats e t al., 1967; Bandy and Ingle, 1970) and clearly indicate that they are unreliable in determining the age of the Pliocene/Pleistocene boundary as determined by biostratigraphic correlation in California. On the other hand, the Pliwene/Pleistocene boundary in the Los Angeles basin has been usually de’ined at the stratigraphic lev1 at which there is a major influx of cold- wafer sinistral populations of Clobigerina pachyderrna replacing the temperate dextral populations of the “upper Pliocene” (Bandy, 1960b; 1964; 19G7a,b; 1968a,b; 1969). This level in California is now known to be stra’i- grephically much higher than the Globorotalia truncatulinoides Datum plane thcct is coincident with the base of the Olduvai Event, but i t occurs near the bare of the Lomita Marl, which has been dated radiometrically by Obrado- vich (1965, 1968) at 3.0 f 0.09 m.y. Bandy et al. (1971a, p. 22) suggest this coiling change within Globigerina pachyderrna from dextral to sinistral may actually correlate with the prominent cooling trend noted near the Brunhes/ Matuyama boundary in the Antarctic a t about 0.7 m.y., in which case the dates on the Lomita Marl must be about three times too old if the planktonic events are properly calibrated to the paleomagnetic time-scale.

Miocene” Neogene Zone N18 with the experimental Mohole section, with the uppermost Mohnian and the Celmontian Stage of California, with the uppermost Messinian of Italy and with the upper Gilbert and lower Gauss magnetic epochs, the upper boundary of which is about 3 m.y. A similar correlation is suggested by Casey (1972) based on radiolarians (however, see discussion below ).

In recent publications Bandy (1969, 1972a,b) has correlated “upper

70

Paleomagnetic calibration of Pliocene zones

A resolution of some of the controversial correlations within the Pliocene which had developed was provided by the important work of Hays et al. (1969). The ranges of selected species of four major microfossil groups (foraminifera, radiolaria, diatoms and silicoflagellates) were compared with geomagnetic reversals back to 4.5 m.y. in eastern equatorial Pacific cores. The establishment of the chronostratigraphic relationships of these fossil ranges allowed the authors to compare biostratigraphic zonations and correlations of previous authors and provided an absolute time framework that could be used in a worldwide correlation of marine sediments.

extinction of all species of Sphaeroidinellopsis with the increased upward abundance of Sphaeroidinella dehiscens) occurred at the top of the Mam- moth Event. This is seen to be the same as the Sphaeroidinella dehiscens Datum of Bandy (1963a,b; 1964), which has been used by some authors to define the Miocene/Pliocene boundary (Bandy, 1963a,b; 1964; Bolli, 1966; Bandy and Wade, 1967; Bandy et al., 1971a). The extinction of Globoqua- drina altispira and Sphaeroidinellopsis multiloba (= S. seminulina) was reported by Ericson et al. (1963) at the Pliocene/Pleistocene boundary, but this event occurs only slightly above Datum V at about 2.8-2.9 m.y. (Saito, in Hays et al., 1969; Glass et al., 1967; Berggren, 1969a).

The difficulty experienced by different workers in determining the position of the Miocene/Pliocene boundary utilizing the criteria suggested by Banner and Blow (1965b) and Blow (1969) is pointed out by Hays et al. (1969, p. 1500-1501) in the following example (core LSDH 78P; Fig. 9).

The oldest zone of major change (planktonic foraminifera1 Datum V - the

POSITION OF

DATUM LEVEL M I O C E N E / P L I O C E N E B O U N D A R Y L E V E L S IN ZANCL~AN

D A T U M B I O S T R A T I G R A P H Y OF T H E P A L E O - T

T I M E S C A L E

MAG i n

m y

W z W V 0 - : ZANCLIAN J

3Om above base c rugorus -+

R C

Fig. 9. Comparative biostratigraphy of the Miocene/Pliocene boundary (see text for further explanation).

7 1

(1) The Miocene/Pliocene boundary determined by Parker (1967) was correlated by the ranges of radiolarian species and planktonic foraminifera at a level between the Gilbert c and b at an age of about 4.2 m.y.

(2) Blow (1969, figs. 38 and 41) placed the N18/N19 boundary at 500 cm in this core and this level is between Gilbert c and b at about 4.1 m.y. at about the same level as that determined by Parker. The Miocene/Pliocene boundary of Blow would be at about 4.3 m.y. (supposedly within Zone N18). It is of interest that Martini and Bramlette (1963) had determined the position of the Miocene/Pliocene boundary on the basis of calcareous nannofossils in the experimental Mohole at a level dated subsequently by Diamond at 4.3 f 0.3 m.y.

(3) However, on the basis of work on equatorial Pacific cores the Mio- cene/Pliocene boundary would be older than 4.5 m.y. inasmuch as specimens of Sphaeroidinella dehiscens forma immatura were found to occur in levels at least that old. Current work suggests that the initial appearance of Sphae- roidinella dehiscens forma immatura occurred about 4.8 m.y. ago (Berggren, 1972a, 1973; Saito et al., 1974).

Bolli and Bermudez (1965) made a revised planktonic foraminifera1 zonation for the middle Miocene to the Pliocene warm-water regions. Applying this zonation to the Indonesian region the Miocene/Pliocene boundary was drawn at the boundary between their Globoquadrina altispiralGloborotalia crassa f o r m is and Glo boquad rina a1 tispira/Globoro talia trunca tulinoides zones. However, as Robinson (1967) and W.H. Blow (personal communication, 1968) have pointed out the Manchioneal Formation is Pleistocene and the presence of G. altispira there is due to reworking. The two species are not found together in natural association (see also Blow, 1969; Bolli, 1970, p. 595). Globigerinoides fistulosa appears just before the extinction of Glo- boquadrina altispira. Thus the Miocene/Pliocene boundary of Bolli and Bermudez (1965) can be correlated with a level between the Mammoth and the Kaena events about 2.9 m.y. ago (Hays et al., 1969, p. 1500). Near the “middle/upper Miocene’’ boundary a coiling change in Pulleniatina occurred from sinistral to dextral, which follows shortly upon the extinction of Globi- gerina nepenthes and is associated with an interval dominated by dextrally coiled Globorotalia tumida. This same sequence of events occurs in the eastern equatorial Pacific cores discussed by Saito (in Hays et al., 1969), and this level is seen to coincide with the Gilbert a at about 3.7 m.y. (see Fig. 8).

The Miocene/Pliocene boundary was subsequently placed at the base of the Globorotalia margaritae Zone by Bolli (1970). The position of this boundary, as determined in the Caribbean sites does not agree with the base of the G. margaritae Zone as drawn in the Mediterranean region and by Beard and Lamb (1968) and Lamb and Beard (1972). This can be seen from the fact that the Miocene/Pliocene boundary is drawn at Site 29 within core 4 (Bolli, 1970, p. 609). All of core 4 lies within the Discoaster surculus Zone, however (Hay, 1970, p. 480) which is no older than mid-Pliocene. The Miocene/Pliocene boundary as determined by Beard and Lamb (1968), Saito

72

(in Hays et al., 1969), Lamb and Beard (1972), and Berggren (1973) is within the Ceratolithus tricorniculatus Zone (see discussion below). This supports the suggestion by Saito (in Hays et al., 1969) that the G. dutertreil G. obliquus extremus Zone is a short zone whose base (= middle/late Mio- cene boundary of Bolli and Bermudez, 1965) corresponds t o his Datum VIII at the top of the Gilbert a (= Cochiti) Event, 3.7 m.y. ago (see Fig. 8).

The relationship of late Neogene planktonic foraminifera1 biostratigraphy in the Caribbean, Gulf of Mexico, and the Mediterranean region has been examined in a series of recent papers by Beard and Lamb (1968), Beard (1969), Lamb (1969,1971), Lamb and Beard (1972). These authors conclude (Beard and Lamb, 1968; Lamb and Beard, 1972) that the Pliocene/ Pleistocene boundary in the Gulf of Mexico occurs a t the first indication of climatic deterioration (that is, glaciation) which corresponds to the extinction of warn water species (Globoquadrina altispira) and the appearance of L L ~ . ? w t c r speck. (Glohorotalia iiiflnta). This occurs at the base of the ILiLAa I<ornial E\ei>t (2.8 11i.y.) in the Gauss Normal Epoch. By implication this aiso marks the base of the Nebraskan Stage in the marine environment (Beard and T,amh. 1968; Beard. 1969; Lamb and Beard, 1972). Cooke (1972) documents evidence, howevrr, t o suggest the presence of a pre-Ne- braskan Sierran glacial event a t this time (see additional discussion in section below on glacial chronology).

The Miocene/Pliocene boundary was shown to correspond to the level a t which Globorotalia tumida, G. multicamerata, and G. margaritae first appear and was calibrated t o approximately 6 m.y. B.P. This level is seen to be close to that applied in the Mediterranean sequence, e.g. Blow (1969), Berggren (1969a,b,c, 1971a,b, 1972a,b, 1973), Hays et al. r1969), Cita (1973,1974). Thus we are confronted with the following interesting situation if we compare the work of the different investigators discussed above (see Fig. 8).

(1) According to the Houston research group, extinction of Globoqua- drina akispira (only slightly subsequent t o the initial appearance at 3 m.y. of Sphaeroidinella dehiscens with a marginal flange) defines the Pliocene/ Pleistozene boundary and occurs a t 2.8 m.y. The Pliocene spans the time interval of ca. 6-2.8 m.y. (Beard and Lamb, 1968; Lamb and Beard, 1972).

(2) In the view of the California workers a close association of several biostratigraphic events - the Sphaeroidinella dehiscens Datum (that is first appearance of Sphaeroidinella dehiscens with marginal flange), the extinction of Sphaeroidinellopsis subdehiscens and Globoquadrina altispira, of Prunopyle titan, and the initial appearance of Pulleniatina obliquiloculata - all of which occur within the approximate time interval of 3-2.7 m.y. in the paleomagnetic tihe-scale, serves to define the Miocene/Pliocene boundary, which is placed at about 2.7-2.8 m.y. (Bandy and Casey, 1969a,b; Bandy, 1971, 1972b,c; Bandy et al., 1971b). The Pliocene/Pleistocene boundary defined on the basis of the evolutionary transition between Globorotalia tosaensis to Globorotalia truncatulinoides is associated with the Gilsa (= Olduvai) at ca. 1.8 m.y. Thus the Pliocene covers from ca. 2.7 t o 1.8 m.y.

73

Facetiously stated, then the Miocene/Pliocene boundary = the Pliocene/ Pleistocene boundary and the entire Pliocene = the Early Pleistocene. In actual fact, the “Pliocene” of Bandy which is equivalent to the “early Pleistocene” of Lamb and Beard is equivalent to the Late Pliocene as typified in the Mediterranean region.

Mediterranean D.S.D.P. results

The Miocene/Pliocene boundary was cored at three sites in the Mediterra- nean Sea on D.S.D.P. Leg 13: the Balearic Abyssal Plain in the western Mediterranean (Site 134), the Tyrrhenian Basin in the central Mediterranean (Site 132), and Ionian Basin in the eastern Mediterranean (Site 125) (Cita, 1973). A sharp sedimentary hiatus (deepwater marls overlying gypsiferous evaporites and interbedded marls) and paleontologc discontinuity characterizes the houndary between Miocene and Pliocene strata. Marine microfossils and brackish-water ostracods and diatoms were recovered in the late Miocene clays, and late Miocene evaporites were cored in the Valencia Trough north of Majorca (Site 122), the Balearic Basin (Site 134), the Sardinian Slope (Site 133), and in the Levantine Basin (Site 129) (Ryan, Hsii et al., 1970, 1973). The regional unconformity at the top of the Messinian has been correlated with a widespread seismic reflector - Horizon M (Wong and Zarudzki, 1969; Ryan et al., 1971; Ryan, Hsii et al., 1970,1973). A sharp contact between the uppermost evaporite beds and the overlying pelagic oozes has been interpreted by Leg-13 scientists as indicative of a sudden inundation of a previously partially desiccated Mediterranean Basin by the Atlantic Ocean, in confirmation of earlier observations indicating regional regression and deep erosion in late Miocene sequences on land (Ruggieri, 1967).

As Cita (1974) indicates, the sudden inundation by the Atlantic Ocean is indicated by a sharp lithologic change in the deeper part of the Mediterra- nean Basin (the Trubi Marl lithology), whereas on the uneroded nart of the continental shelf a distinctive basal conglomerate generally formed as a response to the Pliocene transgression of the sea on the exposed continental margins. Precise, high-resolution biostratigraphy suggests that the oldest Pliocene sediments preserved in the various basins are not synchronous in each instance. The oldest post-unconformity sediments recovered from the deep sea are from the Tyrrhenian Basin (Site 132) and belong to the Sphae- roidinellopsis Acme Zone and to the Ceratolithus tricorniculatus Zone. Paleo- magnetic investigations on D.S.D.P. cores from Leg 13 suggest that this level is referable to the upper part of Epoch 5 (greater than 5 m.y.) (see Appendix, note 3). Several estimates on the age of the Miocene/Pliocene boundary were subsequently made by Cita (1973) :

(1) At site 132 a sedimentation rate of 3.5 cm/1,000 years was found in the Pleistocene. Downward extrapolation of this rate from the Pliocene/ Pleistocene boundary, based on the extinction of Discoaster brouweri, through the continuously cored section to the top of the evaporite sequence

74

yields an estimated date of 5 m.y. for this boundary. (This estimate, though remarkably accurate in light of other information, is founded on the rather weak assumption of a constant rate of sedimentation for over 2 m.y.).

(2) A minimum date of 4.7 m.y. for the Miocene/Pliocene boundary is suggested by that date on the Oriciatico mafic trachyte near Pisa, Tuscany (Tongiorgi, in Selli, 1970) which metamorphoses lower Pliocene clays. A maximum limit of 6 m.y. for this boundary is suggested by the Ischia “green tuff” of that age at Monte Epomeo, Italy (Civetta et al., 1970) which predates the Pliocene inundation (Rittman, 1948) (see Appendix, note 4).

(3) The upper boundary of the Sphaeroidinellopsis Acme Zone is estimated to lie at the Epoch 5/Gilbert boundary at about 5.1 m.y. Extrapola- tion of sedimentation rate to the top of the evaporite below the Sphaeroid- inellopsis acme Zone yields an estimate of 5.25-5.4 m.y. for the Miocene/ Pliocene boundary which lies within the upper part of Epoch 5 leading to an estimate of 5.4 m.y. for this boundary by Cita (1973, 1974).

Cita (1974) points out that the evolutionary transition between Globo- rotalia plesiotumida to Globorotalia tumida has not been recorded in the Mediterranean. She indicates that this apparently occurred when the Medi- terranean was desiccated and that G. tumida did not emigrate into the Medi- terranean after the opening of Gibraltar because of climatic exclusion. The data presented below indicates that the evolution of G. tumida occurred during earliest Pliocene time, however.

Nanno fossil criteria for MiocenelPliocene boundary

In terms of the calcareous nannofossils the Miocene/Pliocene boundary has been placed at the base of the Ceratolithus tricorniculatus Zone (“12) (Bukry, 1972d), within the C. tricorniculatus Zone (Gartner, 1969; Bukry, 1971a, 1973), at the Ceratolithus tricorniculatus/C. rugosus (NN12/NN13) boundary (Bukry, 1972b, c), or within the Ceratolithus rugosus Zone (“13 (Martini and Worsley, 1970; Martini 1971; Bukry, 1971b, 1972a) (Fig. 10).

Gartner (f973) has recently calibrated eleven distinctive calcareous nanno plankton datum levels to the paleomagnetic time-scale. Of particular importance to our discussion of Pliocene biostratigraphy are the following:

Time @.Ye)

1.8 2.1 2.4 3.6 3.95 4.45 4.9 5.7

Datum level

Discoaster brouweri Discoaster surculus Reticulofenestra pseudoumbilica Ceratolithus tricorniculatus Discoaster asymmetricus Ceratolithus rugosus Ceratolithus amplificus Discoaster quinqueramus

First Occurrence (FO), last Occurrence (LO)

LO LO LO LO FO FO FO LO

75

I E O U S MIOCENE N K T O N

BUKRY (197lb)972dYARTlNl 8 WORSLEI 119701 Y b R T l N l 11971) Ceratolithus

f g F b C e r e l o l i l h u i BUKRY(1972d,GARTNER 11973)

BUKRV 11971a,19731 GARTNER 11969) 2:

B U K R Y (1973)

Fig. 10. Placement of Miocene/Pliocene boundary by calcareous nannoplankton biostratigraphers.

Gartner (1973) places the Miocene/Pliocene boundary at the Ceratolithus rugosus Datum (base Zone "13). This was based on an earlier study (Gart- ner, 1969) in which it was shown that the Ceratolithus rugosus Datum occurred in CAP 38 BP in the southwest Pacific at a level (625 cm) within Zone N18 as dated by Parker (1967).

Hays et al. (1969, p. 1500, 1501) have observed that Sphaeroidinella dehiscens ranges well below the age of 4.2 and 4.1 m.y. that they estimate for the level determined by Parker (1967) and Blow (1969), respectively, for the N18/N19 boundary in core LSDH 78P (see Fig. 9). In fact Hays et al. (1969, p. 1501) indicate that the Miocene/Pliocene boundary must be older than 4.5 m.y., the age of the bottom of their core, V24-59, in which Sphae- roidinella dehiscens was present. Gartner (1973, Fig. 2) records the initial appearance of Ceratolithus rugosus within the Gilbert c in cores V24-59 and RC12-66 at about 4.45 m.y. Thus the Miocene/Pliocene boundary, as determined by Parker (1967) and Blow (1969), himself, on core LSDH 78P and by Parker (1967) on CAP 38 BP is younger than the true Miocene/Pliocene using Blow's own criteria. Saito et al. (1974) have recently determined the Sphaeroidinella dehiscens and the Globorotalia tumida Datum at about 4.8 m.y. and 4.9 m.y., respectively, and this is in agreement with our data in the North Atlantic (D.S.D.P. Site 111; Berggren, 1973). The Miocene/Plio- cene boundary (= base Ceratolithus rugosus Zone at - 4.5 m.y.) of Gartner (1973) is too young and actually lies within the Early Pliocene.

cens Datum can be extremely difficult in some instances; thus if the first I t should be pointed out that the recognition of the Sphaeroidinella dehis-

16

appearance of S. dehiscens is not accurately determined miscorrelation by as much as a whole nannofossil zone may result. I t is for this reason that the first appearance of Globorotalia tumida and the extinction of Globoquadrina dehiscens (- 5.0 m.y.) are considered more reliable indicators of the.Mio- cene/Pliocene boundary here. The last occurrence of Globoquadrina dehiscens was recorded at 674-676 cm in CAP 38 BP (Parker, 1967, table 2) a little over a half a meter below the initial appearance of Ceratolithus rugosus at 625 cm (Gartner, 1969. fig. 1).

In terms of calcareous nannofossil biostratigraphy the following data are pertinent. Kukry and Bramlette (1968) recorded Ceratolithus tricorniculatus from the basal Messinian marls in the upper part of the type Tortonian section and the lower Pliocene Trubi Formation of Sicily, the basal Piacen- zian of the Torrente Arda in north Italy, and in the Tabianian nearby, while Gartner (1969) recorded this species from a level correlated with Zone N17 in the Pacific core CAP 38 BP. An overlap in the stratigraphic ranges of Ceratolithus tricorniculatus and C. rugosus was observed between 265 and 545 cm in core CAP 38 RP (Bukry and Bramlette, 1968, p. 152); this interval is of early Pliocene age, equivalent to the basal part of Zone N19 (see Fig. 9).

Stradner (in Ryan, Hsu et al., 1973) does not record Ceratolithus tricorniculatus below the top of the Messinian evaporites, but its absence there is ascribed to adverse environmental conditions by Gartner (1973). The Dis- coaster quinqueramus Zone occurs within the marly intercalations of the evaporites at Mediterranean Site 132. The Ceratolithus rugosus Zone is correlated consistently with the Globorotalia margaritae evoluta lineage zone (Cita, 1973) or with the concurrent range of margaritae and puncticulata at about 4.5-4.6 m.y.

Bukry (1971h) correlated the Miocene/Pliocene boundary with the first appearance of Ccratolithus amplificus, a species intermediate between Cera- tolithus tricorniculatus and Ceratolithus rugosus. Gartner (1973) indicates that C. amplificus appears at a level (18 m ) not greater than 4.9 m.y. in core RC12-66 (see Foster and Opdyke, 1970; Saito et al., 1974). This level is coincidental with the base of Zone N18 as determined by Saito et al. (1974) in RC12-66 (see Fig. 9).

The co-occurrence of Ceratolithus tricorniculatus and Ceratolithus amplificus is restricted t o the lower part of the Lower Pliocene (Zanclian) in Sicily (Gartner, 1973). Ceratolithus amplificus has its initial appearance approximately 6 m above the base of the Zanclian at the newly proposed type section at Capo Rosello (S. Gartner, personal communication, 1973). If this level corresponds to the level dated at about 4.9 m.y. in core RC-66, which is approximately equivalent to the base of Zone N18, then the Miocene/Plio- cene boundary may be said to be somewhat older than this level and an estimate of 5.0 m.y. is suggested here. I t should be recalled that Banner and Blow (1965b) and Blow (1969) recorded the initial appearance of Sphaeroid- inella dehiscens about 1 2 m above the base of the Zanclian also and noted that the Miocene/Pliocene boundary was thus somewhat older and probably

within Zone N18. As we see here it actually lies at the base of Zone N 1 8 or even slightly lower. For all practical purposes it may be drawn at the base of Zone N18. Thus the Miocene/Pliocene boundary lies within the stratigraphic range of Ceratolithus tricorniculatus (NN12), and Bukry (1971a, p. 305), in subdividing the C. tricorniculatus Zone into three subzones - from the bottom): Triquetrorhabdulus rugosus, Ceratolithus amplificus and C'erato- lithus rugosus - drew the Miocene/Yliocene boundary at the base of the C. amplificus Subzone.

In summary it would appear that a set of multiple criteria have been developed for helping to correlate the Miocene/Pliocene boundary in a manner similar to that suggested for the Pliocene/Pleistocene boundary (Hays and Berggren, 1971). These criteria include: planktonic foraminifera - extinction of Globoquadrina dehiscens, first evolutionary appearance of Sphaeroidinella dehiscens and Globorotalia tumida; calcareous nannoplankton - first evolutionary appearance of Ceratolithus amplificus. These multiple criteria occur just above the end of Epoch 5 on the paleomagnetic time-- scale, a point dated at about 5.1 m.y. (Foster and Opdyke, 1970). For practical purposes the Miocene/Pliocene boundary may be said to occur within the interval of 4.9-5.1 m.y. (see Fig. 4).

The ages on the remainder of Gartner's (1973) datum levels agree well with those obtained in current studies on the North Atlantic (Orphan Knoll, Site 111; Berggren, 1973) with the exception of the Reticulofenestra pseudoumbilica (extinction) Datum at 2.4 m.y. This level occurs near the bottom of Zone N21 (Gartner, 1969) and below Datum V, Sphaeroidinel- lopsis extinct*ion, at the end of the Mammoth Event, about 2.95 m.y. of Saito (in Hays e t al., 1969) in D.S.D.P. core 3/12C/4/2 near Cape Verde (Cita, 1971), and corresponds to the initiation of major, if not continental glaciation in the Northern Hemisphere (Berggren, 1972b; Cooke, 1972). Thus we consider the extinction date of 2.4 m.y. suggested by Gartner (1973) as somewhat too young and, indeed, Gartner (1973) himself notes that near the top of its range this species is rare and represented mostly by relatively small specimens. It may be either that: (a) the extinction level of R. pseudoum bilica determined by Gartner is too high and that in the upper part of the stated range of this species its presence is due to reworking, or (b) that the extinction level at 2.4 m.y. is real and that previous determinations on the R, pseudoumbilicalD. surculus boundary are incorrect due to failure to recognize the latest part of the range of R. pseudoumbilica. The weight of the evidence cited above suggests that the correct alternative lies in (a) above and for this reason we assign an age of - 3 m.y. to the R. pseudoumbilicalD. surculus boundary here (see Fig. 4).

Pliocene geochronology of the Western Pacific

Kennett et al. (1971) established a paleomagnetic chronology for the Pliocene to early Pleistocene sequence of the marine sediments of New

Zealand. Major temperature fluctuations were recognized from the upper Gauss (mid-Pliocene) to the middle Matuyama (Early Pleistocene). The first observed major Pliocene cooling in the Austral seas spans the Gauss/Matuya- ma boundary at 2.4 m.y. Pronounced cold phases were observed in the Middle Pliocene Waipipian and Late Pliocene Mangapanian stages and at the beginning of the Pleistocene in the lower Hautawan Stage.

The authors suggest a possible relationship between paleoclimatic and geomagnetic polarity changes, since cooling trends commonly begin at or shortly after geomagnetic polarity changes at the end of the Kaena Event, 2.8 m.y., near the base of the Olduvai (= Rgunion Event in Kennett et al., 1971) at 2.13 m.y., in the upper “Olduvai” (= RLunion Event) at 1.98 m.y.; near the base of the Gilsa (= Olduvai of this work) at 1.79 m.y. Cooling maxima also coincide with the geomagnetic reversals at the Gauss/Matuyama boundary at 2.43 m.y. Climatic changes have been noted at, or close to, 6 of 8 Late Neogene polarity changes (see also Heirtzler, 1968; Kennett and Watkins, 1970).

Kennett e t al. (1971) correlated the Opoitian Stage with the Tabianian and the Waipipian and Mangapanian with the Piacenzian, and the Astian + Calabrian with the Hautawan. The top of the Tabianian (= Opoitian) was put at about 2.5 to 2.6 m.y., but if the top Tabianian is approximately equivalent to the upper limit of Globorotalia rnargaritae (as suggested by Cita, 1974) then the Tabianian/Piacenzian boundary is at about 3.4 m.y. or a bit lower. I t is indeed probable that the Opoitian is more or less equivalent to the Tabianian but the top of the Tabianian is probably not as high as the top of the Opoitian. Globorotalia inflata occurs in the Piacenzian but has not been recorded in the Tabianian. It appears for the first time in the upper part of the Opoitian (Kennett et al., 1971, p. 279, ref. 32). (In the North Atlantic Globorotalia inflata appears at about 3 m.y. ago coincident with the first glacially rafted sediments.) These data support the suggestion that the top of the Tabianian is not younger than 3 m.y. and are incorporated in the correlations between the New Zealand and European time-stratigraphic units that are shown in Fig. 1.

In Australia the Pliocene/Pleistocene boundary has been traditionally placed a t the base of the Werrikooian Stage (Gill, 1957, 1961,1968).

Asano et al. (1974) pointed out that there are two distinct biofacies in late Neogene sequences of Japan: the tropical and subtropical one on the eastern side of Japan and a cool temperate one on the western side of Japan. The distribution of these biofacies suggests current patterns similar to today, especially on the Pacific side. In the case of the Japan Sea side a cool-water biofacies was more significant during the Late Neogene than at the present time. A different planktonic foraminifera1 zonation has been developed in the Japan Sea from that of the Pacific side (Shinbo and Maiya, 1971).

Several datum planes have been described and recognized in the Pliocene and Pleistocene sequence on the Pacific side of Japan. The position and characteristics of these datum planes and their estimated ages are shown in Table XI below.

79

TABLE XI

Relationship of calcareous planktonic datum levels t o paleomagnetic time-scale in Japan (modified from Asano e t al., 1974)

Datum

A

B

C

G

H

Datum level

first evolutionary appearance of Sphaeroidinella dehiscens from Sphareoidinellopsis subdehiscens

extinction of Sphenolithus abies

initial evolutionary appearance of G. tosaensis from its ancestor G. crassaformis

Pulleniatina coiling change from dextral t o sinistral

initial appearance of Gephy rocapsa

first evolutionary appearance of G. truncatulinoides from its immediate ancestor G. tosaensis

extinction of discoasters and sinistral to dextral coiling change in Pulleniatina

extinction horizon of Pseudoemiliania lacunosa

Characteristics

Pulleniatina dominated by sinistrally coiled forms

Pseudoemiliania lacunosa appears near this level

Pseudoemiliania lacunosa abundant a t this level; dextrally coiled Pulleni- atina (rare); above datum discoasters become reduced in number and D. brouweri dominant

in approach of datum discoasters sharply reduced and only brouweri present

Estimated age (m.y.)

4.8

“3.0

3.0

2.3

2.0

1.8

1.65

0.35

Late Neogene planktonic foraminifera biostratigraphy of Taiwan has been discussed in several papers by Chang (1960,1962,1967,1974) and Huang (1967,1974). In general the sequence of planktonic foraminiferal assemblages is similar to that which occurs in the Indo-Pacific tropical area, and the zonation scheme used in those areas are of use in Taiwan also (Huang, 1974).

THE PLEISTOCENE EPOCH AND THE PLIOCENE/PLEISTOCENE BOUNDARY

A summary of investigations dealing with Quaternary boundaries and

80

correlations up to 1967 was presented by Hays and Berggren (1971). On the basis of data available at the time the following conclusions were reached with regard to the position and age of the Pliocene/Pleistocene boundary.

(1) The boundary can be determined paleontologcally by using extinctions or appearances of one or more of a number of microfossils.

(2) These events, though not simultaneous, are linked with the Olduvai Normal Event of the Matuyama (at that time estimated to have an age range of 1.80-1.95 m.y.; but see below).

(3) On the basis of the above, the Pliocene/Pleistocene boundary was determined to be about 1.85 m.y. old as defined by the first evolutionary appearance of Globorotalia truncatulinoides (Berggren et al., 1967).

Beginning of the Pleistocene in the Calabrian stratotype

One of the criteria selected for the correlation of the Pliocene/Pleistocene boundary is the extinction of discoasbers, i.e. Discoaster brouweri (Ericson et al., 1963; Riedel et al., 1963; Wray and Ellis, 1965; Berggren et al., 1967; Hay et al., 1967; Hays and Berggren, 1971). The extinction of D. brouweri was shown to be related to the Olduvai Normal Event by several authors (Berggren et al., 1967; Glass et al., 1967). More recently the relalionship of this extinction level to the Olduvai has been reaifirmed in equatorial Pacific cores by Hays et al. (1969), in the Philippine Sea (Takayama, 19i313) and to a level at Le Castella, Italy, interpreted by Emiliani e t al. (1961) as 1:orrelating to the base of the Calabrian (Takayama, 1970).

An investigation by Takayama (1969, 1970) on the calcareous nannoplankton at Le Castella has an important bearing on the paleontoiogic recognition of the Pliocene/Pleistocene boundary. He indicates that the first appearance of the genus Gephyrocapsa occurs in the Upper Pliocene (sample C4, about 50 m below the base of the Calabrian) at Le Castella, but that specimens of the genus become abundant in the Calabrian itself. Reworked discoasters are present throughout the Pliocene and Pleistocene at Le Castel- la but Takayama (op. cit.) indicates that the discoasters probably became extinct near the level of sample C3 (just below the Pliocene/Pleistocene boundary). McIntyre e t al. (1967) had observed that in deep-sea cores in the North Atlantic the first appearance of Gephyrocapsa occurs near the extinction level of discoasters and this was confirmed by Takayama (1970) in core V21-98 from the Philippine Sea. Thus, there is a slight overlap in the range of Gephyrocapsa spp. and Discoaster brouweri at Le Castella, although in view of the high sedimentation rate represented in this area the overlap may indeed be more apparent than real; at any rate it is probably of short duration.

Bandy and Wilcoxon (1970) have also investigated the distribution of microfossils a t Santa Maria di Catanzaro and at Le Castella. They report:

(1) Discoasters become extinct at Santa Maria di Catanzaro, the stratotype of the Calabrian, near the base of the “sandy Calabrian”. The “sandy Ca-

81

labrian” of Bandy and Wilcoxon (1970) is also referred to by Selli (1971, p. 58). It is 40 m thick and encompasses samples 38 to 30 of Bandy and Wilcoxon (1970, fig. 4). It is not the same lithologic unit referred to by Bayliss (1969, p. 138) as “9 (vel 5.2) * meters of interbedded silts and soft sands .... etc.” which is at the base of the stratotype.

(2) Pseudoemiliania lacunosa, Gephyrocapsa caribbeanica and Discolithina japonica range concurrently through the 80-m section represented.

(3) Hyalinea balthica first appears in the “sandy Calabrian”. (This agrees with Bayliss (1969, fig. 3) who recorded the first appearance of H. balthica in the lowest exposed parts of the section, that is to say the so-called “sandy Calabrian” along the road to the west of Santa Maria di Catanzaro.)

(4) Globorotalia truncatulinoides first appears in samples about 80 m above the lowest occurrence of H. balthica. If the base of the “sandy Cala- brian” of Bandy and Wilcoxon (1970) is the same as the base of the section described by Bayliss (1969), then the level from which Bandy and Wilcoxon (1970) record the first truncatulinoides would be at about the level of sample By 42 (Bayliss, 1969, fig. 2 on p. 136), i.e.-about 21 m above the base of the stratotype Calabrian. In fact this is precisely the level at which Bayliss (1969, figs. 2, 3) recorded the first truncatulinoides.

sequence at Santa Maria di Catanzaro represents most of the section exposed there but only part of which represents the stratotype. The initial appearance of Hyalinea balthica in the lower part of the “sandy Calabrian” is therefore not a defining characteristic of the Calabrian Stage/Age, inasmuch as Bayliss (1969) has shown that this form ranges to the bottom of the exposed section at Santa Maria di Catanzaro, at least 67 m below the first appearance of Arctica islandica.

The choice of Hyalinea balthica is also unfortunate as the distribution of this form is strongly controlled by environment, so that it can hardly be used for the delineation of a time-equivalent horizon (Poag, 1972). A similar sequence of events was recorded at Le Castella (Bandy and Wilcoxon, 1970, p. 2942, 2943). A single specimen of Globorotalia truncatulinoides was reported in a sample about 2 m above the base of the Calabrian as determined by the first occurrence of Hyalinea balthica. However, the highest occurrence of discoasters comes between these two levels so that it would appear that we are indeed dealing here with a horizon at or near the lower Calabrian boundary at Santa Maria di Catanzaro. The overlap in Gephyrocapsa caribbeanica and discoasters (primarily D. brouweri) in the lower part of the section at Le Castella mirrors the findings of Takayama (1969, 1970). The findings of Bandy and Wilcoxon (1970) support the conclusions that the upper Santa Maria di Catanzaro section and that of Le Castella are age- equivalent .

It would appear that the samples collected by Bandy and Ascoli from the

Smith (1969) had previously investigated the discoaster distribution at Le


82

Castella and Santa Maria di Catanzaro and concluded that the two localities were not stratigraphically equivalent; Le Castella was considered to be uppermost Calabrian, Emilian and basal Sicilian in age. His anomalous results were interpreted by Nakagawa et al. (1971) as possibly due to incomplete sam- pling. Whereas Smith (1969, p. 580) indicates that “approximately 40 m of clays and shales are exposed just west of town (Le Castella)”, Nakagawa et al. (1971, p. 362) indicate that about 500 m of Pliocene-Pleistocene sediments are exposed in the cliffed coastline extending NNW-SSE of Le Castel- la; of this, the lower 170 m are middle Pliocene, the middle 250 are late Pliocene, and the upper 100 m are Pleistocene (Calabrian).

On the basis of the study of the calcareous nannofossil flora, Smith (1969) concluded that the major part of the stratigraphic sequence exposed at Le Castella belong to the Emilian (warm) Stage and assigned the uppermost beds to the Sicilian (cool) Stage. The joint occurrence of Gephyrocapsa Carib beanica and Discoaster brouweri throughout the section was recorded. The Emilian, in turn, was correlated with the marine Aftonian interglacial Stage of North America. However, the Sicilian Stage of Italy has been correlated with the “Cromerian” mammalian fauna of N. Europe (Ambrosetti, 1967; Bonadonna, 1968), the base of which has, in turn, been calibrated against the Brunhes/Matuyama boundary (0.7 m.y.) by Van Montfrans (1971a,b). The youngest reliable dates on the Calabrian Stage are about 1 m.y. (Ambrosetti et al., 1972), which is also approximately the upper limit of the Villafranchian mammalian age. These dates suggest that the Emilian marine Stage is a short interval within the latest Matuyama, approximately equivalent t o the interval between the Jaramillo Normal Event and the base of the Brunhes Normal Epoch. Discoasters, however, became extinct near the top of the Olduvai-Gilsa Normal Magnetic Event (about 1.6 m.y. ago); the overlap of Gephyrocapsa caribbeanica and Discoaster brouweri in deep- sea cores occurs in the interval between about 2 m.y. and 1.6 m.y. (see also Azzaroli, 1970).

Nakagawa et al. (1971) have presented the preliminary results of paleomagnetic measurements on the Pliocene-Pleistocene section exposed at Le Castella. The presence of the radiolarian Eucyrtidium elongatum peregrinum, the diatom Thalassiosira conuexa, and the predominance of Discoaster brouweri in the lowest sample ((310) of the “Pliocene superiore” suggests that the base of the Matuyama Reversed Polarity Epoch is within this interval. This is because the radiolarian species appears t o become extinct worldwide within the Gauss Normal, just above the Kaena Event, a t about 2.7 m.y. and the diatom species disappears within the early part of the Matuyama Reversed Epoch, a t about 2.2-2.3 m.y. (Hays et al., 1969; Burck- le, 1972).

The Japanese authors found reversed polarity in samples taken at and near the Pliocene/Pleistocene boundary. By plotting the possible age-range of the stratigraphic position of each sample against the magnetic stratigraphy vs the stratigraphic distance above or below the boundary and connecting all the

83

samples examined by an oblique line, they conclude that the Pliocene/ Pleistocene boundary is slightly younger than the Gilsa Event (Olduvai of this study), but they do not exclude the possibility that it might lie within a reversed split event near the middle of the “Olduvai” Normal Polarity Event (Re‘union Event of Gromme‘ and Hay, 1971). Micropaleontologic data (Berg- gren et al., 1967; Hays et al., 1969; Saito, 1969a,b; Hays and Berggren, 1971) strongly indicate that the Pliocene/Pleistocene boundary is associated with the first major paleomagnetic event older than the Jaramillo, namely the Olduvai-Gilsa. The paleomagnetic and micropaleontologic data are seen to be in general agreement. Nakagawa et al. (1971) indicate that their results are tentative and preliminary and that more detailed investigations are being conducted on the Le Castella section.

gy here. Nakagawa et al. (1969) have expressed doubt about the identification of the Olduvai Event in deep-sea cores because of the relation of sediment thickness to the absolute age of the magnetic reversals and because of the problem of recognizing this short event in a section of low sedimentation rate. In most cases only a single normal polarity event is seen in deep-sea cores between the Jaramillo and the base of the Matuyama Reversed Polarity Epoch. This is the Olduvai Normal Event of Berggren et al. (1967), Glass et al. (1967) and Hays et al. (1969), which corresponds to the Gilsa Normal Event of Cox (1969).

Gromme‘ and Hay (1971) showed that the normally magnetized volcanics upon which the Olduvai Normal Event is based had K-Ar ages such that, within the extremes of analytical probability limits, “...the span of time represented by these rocks might be as great as 1.6 to 1.9 m.y. or as little as 1.70 to 1.75 m.y.” (p. 183). These dates did indicate, however, that the normally magnetized material at ca. 2.0 m.y., which in deep-sea cores had been identified with the Olduvai Event, were too old to have this correlation, and that on the other hand the lavas dated at 1.60 f 0.5 m.y. upon which McDougall and Wensink (1966) had founded the Gilsa Event had actually been erupted during the Olduvai polarity event. As the name Olduvai has priority, in this context Gilsa = Olduvai. Gromme‘ and Hay (1971) went on to propose that the spurious “Olduvai” event be re-named the Rkunion Event (s).

Continued work on the Olduvai volcanics was reported by Curtis and Hay (1972), who based their conclusions on more than 50 individual K-Ar analyses. The uppermost part of the normally magnetized sequence appeared not to be younger than 1.71 m.y. I t thus seems possible that a short reversed interval separated the Olduvai Normal Event, in the original sense, from a period of normal polarity at ca. 1.61 m.y. (Cox, 1972; Dalrymple, 1972; Watkins, 1972). In this context, Olduvai = Olduvai + Gilsa, covering a time span of roughly 0.21 m.y. from 1.82 to 1.61 m.y. and is shown as a single such normal polarity interval on the chart (Fig. 11, p. 93).

It is important to clarify our use of paleomagnetic epoch-event terminolo-

A preliminary re-examination of the basalts and tuffs at Olduvai Gorge by

84

Lamb (1969, p. 563) records Hyalinea balthica “..in the lower samples of the Calabrian (i.e., in the lower sandstones in the Calabrian at this locality), whereas Globorotalia truncatulinoides occurs first some 30 meters higher in the section, or some 70 meters below the horizon at which it was reported by Bayliss”. Lamb (1969) is somewhat ambiguous in this report for he does not indicate where his material was collected with reference to the three stratigraphic sections measured by Bayliss (1969, fig. 3, p. 137-139) which he is discussing. Lamb (1969) indicates in a subsequent paragraph that “... Globorotalia truncatulinoides appears for the first time somewhat above the base of the Calabrian ...”, evidently meaning the definition based on the first appearance of H. balthica. In a subsequent paper, however, Lamb and Beard (1972, p. 18) revise this to state that in sections along the road from Santa Maria di Catanzaro t o Caraffa di Catanzaro and Cortale, “ ... Hyalinea balthica occurs in the lower samples of the Calabrian (i.e., the lower sandstones at this locality), whereas Globorotalia truncatulinoides occurs first some 30 meters higher in the section, which is some 22 meters below the horizon at which it was reported by Bayliss ...”. Bayliss (1969) recorded the first occurrence of G. truncatulinoides about 30 m (vel 21 m, fide D. Bayliss, personal communication, 1972) above the base of the stratotype Calabrian as defined by the marker bed G - G ’ of Gignoux (1913, p. 34). If Lamb found G. truncatulinoides 22 m (let alone 70 m) below Bayliss’ recorded level it would be in basal Calabrian sediments very close to the G-G’ equivalent sandstones (see below) and thus at or near the currently accepted base of the Pleistocene.

of this definition and go on t o say that, “...although the base of the ‘sandy’ Calabrian is not exposed, additional samples of some 10 meters of lower ‘sandy’ Calabrian above a 30 m covered interval from the churchyard of Santa Maria di Catanzaro provide data on the lower interval. Below the covered interval is more than 30 meters of highly fossiliferous siltstone considered to be Pliocene. Thus, on the basis of foraminifera1 evidence, the base of the Calabrian stage is not completely exposed at Santa Maria di Catanzaro because it falls within the 30 meters of covered section below the lower sandstones. This covered interval has Pliocene strata below and Pleisto- cene strata above ...” (italics added).

I t is important to stress that Lamb (1969, p. 563) and Lamb and Beard (1972, p. 18) believe that the initial appearance of Hyalinea balthica stratigraphically below Gignoux’s marker G-G’, “...effectively nullifies Gignoux’s boundary...”. We are of the opinion, however, that these authors have con- fused definition with recognition. In the rules of stratigraphic nomenclature, a lithostratigraphic plane such as that formed by marker G-G’ and its lateral equivalent must be the basis of the definition of a biochronostratigraphic entity such as the Calabrian Stage/Age. Accordingly, the base of the Cala- brian Stage is definitely exposed at Santa Maria di Catanzaro, since i t is clearly stated by Gignoux (1913, p. 35) t o be the base of his marker bed

However, Lamb and Beard (1972, p. 18-19) do not recognize the validity

85

G-G’. This forms a prominent outcrop on Punta dei Brigante 250 m NE of Santa Maria di Catanzaro, and is a calcarenite band containing abundant Arctica (= Cyprina) islandica which passes laterally into friable sands and silts. Various paleontological criteria have subsequently been used in attempts to recognize this level elsewhere but they are in no way part of the definition of the stratotype. The obvious ecological significance of the cold- adapted shallow-water foraminifer Hyalinea balthica appeals to authors who would equate the beginning of the Calabrian, and even more, the beginning of the Pleistocene, with the onset of global glaciation, and this preconcep- tion evidently has influenced the conclusions of Lamb and Beard (1972), but at most they have only presented an argument for revising the biostratigraphical context within which the base of the Calabrian, as originally defined, can be correlated.

It is also important to note that the “sandy Calabrian” referred t o by Lamb and Beard (1972) and other authors lies below Gignoux’s marker G-G’ and is thus neither Calabrian nor Pleistocene. With this in mind, let us review the data of Bandy and Wilcoxon (1970), Lamb (1969), Bayliss (1969) and Lamb and Beard (1972) in an attempt to clarify the problems of the Pliocene/Pleistocene boundary in its agreed association with the base of the stratotype Calabrian at Santa Maria di Catanzaro.

(1) Hyalinea balthica is recorded by all of the above authors throughout the section exposed in the sea cliffs at Santa Maria di Catanzaro, both above and below Gignoux’s marker G - G ’ and its equivalent facies (although, according to Bayliss, 1969, p. 139-140, “...no specimens of ‘Hyalinea balthica’ were found in the sandy horizons which are laterally equivalent to the calcarenite on Pta. dei Brigante where Arctica islandica has been found...”, meaning the marker bed itself). H. balthica is not recorded from fossiliferous siltstones below the 30-m covered interval in the Santa Maria di Catanzaro churchyard, which lies beneath the “sandy Calabrian”. It therefore makes its local appearance in the Upper Pliocene.

(2) Bayliss (1969) made an exhaustive examination of samples from three measured sections at Santa Maria di Catanzaro, in which Globorotalia truncatulinoides was found only in Calabrian sediments, beginning about 21 m above the marker bed G - G ’ and its lateral equivalents. The record of Bandy and Wilcoxon (1970) appears t o agree with that of Bayliss. However, Lamb (1969) and Lamb and Beard (1972) indicate that this species is recorded about 30 m above the lowest observed occurrence of Hyalinea balthica, which would place it immediately below Gignoux’s marker G-G’. Whether or not this discrepancy can be resolved i t can be concluded that the initial appearance of Globorotalia truncatulinoides, a planktonic form, is a better guide t o recognizing the proximity of the Pliocene/Pleistocene boundary in other parts of the world than the benthic, ecologically sensitive Hyalinea balthica with its lithofacies-linked distribution.

86

Climatic definition o f the Pleistocene

Beard and Lamb (1968) indicate that the Pliocene/Pleistocene boundary in the Gulf of Mexico occurs at the first indication of severe climatic deterioration (i.e. glaciation), which corresponds to the extinction of warm-water species (Globoquadrina altispira) and the appearance of cold-water species (Globorotalia inflata). This occurs at the base of the Kaena Normal Event (2.8 m.y.) in the Gauss Normal Epoch. In their interpretation, this would also mark the base of the Nebraskan Stage in the marine environment (Beard and Lamb, 1968; Beard, 1969; see also Poag, 1972, for an opposing interpretation).

Furthermore, Beard (1969, p. 588) states that “the Pliocene/Pleistocene boundary must be defined on the basis of climatic deterioration as reflected by the change from warm- to cold-water faunas rather than on the basis of evolutionary changes”. Lamb (1969, p. 568) exhibits a similar misunder- standing of the basis for correlation of chronostratigraphic boundaries by observing that the “coiling changes in the G. menardii complex enjoyed popularity for defining the Pliocene/Pleistocene boundary until McIntyre et al. (1967), as well as others, discovered that this datum is the beginning of a warm period following a cool period and, thus, is not a satisfactory horizon at which to begin the Pleistocene”.

nition of chronostratigraphic boundaries (see Hedberg, 1961, 1968; George et al., 1969; Report of Stratotype Working Group, Comm. Mediterr. Neo- gene Stratigraphy, Proc. 4th Session, Bologna (1967), 1970). Chronostrati- graphic boundaries are not determined by “convenience”. They are. They exist. The most reliable means for the recognition and correlation of chronostratigraphic boundaries is the fossil record, and Van Hinte (1969, p. 271) has stated the case well in observing that “a succession of phylozones is the most reliable tool in correlation and age determination, because it directly reflects the irreversible evolution of life on earth providing maximum exclusion of the environmental factor” (italics ours). In actual fact the initiation of glaciation plays no role in the definition and determination of the Pliocene/ Pleistocene boundary (Hays and Berggren, 1971). The marine Calabrian Stage is a chronostratigraphic unit with a clearly defined lower lithostratigraphic boundary defined locally by a stratotype section (Santa Maria di Catanzaro). Even the Pliocene/Pleistocene boundary has its own type section (Le Castella), although this seems unnecessary, if not illogical, to us. There is no significant climatic change associated with the Pliocene/Pleistocene boundary although there are numerous oscillatory climatic cycles both below and above this boundary in the marine record (Hays et al., 1969; Ruddiman, 1971; Keany and Kennett, 1973). Correlation of climatic cycles cannot be made without recourse to an independent time-scale afforded by either trans-specific evolutionary changes, extinctions and/or an absolute time-scale. For instance, if climatic cycles are out of phase from one area to

These statements are remarkable in the light of current work on the recog-

87

the other there is no means of recognizing the diachronism independently. In short, climatic changes can hardly be considered suitable for the definition and determination of chronostratigraphic boundaries, and as a simple matter of logic should not be incorporated as primary criteria in an otherwise bio- logically based system.

The investigations of Berggren et al. (1967), Glass et al. (1967), Hays et al. (1969), Hays and Berggren (1971) and Poag (1972) indicate a close relationship between the initial appearance of Globorotalia truncatulinoides and the extinction of discoasters. This relationship has been further documented in the course of several cruises of the “Glomar Challenger” (Deep Sea Drilling Project Initial Reports), where it is shown to occur with the Olduvai-Gilsa Event (between 1.61-1.82 m.y.). In the stratotype Calabrian Globorotalia truncatulinoides has been reported 21 m above the base of the Calabrian (Bayliss, 1969). Lamb (1969, p. 563) records Hyalinea balthica “in the lower samples of the Calabrian (i.e., in the lower sandstones in the Calabrian at this locality), whereas Globorotalia truncatulinoides occurs first some 30 meters higher in the section, or some 70 meters below the horizon at which it was reported by Bayliss”. Lamb (1969) is somewhat ambiguous here for he does not indicate where this material was collected, at “Santa Maria di Catanzaro or at Caraffa di Catanzaro and Cortale” (loc. cit.). In this connection the reader should see Bayliss (1969, fig. 3, p. 137-139) for the three stratigraphic sections measured by Bayliss, and their coordinates. Lamb (1969) indicates in a subsequent paragraph that “Globorotalia truncatulinoides appears for the first time somewhat above the base of the Calabrian ...”, evidently meaning the first local appearance of H. balthica.

Age of the base of the Pleistocene

If we now try to determine the age of the base of the Calabrian Stage (i.e. the Pliocene/Pleistocene boundary) as stratotypified at Santa Maria di Catan- zaro we can proceed along the following line of reasoning:

(1) The presence of the radiolarian species Eucyrtidium elongutum peregrinum, the diatom species Thalassiosira conuexa and the predominance of the discoasterid D. brouweri in the lowest sample (C10) of the “Pliocene superiore” at Le Castella suggests that the base of the Matuyama Reversed Polarity Epoch is within the Late Pliocene (Nakagawa et al., 1971). E. elongatum peregrinum became extinct within the Gauss Normal, just above the Kaena Event, at about 2.7 m.y.; T. convexa became extinct within the early part of the Matuyama Reversed Epoch, at about 2.2-2.3 m.y. (Hays et al., 1969). The Pliocene/Pleistocene boundary is thus younger than 2.2-2.3 m.y.

(2) The first appearance of Gephyrocapsa spp. occurs in sample C4, about 50 m below the base of the Calabrian at Le Castella as determined by Emilia- ni et al. (1961) but specimens of the genus first became abundant in the Calabrian itself (Takayama, 1969,1970; see also Bandy and Wilcoxon, 1970). McIntyre et al. (1967) observed that the first occurrence of Gephyro-

88

capsa spp. occurs near the extinction level of discoasters and this was confirmed by Takayama (1969) in core V21-98 from the Philippine Sea.

(3) The extinction level of discoasters is near the top of the Olduvai Event (Berggren et al., 1967; Phillips et al., 1968; this work). (4) Saito (1969a,b) has noted that the extinction of Globigerinoides

obliqua occurs near the Pliocene/Pleistocene boundary as determined by Emiliani et al. (1961) at Le Castella and that in deep-sea cores this event occurs near the base of the Olduvai Normal Event, dated a t about 1.9 m.y. Colalongo (1968a) and Dondi and Papetti (1968) made similar observations on the extinction of this form near the Pliocene/Pleistocene boundary in the Romagna Apennines and the Po Valley, respectively. In our material G. obliqua becomes extinct at the base of the Olduvai-Gilsa, dated at about 1.82 m.y.

(5) Globorotalia truncatulinoides makes its initial appearance slightly above (Bayliss, 1969) or slightly below (Lamb, 1969; Lamb and Beard, 1972) the base of the Calabrian as defined in Gignoux’s bed G-G’ (1913).

Although an evolutionary sequence from G. tosaensis to G. truncatulinoides has not been demonstrated in the Mediterranean region, this initial appearance may be close to the initial evolutionary appearance seen in deep- sea cores. In deep-sea cores the initial evolutionary appearance of G. truncatulinoides occurs within the Olduvai-Gilsa Normal Event (1.61-1.82 m.y.) (Berggren et al., 1967; Berggren, 1968).

(6) Thus we see that the following biostratigraphic events are closely associated with the Olduvai-Gilsa Normal Event (1.61-1.82 m.y.): initial evolutionary appearance of Glo borotalia truncatulinoides and of Gephyro- capsa spp., the extinction of Globigerinoides obliqua and of discoasters. G. truncatulinoides has not been recorded in the Mediterranean region in pre- Calabrian levels so that the Pliocene/Pleistocene boundary cannot be older than the Olduvai Normal Event (- 1.8 m.y.). The present evidence suggests that the multiple criteria listed above occur within a relatively short time- interval which includes the base of the Calabrian as stratotypified at Santa Maria di Catanzaro (Gignoux, 1910,1913). This same evidence also strongly supports correlation of the base of the stratotype Calabrian at Santa Maria di Catanzaro to the base of the Calabrian at Le Castella as determined by Emiliani et al. (1961).

The use of Hyalinea balthica in correlating the base of the Pleistocene is seen to be unsound as it occurs well below the base of the stratotype Cala- brian at Santa Maria di Catanzaro (Bayliss, 1969, fig. 3 on p. 137, 139, 140, 142) even though only slightly below the marker bed of Emiliani et al. (1961) at Le Castella (Lamb and Beard, 1972, p. 23, fig. 10 on p. 24). Its distribution appears to be strongly controlled by environmental factors, and as a shallow-water dweller it is of limited value outside the Mediterranean littoral deposits (Selli, 1967; Poag, 1972).

of Globorotalia truncatulinoides and the extinction of Globigerinoides It would seem that the close coincidence between the earliest appearance

89

obliqua and discoasters offers a more reliable means of correlating the Plio- cene/Pleistocene boundary. Other paleontologic events can also be compared closely with these three events in various regions so that this boundary should now be recognizable over a large part of the world. Inasmuch as these paleontological events are also related to the Olduvai-Gilsa Normal Event, it might be appropriate to suggest that the Pliocene/Pleistocene boundary is located within the time span of this event (1.61-1.82 m.y.). This is approximately 200,000 years, a geologically short time and one which would allow the realistic use of multiple biostratigraphic criteria in recognizing the boundary where paleomagnetic data are not available.

a number of papers by Fleming (1950), Jenkins (1964,1971, 1973), Vella and Nicol (1970,1971), Kennett et al. (1971), among others (for a more complete bibliography, see Jenkins, 1973). Jenkins (1973) has summarized investigations on the position of the Pliocene/Pleistocene boundary in New Zealand and Australia and indicates that the placement of this boundary in these two countries has proceeded along two lines of investigation: climatic- deterioration criteria and evolutionary changes in the fossil biota. The criticism of Jenkins (1973) that the initial appearance of Globorotalia truncatulinoides in the Mangaopari section of New Zealand and its correlation at the base of the “Gilsa” (= Olduvai) to the base of the type Calabrian (Kennett et al., 1971) is “unacceptable” because G. truncatulinoides occurs only at the top of the type Calabrian is inaccurate (see above). The initial occurrence of G. truncatulinoides only 21 m above the base of the Calabrian in a basin of rapid sedimentation suggests that its evolutionary appearance (if not its first appearance in the Calabrian) is almost synchronous with the base of the Pleistocene and certainly can serve as a useful criterion in recognizing the Pliocene/Pleistocene boundary elsewhere.

In New Zealand the climatically determined Pliocene/Pleistocene boundary was drawn in the Late Pliocene Waipipian Stage at about 2.5 m.y. B.P., whereas the paleontologically defined boundary near the base of the Hauta- wan Stage is correlated with the base of the “Gilsa” at about 1.8 m.y. (Ken- nett et al., 1971).

As a result of their investigations of the New Zealand section Kennett and his colleagues (1971, p. 179) conclude that “correlations of a climatically defined Pliocene/Pleistocene boundary when first marked climatic cooling is used are unlikely to be valid. The Pliocene epoch would be eliminated if the first marked late Cenozoic climatic cooling is taken as the beginning of the Pleistocene, because the first known major cooling took place in late Mio- cene time.”

The Pliocene/Pleistocene boundary in New Zealand has been discussed in

CHAPTER 6

LATE NEOGENE MAMMALIAN BIOCHRONOLOGY AND K-AR TIME- SCALE

Fig. 11 shows the relationship of Late Neogene mammalian biochronology to marine planktonic zonation, the K-Ar time-scale, and the geomagnetic polarity reversal sequence in the interconnected faunal realm of Eurasia, Africa, and North America. The radiometric ages applied to the mammalian succession are taken mainly from the research published by Evernden et al. (1964) for North America, Bishop et al. (1969) for East Africa, and numerous sources listed by Van Couvering (1972) for Europe and North Africa. Calibration of the marine sequence is discussed elsewhere in this study, but in this section we include K-Ar and paleomagnetic ages on near-shore and estuarine marine biostratigraphical sequences relating to the mammalian chronology, principally those listed by Konecnf and his co-workers (e.g., Konecnf et al., 1969), by Choubert et al. (1968), and by the Netherlands workers Van Montfrans (1971a) and Zagwijn et al. (1971).

B IOCHRONOLOGY VS BIOSTRATIGRAPHY

It is an often overlooked fact that the further apart two contemporaneous fossil sites may be, the less is the value of lithostratigraphic criteria in correlating them chronologically. Because of the relatively wide spatial dispersal of vertebrate fossils and vertebrate fossil localities this fact becomes apparent in all but most prolific exposures, so that the “zone” or “age” in mammalian paleontology dispenses with a lithostratigraphic basis (Wood et al., 1941; Tedford, 1970) and purely biochronological sequences of Cenozoic mammal ages with convenient paleontological characteristics are in widely accepted use in North America, the Himalayan foothills, and Europe, although in the last named area a consensus on the philosophy and terminology of Cenozoic mammalian biochronology has yet to be reached and many versions are in active contention.

In biochronology, emphasis is naturally placed on events which assist in correlating the regional sequence, such as major environmental changes or the rapid spread of a distinctive taxon. In our discussion of mammal chronology of the Late Neogene the most stressed events, referring to the Mediter- ranean area, are:

92

Time

0.6 m.y.

?1.6 m.y

ca. 2.5 m.y. ca. 4 m.y.

ca. 5 m.y.

ca. 10 m.y.

ca. 12 m.y.

ca. 12.5 m.y.

ca. 14 m.y.

Events

initiation of major continental glaciation (Mindel) in Euro- pean subpolar regions (end Biharian) initiation of major montane glaciation in European subpolar regions (Middle Villafranchian) appearance of Elephantidae in Europe (Early Villafranchian) appearance of microtine rodents in southwest Europe (mid-Ruscinian) inundation of Mediterranean Basin (earliest Ruscinian) preceded by desiccation of isolated Paleo-Mediterranean Sea (Late Turolian) initiation of relatively cool, summer-dry climate in temperate regions (Early Turolian) appearance of murid rodents in Paleo-Mediterranean area (Early Vallesian) appearance of Hipparion in Paleo-Mediterranean area (earliest Vallesian) isolation of Paratethys basins (earliest Oeningian)

The colonization of temperate lands by Homo quring the latter part of the 9 Neogene was undeniably influential on mammal pbpulations, but the timing

of the successive cultural stages during which this influence progressively increased is still very uncertain (Isaac, 1972) and its relationship to a general history of continental biochronology is beyond the scope of this review.

In treating the transitions between mammalian ages we follow the recommendation of the Stratigraphic Committee of the Geological Society of Lon- don (George et al., 1969) that “base defines boundary”, and consequently devote attention principally to the earliest part of each succeeding age. Quo- tation marks indicate a term that is used apart from its primary definition, i.e. “Vindobonian” land mammal age.

The K-Ar dates quoted in this paper are mostly given without the analytical uncertainty figure, or “error”, since different analysts mean different things by this number and some do not publish it at all. Readers should bear in mind that the radiometric age of a sample is only the tip of an experimental iceberg. All dates should be evaluated in context, if possible: the number and quality of confirmatory determinations, number and quality of stratigraphically related determinations, paleontological and geological control, physical condition of the samples, and so on. If it were possible to stan- dardize and quantify all these documentary variables they would make a more meaningful “uncertainty” figure than the generally reported one.

The 40K decay constants used to calculate K-Ar ages are not yet stan- dardized, and we have converted the reported ages in some papers to the “western” values obtained from using X E = 0.585 . 1CJ’’ and A,= 4.72.1CJ’O. We estimate that laboratories using the Soviet Academy constants will report ages about 6.5% older than the “western” laboratories (e.g., Bagdasarjan et al.,

NANNDR ANKTON 'ALEOMAG- in

TRATIGRAW my ERICSON 8 WOLLIN MARTINI

ZONATION NETIC

119681

1FiIOCENEI v r"- " 2 1

V NN 20 __ __

0 5 - N23 U

...7..- T

-"IY-- - -10-

a z

s NN 19

R . N22

"12 5 0 ~

EPOCH

EPOCH

N I7

N N l l

EPOCH

~

EPOCH

iPOCHS

I30

i6

EIBWEAN- AFRICAN

FAWAS

-

w z W 0 0 I- in W J P

-

w z W

0

0

J

0

-

- L

W z W

0

0

I -

EWKlPEAN

LAM) MAMMAL

AGES

Fig. 11. Relationship of Late Neogene mammalian biochronology to marine planktonic zonation, the K-Ar time-scale and the geomagnetic polarity reversal sequence in the interconnected faunal realm of Eurasia, North Africa and North America (see text for further explanation).

pp. 93-96. ONm-USPIAN NORTH NORTH AMERICAN WEST I PLANKTONIC T

CALIFORNU W S T FORAMINIFERAL in IEG~ON.X)VIET in AMERICAN KIAR DATES

AGElSTeGES AGES

MAMMAL

m y UNION MARINE

STAGES DATUM LEVELS C CONTINENTAL MARINEI,

97

1971), and that Swiss and West German workers will report ages about 1% younger (,e.g., Lippolt et al., 1963; Horn et al., 1972) over the approximate range 0-30 m.y.

EUROPE AND AFRICA

In the following pages we will concentrate on the Late Neogene of the Mediterranean drainage basin, with secondary attention to Northern Europe and sub-Saharan Africa. It is in this area that marine and continental biochronology and K-Ar dating are best combined, and where the classical sequences of the Neogene can be directly referred to.

It is necessary to briefly recapitulate the history of biostratigraphical and biochronological terminology used in Europe. Until relatively lately the subdivisions of European Cenozoic land mammal history were known under an array of marine stagefage names representing various places in western Eura- sia where there were real (or reputed) correlations between land mammal fossil beds and the stage stratotype. Vertebrate paleontologists then found it necessary to agree on new, mammalian definitions for these stage/age names to settle correlation and boundary problems in continental sequences. The inevitable result was that the continental and marine stratigraphic terminology, despite a beguiling similarity, came to have separate existences.

The inconsistency between marine and non-marine definitions of the lithostratigraphically based stages/ages has caused vertebrate paleontologists to propose or adopt a variety of new terms for subdividing Cenozoic land mammal history (e.g., Kretzoi, 1962 i. al.; Thaler, 1966; Tobien, 1970a; Cicha et al., 1972). Because of the limitations of the vertebrate record these new names (like their marine-derived precursors) are effectively biochronological units, despite attempts by some authors to designate lithostratotypes, and correspond in usage to the explicit regional biochronological units of North America. The two sets of names which have so far proved the most popular substitutes for classical mammal age names derived from marine stratotypes are firstly the small-mammal “zones” originated by Thaler (1966) which have so far been applied mainly in Spain and France, and sec- ondly a more or less ad hoc substitution of new names for the old to express the difference between the marine and land mammal definitions. The second system translates into the earlier vertebrate literature with little difficulty, and we use it here as a basis for organizing our discussion, with a few original modifications.

We begin the discussion of European and African biochronology with middle Miocene levels about 14 m.y. old in order to review the wide variety of alternative ideas about the relationships and dating of Late Neogene faunas in this region.

Oeningian (Heer, 1858)

This period was based on extensive collections of non-marine plants, in-

98

sects, lower vertebrates and mammals from the Oberen Susswasser-Molasse of northern Switzerland and southern Germany. According to the summary of Tobien (1971), the Oeningian assemblage has been placed in Sarmatian, Tortonian, or Messinian age by various authors. Studies by Fahlbusch and Gall (1970) indicate that the molasse-facies near Munich contains “Torto- nian” (that is, Late “Vindobonian”) mammals, but the typical Ohningen fauna (cf. Thenius, 1959) belongs to that period in time which immediately antedates the invasion of Hipparion in the Paratethyan-West Tethyan region, and is younger than Late “Vindobonian” mammal faunas of the Vien- na Basin, or the level of La Grive-St. Alban in France. Other local faunas of this period include those of Opole, Poland (Kowalski, 1967), St. Gaudens, France (Thaler, 1966), Caste1 de Barber& and local equivalents near Barce- lona, Spain (Crusafont, 1972) and those placed in the Sarmatian in German- speaking countries (Thenius, 1959). Unfortunately the name “Sarmatian” is gravely compromised by variations in usage outside the Vienna Basin where it is generally applied to brackish-marine deposits (cf. Papp, 1969) ranging in age to considerably younger than the Hipparion Datum (Gabunia and Rubenstein, 1968) despite its original definition. On the other hand the type fauna of the Oeningian is strictly non-marine, with both large and small mammals well represented, and has the added advantage of good radiometric calibration (see below). We therefore informally extend the meaning of Oeningian in this study to include most of the land mammal faunas attributed to Sarmatian by Thenius (1959), specifically those mentioned by name above.

The term Maremmian, proposed by Lorenz (1968) and used by Van Couvering and Miller (1971) for the Oeningian interval, is apparently based on a swamp biofacies of Vallesian age (see below) and should be discon- tinued.

have been dated by Lippolt e t al. (1963) so that the “Sarmatian/Vindobo- nian” boundary below the stratigraphic level of the Ohningen fossil beds is given at approximately 14 m.y. (Heilsberg, Bischofzell, Jungkernbuhl) and the uppermost part of the formation, the Jungere Nagelfluh, is dated at 12.6 m.y. (Hohenstoffeln). A basaltic horizon from the Frankfurt area, also correlated to the lowermost “Sarmatian” stage, has been dated by Horn et al. (1972) at three sites (samples 25, 26, 30) which give K-Ar ages from 13 to 14 m.y. In Czechoslovakia, the oldest (corrected) age given to the Sarmatian/ Vindobonian transition is on the Kr’alovce tuff, about 14 m.y. (Bagdasarjan et al., 1971, samples AV-29, AV-30; D. Vass, personal communication, 1972). Other Slovakian volcanics attributed to Sarmatian levels, which in this area range well above the Hipparion Datum which defines the top of the Oeningian, date from approximately 13 to 10.5 m.y. (Ruskov, Sazdice, Horsa-Brhlovce, Vinne, Vinicky, Star5 Kremnicka: see Konecn? et al., 1969 Bagdasarjan e t al., 1971) on a corrected scale. We can therefore set the age of the earliest Oeningian, which should be roughly equivalent to the base of the

Tuffs in the Oberen Susswasser-Molasse of the upper Rhine Valley (Hegau)

99

Central European Sarmatian Stage/Age, between 13.5 and 14.0 m.y. with reasonable confidence.

Beds with Oeningian mammal faunas immediately pre-dating the Hippa- rion Datum rest with structural conformity upon sequences with planktonic foraminifera in the Valles-Penedes graben near Barcelona and in central Tuni- sia. In the Bled Douarah, west of Gafsa, Tunisia, the mammal-bearing Beglia Formation overlies marine clays of the Mahmoud Formation said by J. Salaj to contain planktonic foraminifera of Zone N12 in the upper part (P. Robin- son, personal communication, 1972), although the published information (Salaj and Stranik, 1971) suggests a microfauna which could be as young as Zone N 1 0 below the “sables grkseuse” (Beglia Formation). In the thick Mio- cene exposed in the Rierussa arroyo near Gelida, Barcelona, the sequence of foraminifera1 faunas in the marine beds below the Hostalets fossil beds (Truyols and Crusafont, 1951) is reported by Dr. Salaj to end in faunas of Zone N13 (P. Robinson, personal communication, 1972), although this study is in a preliminary stage. In both places only the later part of the Oeningian may be represented, and it is possible that this level may be younger than the N13/N12 boundary, whereas the oldest Oeningian could be as old as the N11/N10 boundary (see below).

The position of the Oeningian relative to marine planktonic zonation can also be referred to recent studies in the Vienna Basin. Here, the lower Sarma- tian facies (early Oeningian) in its typical development is made up of brackish-water formations which appear to rest with a nearly isochronous contact on marine Badenian (Upper Vindobonian) strata. The base of the Sarmatian is generally assumed t o be younger than the N14/N13 boundary because of the reported presence of Globigerina nepenthes in the Upper Badenian (e.g., Cicha et al., 1972) but investigations by F.W. Steininger (personal communication, 1972) indicate that this is a mistaken identification of a member of the earlier G. druryi group. On this basis the Badenian is probably not younger than the stratotypical Langhian and early Serravallian (Zones N9-N 13).

Two vertebrate faunas are of great importance for biostratigraphic correlation of the lower Middle Miocene continental sequences and mammal ages: firstly the vertebrate fauna of Neudorf an der March, Spalte 1 und 2 (Dgvhs- kaT Nova Ves) near Bratislava, Czechoslovakia, which occurs in fissure fillings and is covered by marine sediments of late Badenian age. The evolutionary level of the micromammals, the geologic and paleogeographic relationships are the main reasons for correlating this locality (i.e., the fissure fillings of Neudorf an der March) with the Early Badenian (uppermost N8-N9/N10; Cicha et al., 1972). The base of the Badenian of the Central Paratethys is defined by the first appearance of Praeorbulina (P. glomerosa s.1.) and overlies the Carpathian Stage which is characterized in its uppermost part by Globigerinoides sicana. A second rich vertebrate fauna was described by Thenius (1952) and Steininger (1967) from marine sediments of late Bade- nian age from the locality Neudorf an der March, Sandberg. This fauna is

100

characterized by the first appearance of Protragoceras and Dryopithecus and correlates with Thaler’s (1966) Zone de Sansan, which is older than La Grive, which is in turn older than the typical Oeningian fauna. The Neudorf an der March, Sandberg level (Late Badenian) is approximately equivalent to zones NlO/Nll-N13 (fide F. Steininger, personal communication).

Cicha et al. (1972) thus equate the La Grive mammal age to the Sarma- tian, but this is surely a “chartefact” generalization based on the Sandberg =

Zone de Sansan = Upper Badenian equation in idealized form. At any rate, with a “base Oeningian” at 13.5 m.y. minimum, and later “Burdigalian” pre-Sansan mammals dating from about 16 m.y. (Van Couvering, 1972) the age of the Sandberg locality can be extrapolated to about 14.5 m.y., although this introduces a serious conflict with evidence from the Pacific Basin for a 14--15 m.y. age for the Orbulina datum, seen also in lower Badenian.

Since the G. nepenthes datum at N14/N13 is no longer restricted to a pre-Sarmatian age (i.e., older than ca. 14 m.y.) it can now be moved younger (Fig. 11) in accordance with the evidence of Dymond (1966), Turner (1970) and Page and McDougall(l970). This change would also accomodate the determinations by Ikebe et al. (1972) which place the Orbulina datum (Zone N9) at ca. 15 m.y., but this Japanese date does not agree with the calibration of the Slovak-Armenian group, who place the Orbulina datum in lower Badenian strata at ca. 18 m.y. (Konecny et al., 1969) and basal upper Bade- nian (“Boliuina-Bulimina Zone”) equivalent to Sandberg at ca. 15 m.y. (Bagdasarjan et al., 1971, samples AV-27, AV-28) taking into account our estimated correction factor. While this second date agrees fairly well with the previously extrapolated age of the Sandberg local fauna in mammalian chronology it would be rash to ignore all the uncertainties as to continental and marine calibrations and radiometric decay constants and conclude that the conflict is thereby solved. We can say, nevertheless, that the Early Oeningian (ca. 13.5 m.y.) lies well above Zone N10, the lowest possible zone associated with the Sandberg pre-Oeningian fauna, and (from Hostalets if not Bled Douarah) probably close to if not above the N13/N12 boundary in the Early Serravallian. Although it may be that the earliest Sarmatian is time-equivalent with the later part of the Zone de La Grive, the direct correlation of Sarmatian and Zone de la Grive (Cicha et al., 1972) is not in accord with mammalian evidence linking Sarmatian vertebrate faunas mainly to Oenin- gian, nor with the radiometric evidence appears to indicate an interval between ca. 15 m.y. and 13.5 m.y. of post-Sansan and pre-Oeningian time, most if not all of which is also pre-Sarmatian time.

In summary there appears to be a slight discrepancy in the calibration of the marine-continental biostratigraphies to the time-scale in the vicinity of the Orbulina Datum. The correlation of the Neudorf an der March, Sandberg fauna (which is older than the typical Oeningian fauna) within the N10-Nl2 interval, suggests an age of about 14.5 m.y. for a post-Orbulina level and about 15 m.y., or slightly older for the Orbulina Datum. On the other hand, the available data in the marine record suggests an age of about 1 4 m.y. for

107

the Orbulina Datum, and this estimate has been used here in the construc- tion of Fig. 11. The discrepancy here is probably due partly to the range of experimental error in K-Ar dating and partly to imprecision in the marine- continental correlation involved. For the moment we leave this unresolved until additional data are available (see Appendix, note 1). Suffice to state here that the dates of 15.5-15.0 m.y. for the Late Badenian (NlO-Nl3) and 17.5--18.0 m.y. for the Early Badenian (Orbulina Datum) which are used in the Paratethys region are clearly too old.

the European Oeningian have been obtained from Muruyur, ca. 13.5 m.y. (W.W. Bishop, in Coppens, 1972) and from the Fort Ternan quarry, ca. 14 m.y. (Evernden et al., 1964; Bishop et al., 1969), both in west-central Kenya. The Muruyur fauna, part of the Lake Baringo area research sponsor- ed by Bedford College (London) and Leicester University, is as yet under study, and not all of the Fort Ternan collection has been published, but Churcher (1970) and Gentry (1970) describe an abundant and varied rumi- nant fauna from the latter site which is said to be at the same evolutionary level as “latest Vindobonian or earliest Sarmatian” faunas of Europe and Asia. A. W. Gentry (personal communication, 1972) furthermore tentatively places the bovid assemblage of Beni Mellal, Morocco, as approximately equivalent in age to that of Fort Ternan, while Jaeger and Martin (1971) independently concluded that the small mammals from Beni Mellal were slightly older than the Vallesian. Considering all the uncertainties, Gentry’s paleontological correlation of the 14 m.y. old Fort Ternan fauna agrees remarkably with the German and Slovakian radiometric evidence for the age of the Early Oeningian. On the other hand, the bovids near the Chinji-Nagri boundary in the Siwaliks of India and Pakistan, which Gentry (1970) also compares with those of Fort Ternan, have been reported to be associated with Hipparion (Pilgrim, 1938) and would therefore be at least 1.5 m.y. younger on the basis of a 12.5-m.y. Hipparion Datum. Hussain (1971) points out, however, that the earliest unequivocal appearance of Hipparion is above the base of the Nagri and is thus more in accord with Gentry.

In Africa south of the Sahara, collections of the same radiometric age as

Vallesian (Crusafont, 1950)

This period of time is characterized by the first Hipparion faunas of the Mediterranean world. The base of the Vallesian “stage” was identified by Crusafont with the first appearance of Hipparion in Spanish land mammal faunas. Thus, pre-Vallesian Hipparion remains in other parts of the Old World can logically be expected to exist according to the limited sense of Crusafont’s definition, but the rapidity with which this equid colonized the temperate regions of the Northern Hemisphere has been widely remarked (e.g., Teilhard de Chardin and Stirton, 1934; Repenning, 1967; Gabunia and Rubenstein, 1968). Although Crusafont was only attempting to make a local re-definition of the early “Pontian” age, also known as “Maeotian”, the

1 0 2

usefulness of the Hipparion Datum liberated in this way from ambiguous and difficult correlations to brackish-marine molluscan faunas of the Paratethys has led other European workers to extend the Vallesian throughout the Mediterranean basin. Correlation to the Spanish faunas will undoubtedly cause some problems as the record improves, but for the present we will treat the Hipparion Datum as essentially isochronous in central and western Eurasia and North Africa.

faunal successions in this region is closely approached in several places. The most significant of these are in the Valles-Penedes basin near Barcelona (Hostalets and Can Ponsig local fauna sequences) where the Vallesian may be said to be typified, and at Howenegg in southern Germany where Hipparion- bearing strata rest on upper tuffs of the Oberen Susswasser-Molasse. The Howenegg fossil beds are directly dated by hornblende lapilli at 12.5 m.y., and the uppermost molasse tuffs at 12.6 m.y. (Lippolt et al., 1963). Thus the Howenegg fauna appears to succeed the radiometrically calibrated “type fauna” of the Oeningian superpositionally with little lapse in time; it is from this occumence that an age of 12.5 m.y. was proposed for the Hipparion Datum in western Eurasia (Van Couvering and Miller, 1971).

The Hipparion Datum is also documented in superposed successions of local faunas in northern Turkey (Ozansoy, 1965), southwestern Turkey in the Sekkoy Member of the Pisdic Formation (Sickenberg and Tobien, 1971), central Tunesia in the Beglia Formation (Robinson and Black, 1969) and possibly in Portugal in the Lissabon localities (Thenius, 1959). According to modern workers (Forstgn, 1968, 1972; M.T. Alberdi. written communication, 1972), the earliest North African Hipparion, H. africanum Arambourg, is synonymous with the first-appearing species in Europe, H. primigenium (v. Meyer) ( see Appendix, note 6).

Furthermore, Vallesian-like collections of mammals from Morocco (Oued Zra), Algeria (Oued el Hammam) and the upper molasse of northern Tunisia (Kechabta Formation) include specimens of primitive Hipparion together with Progonornys, the first murine rodent immigrant into the Paleo-Mediter- ranean Basin (Jaeger and Martin, 1971; J.J. Jaeger, written communication, 1972; P. Robinson, personal communication, 1973) as do most Vallesian faunas of Spain and France (H. de Bruijn, written communication, 1973). Progonomys has also been found in the Vallesian local fauna at Kastellios Hill, Crete (De Bruijn et al., 1971), so it seems likely that the northern and southern shores of the basin shared a similar climate and mammal fauna during this period. Marine microplanktonic faunas of the central and eastern Paleo-Mediterranean (Israel, Crete, Libya, Italy) were essentially continuous with those of the Caribbean faunal realm during the Vallesian Age (that is, later Serravallian-early Tortonian) (Cita and Blow, 1969; Berggren and Phillips, 1971) so that normal interchange of land mammals would necessarily have been by way of the Levant rather than across a Betic (Gibraltar), Sicilian or Sardinian-Corsican land bridge. However, tectonism was active

The “true”, directly observed, first appearance of Hipparion in vertebrate

103

along the continental margin of the Maghreb during the Middle and Late Miocene (e.g., Glaqon and Rouvier, 1972), and intrusive and volcanic rocks associated with the tectonism have been dated from ca. 10 m.y. to ca. 7 m.y. (Choubert et al., 1968; Bagdasarjan et al., 1973). It is possible that Vallesian migrants may have been afforded short-lived trans-Mediterranean passage- ways during the tectonic activity.

South of the Sahara, faunas of the same age as the Early and Middle Vallesian (if not Late Vallesian) do not appear to include Hipparion. Many new localities found by M. Pickford in the Ngorora Formation near Lake Baringo, Kenya, during the 1972 summer season represent a long sequence of local faunas of which only the youngest contain Hipparion (W.W. Sishop, personal communication, 1972). According to Bishop and Chapman (1970) the upper fossiliferous sediments of the Ngorora Formation should not be any older than about 11 m.y. From this it could be argued that latitudinal zoogeographic zones (i.e. desertic) had delayed the spread of Hipparion south of the Mediterranean littoral, as they may also have prevented the advance of the pre-Vallesian equid Anchitherium in the earliest Miocene *. Other appearances of equids in the sub-Saharan fossil record are numerous but are attributable with certainty only to levels younger than the Ngorora beds (C.S. Churcher, written communication, 1973).

The oldest Hipparion remains directly associated with marine beds seem clearly to be of pre-Tortonian age. We have already pointed out that no planktonic microfaunas younger than Zone N13 have been identified in marine strata underlying beds where the Oeningian/Vallesian transition (Hipparion Datum) is observed. GuGrin et al. (1972) have recently described the occurrence of two Hipparion teeth of “primitive” aspect which were included in gravels below marine beds said to contain a planktonic microfauna like that of the Tortonian stratotype. More elaborate evidence, however, is afforded by the work of De Bruijn et al. (1971) at Kastellios, Crete, where beds with large and small mammals of probable late Vallesian age are intercalated with beds containing planktonic microfaunas which these authors place in lowest Zone N16 (basal Tortonian). Other interpretations of the published marine microfauna might make it as old as the N14/N15 boundary, and M. Freudenthal (written communicaticn, 1972) considers that the Kastellios small mammals might be younger than Late Vallesian, so the correlation of Late Vallesian to early Zone N16 is probably on the con- servative side. The Hipparion Datum itself is therefore probably equivalent to Serravallian marine levels, and may be close to the G. nepenthes Datum (see above).

An even more extreme opinion on the age of the Vallesian relative to Miocene marine beds is presented in a preliminary report by Freudenthal (1971) on a fossil mammalian fauna from the Gargano Peninsula, prov. Foggia (Italy). The mammals represent an island population, preserved in fissure deposits, derived in this author’s opinion by isolation of a Late Valle- sian or slightly younger fauna. As sea level rose, the fissures were progressive-


104

ly covered by marine calcarenites which are labelled “probably Serravallian” on a recent geological map of Italy. Freudenthal(l971, p. 6) interprets this relationship to mean that Vallesian is older than Serravallian, but in our view this is only a speculation until the local marine microfauna has been re- examined.

At Plakia, Crete, a small, rodent-dominated local fauna estimated to be “slightly older than the first arrival of Hipparion in Europe’’ (that is, Oenin- @an) is described by De Bmijn and Meulenkamp (1972). The fauna is collected from the base of a Neogene succession in which higher levels contain local foraminifera1 assemblages correlative to the lower or middle part of the Tortonian stratotype. An unusual glirid species in the mammal fauna, found elsewhere only in pre-Vallesian sites in Spain, suggests that the absence of Hipparion here is not accidental, but the geological relationships do not preclude a late Serravallian age for the lower Plakia section.

The vertebrate faunas of Baccinello, prov. Grosseto (Italy) consist of a lower level (faunas V1 and V2) associated with brackish-marine molluscs, succeeded upwards by normal-marine beds in which a sandstone containing a Hipparion fauna (V3) is intercalated (Lorenz, 1968). The lower vertebrate assemblage is found in lignite seams and associated paludal sediments, and includes the famous remains of Oreopithecus bambolii Gervais. Because the molluscs of this general level were attributed to “Helvetian” age, and because Hipparion was determined to be confined to the higher (V3) level, Hurzeler (1958) and Lorenz (1968) considered the Vl-V2 fauna to be of pre- “Pontian” (pre-Vallesian) age, representing an interval which was proposed to be named Maremmian. Advances in the knowledge of the small-mammal biochronology of southern Europe have since shown that the murid found in the Vl-V2 assemblage, Parapodemus cf. vireti Schaub, is descended from Progonomys of early Vallesian age (cf. Michaux, 1969); the absence of Hipparion from the swamp facies here is probably ecological and the Oreo- pithecus fauna of Baccinello is most likely late Vallesian (contra Van Couver- ing and Miller, 1971). Nothing further need be said of the brackish-marine molluscan assemblage, except that it represents another example, if one be needed, of the limitations of such facies-bound assemblages for biochronological correlations.

Turolian (Crusafont, 1965)

This period is based on collections from the Calatayud basins of north- central Spain. It replaces the earlier term “Pikermian”, which Crusafont (1950) had proposed as a replacement of the “Pontian sensu stricto”, which had a variable meaning among vertebrate paleontologists who wished to acknowledge that some early Hipparion faunas were “pre-Pontian,” or “pre- Pliocene”, in age. The Spanish Turolian faunas clearly evolve directly from those of the Spanish Vallesian but in addition show very striking differences: e.g., a much more diversified cricetid and murid rodent fauna; a sharply

105

accelerated trend towards hypsodonty among such herbivores as equids, rhinoceroses, and iodents; and a marked increase of open-country types such as antelopes, hyaenids, and macaques relative to streamside and forest-adapted forms such as beavers, otters, tapirs, and pongids. This sort of chauge is paralleled in faunas of similar evolutionary grade all over Eurasia (cf. The- nius, 1959). The change was least noticeable in North European faunas, whereas typically endemic African taxa (ostrich, giraffids, hyraxes, aardvark) first advanced at about this time into southeastern Europe and the Asian steppes, suggesting that the Turolian began with the onset of a climate change of continental magnitude in which ecofacial boundaries shifted progressively northward (Tobien, 1970b). The effects of the post-Vallesian climate change on the fossil record is therefore probably recognizable as a resonal biochronological datum (albeit a relatively slow rather than sudden event). In the temperate areas it seems to have been expressed as a change towards continental ecologies which were better adapted to drier summers and higher fire hazard (i.e. grasslands) if not to higher mean annual temperature. In fact the paleobotanical and paleooceanographical evidence appears to point to an overall cooling trend during this period (cf. Nagy, 1970; see also discussion of paleoclimatology in Chapter 7). Because this transition is very well recorded in the Spanish Neogene, and continues the regionally correlated Vallesian sequence, we prefer to extend the Turolian as a regional land mammal age in place of the “Pontian sensu stricto” but with essentially the meaning that this term has had to vertebrate paleontologists.

Los Mansuetos (Teruel) has been designated the “stratotype” of the Turolian “stage” (De Bruijn and Mein, 1968; Freudenthal, 1968), but despite the thickness of the section only a short interval of evolution in Euro- pean mammal faunas is represented in the Los Mansuetos levels, and this should preferably be understood as the Turolian type locality. The characteristics which are said to define the Turolian mammal age such as the presence of Ruscinomys s.1. (Thaler, 1966; but see Mein and Freudenthal, 1971) or the absence of Cricetodontidae (De Bruijn and Meulenkamp, 1972) need not apply at the Turolian/Vallesian transition. A number of southwestern Euro- pean localities yield mammalian assemblages which are markedly more progressive and grasslands-oriented than those of typical Vallesian sites, and which are also earlier than Los Mansuetos (see Fig. 11). For this reason our boundary is indicated at the approximate evolutionary stage of Piera, a site stratigraphically overlying Hostalets de Pierola and distinctly more advanced in its paleontology (P. Robinson, personal communication, 1973), in order to link the Valles-Penedes sequence to that of Teruel. Los Mansuetos thus characterizes the fully developed Turolian age of mammals and not its beginning, and sites such as Montredon and Masia del Barbo are Early Turolian (De Bruijn et al., 1971; Benda and Meulenkamp, 1972). Unfortunately the taxonomy of Hipparion of this general age is very complicated and controversial, but most active workers agree in distinguishing two (or more) species in nearly every Turolian fauna in Eurasia, one being much more gracile than

106

the other (Forste‘n, 1968; Sondaar, 1971; Hussain, 1971). Correlation of Turolian age faunas therefore depends mainly on estimates of the evolutionary grade of collected assemblages and on paleoclimatological arguments, and the exact relationship of local faunas outside the geographical range of the small-mammal species of the designated Turolian (= Zone de Terual of France) is a matter for much future work.

On the island of Samos, a well-known grasslands-type Hipparion fauna is correlated to the Pikermi and Salonika local faunas on the Greek mainland and with the Turolian of Spain (Thenius, 1959; Freudenthal, 1970), although the Samos fauna is more strongly influenced by Asian and African elements than those further west. At a higher level in the Samos sequence (Quarry 5) an even more “steppic” fauna is found which closely resembles that of Maragha in Iran (Sondaar, 1971; Gentry, 1971). Tobien (1970b) argues that the resemblance to Maragha is one of facies rather than synchroneity, and i t is certainly true that there is an overall decrease in “steppic” elements a t all levels going northwestwards. The main Samos fossil bed (Quarries 1-4) is directly dated at ca. 8.5 m.y. as part of a stratigraphic sequence of dated tuffs (Van Couvering and Miller, 1971). The Melka el Ouidane ( Camp-Berteaux) Hipparion fauna of Morocco has also been dated by Choubert et al. (1968); according t o its age of 7.4 m.y. it should lie in the Late Turolian but it is as yet a small fauna which has not been well described (cf. Thenius, 1959; J.J. Jaeger, written communication, 1972). Relying mainly on the Samos dates t o give an upper limit to the age at which the continental climate began to change as described above, we estimate the beginning of the Turolian age at some time between 10 and 9.5 m.y.

The consensus of very limited evidence linking Turolian mammals with the Late Neogene marine sequence is that the Turolian is roughly equivalent to the stratotypical Tortonian in time. Firstly, the upper Baccinello mammal level (V3 fauna) includes a gracile Hipparion and other taxa (including a macaque-like form) associated with the murid Anthracomys majori Schaub, which is not known t o occur earlier than late Turolian levels in Spain (Michaux, 1969). The foraminifera1 mark associated with the V3 fauna yield planktonic species which are best (if not certainly) ascribed to Zone N16 (Lorenz, 1968; Van Couvering and Miller, 1971). A similar association has recently been described at Aspe in Alicante, southeast Spain, where a medial Turolian mammal fauna (in the sense of our definition) has been found in sands interdigitated with marine strata with an upper Zone N16 microfauna, equivalent t o mid or late Tortonian age (Montenat and Crusafont, 1970; Montenat and Martinez, 1970). Again, in Macedonia, Gramann and Kockel (1969) report a Pikermi-like Hipparion fauna in conglomerates interfingering with the base of a marine section in the Strimon Basin. The foraminifera in this sequence comprise a depauperate, brackish-influenced assemblage in which only the benthonic form Uvigerina bononiensis compressa appears to have diagnostic value, as Meulenkamp (1969) has identified it in the mid- Tortonian of the type section (?mid Zone N16). Radiometric evidence from

107

other parts of the world (e.g. Ikebe et al., 1972; Page and McDougall, 1970, 1974) suggests that the foraminifera1 faunas of earliest Tortonian age may be as young as 10-9 m.y., but the evidence from Kastellios (see above) indicates that the Vallesian overlaps this part of the Tortonian marine sequence. Other evidence, cited elsewhere in this study, also indicates that base Tortonian is ca. 11 m.y. old and thus predates our estimated age for earliest Turolian.

Ruscinian (Kretzoi, 1962)

This period was proposed at the Lyon 1971 meeting of the Committee on Mediterranean Neogene Stratigraphy (V‘ Congrks du Ndogkne Me‘diterra- ne‘en) as the name for the period in European mammalian chronology between the Turolian and the Villafranchian (Coppens, 1972) and is so used in this paper (Figs. 11 and 12). The name is a latinization of Rousillon, a well- known local fauna of southern France (Thenius, 1959) which is equivalent in age to Montpellier and others placed in the Zone de Perpignan by Thaler (1966). Ruscinian was intended by Kretzoi to identify assemblages from the period just older than that represented by the Cshnota 1.f. in Hungary, which is itself apparently just pre-Villafranchian in age (fide Tobien, 1970b). In the C.M.N.S. usage, which we adopt, the Ruscinian would therefore include the level of Cshnbta, so that the “CsArn6tian” age proposed by Kretzoi (1962) is submerged.

Rodents living during the Ruscinian age continued the trend towards diversification and hypsodonty exhibited by Turolian line ages to the extent that in southwestern Europe many newly evolved genera (e.g., Apodernus, Occitanomys, Ruscinomys s.s., Castiltomys, and others) have been distinguished in collections just younger in evolutionary grade than the Arquillo local fauna of latest Turolian age (Michaux, 1969; Mein and Michaux, 1970; Gu&in and Mein, 1971). Some characteristic large mammals of the post- Turolian local faunas are Hipparion crassum Gervais, A nancus arvernensis Croizet e t Joubert, Dicerorhinus megarhinus Cristol and Tapirus arvernensis Croiz. et Joub. The quasi-proboscidean Deinotheriurn becomes extinct north of the Sahara during the Ruscinian and microtine rodents make their first appearance, among other changes which may have been brought about by a continuing deterioration in Holarctic climates.

No direct radiometric dating of Ruscinian age fossil beds is available in Europe. Correlation t o the dated East African local faunas is difficult because of the increased differences between tropical and temperate faunas due to climate deterioration, and with those of North America because of geographical distance. However, i t is now widely accepted that the Paleo-Mediter- ranean Sea Basin was isolated from the world ocean during the Messinian Age (Ruggieri, 1967; Hsii et al., 1973), and igneous rocks associated with Messinian-like evaporite deposits and deep pre-Pliocene erosion have been dated from 8.0 to 6.0 m.y. a t Mellila, Elba, and Ischia (Tongiorgi and Ton- giorgi, 1964; Choubert e t al., 1968; Civetta et al., 1970). Research in south-

108

ern Spain and in the Rh6ne Valley appears to identify early Ruscinian land mammals with the Messinian regressive event:

(1) In the upper Guadalquivir Basin and further east in Murcia and Ali- cante, strata with normal marine planktonic microfaunas of post-Tortonian and pre-Zanclian age indicate connections to the open Atlantic during the Messinian. Studies by C. Montenat in Alicante suggest that beds of this age correlate with mammal-bearing sediments at La Alberca which yield an early Ruscinian rodent assemblage (Montenat et Crusafont, 1970; emended by M. Crusafont, personal communication, 1972).

(2) Early? Ruscinian local faunas in southern Spain at Alcolea de Cala- trave (Molina et al., 1972) and Almoradier (De Villalta Comella and Crusa- font, 1957), but not at Alcoy of earliest Ruscinian age (Thaler et al., 1965; Gukrin and Mein, 1971) contain North African elements such as the rodent Paraethomys and the equid H. rocinantis Vill. et Crus. (L. Thaler, personal communication, 1972; E. Aguirre, personal communication, 1972). I t is entirely possible that these and other mammals (e.g., North African rhinoceros according to personal communication from E. Aguirre, and gerbils according to personal communication from H. de Bruijn) may have entered southern Spain across a dry land bridge when the Betic seaway to the Paleo-Mediterranean was blocked.

(3) The lowest strata of the post-Messinian marine infilling at the mouth of the Rhane paleo-canyon contain earliest Pliocene microplankton with Sphaeroidinellopsis; according to Ballesio (1971) these beds should be correlated with fossiliferous continental deposits at Hauterives, 100 km or more upstremi, which yield a fauna close to the Turolian-Ruscinian transition (newly proposed Zone de Hauterives of Gugrin and Mein, 1971, said to be just pre-Alcoy in age). Allowing for a certain amount of justifiable leeway in each of the three cited continental-marine correlations, we would conclude that the early Ruscinian faunas are probably a little older than the oldest Pliocene marine faunas of the Mediterranean Basin, and estimate an age for the beginning of the Ruscinian of ca. 5.5-6 m.y. The Hauterives level, with Apodemus primaeuus the oldest known field mouse, should be included among the earliest Ruscinian faunas; Alcolea and La Alberca, as well as Alcoy, may be somewhat older.

According to L. Thaler (personal communication, 1972), the Ruscinian should begin with the immigration of microtine (arvicoline) rodents such as Mimomys into central Europe. Nevertheless, the first microtines did not reach the south of France until a little before the level of Sbte and Nlmes local faunas (Hautimagne sub-zone of Zone de Perpignan) in what we would call late Ruscinian time, and after the level of Montpellier and of Rousillon itself (Thaler, 1966; Michaux, 1971a). I t might be expected that a group which today contains so many cold-adapted forms might have exhibited an original preference for northern latitudes, and in fact the earliest known microtine in Europe may be the Promimomys from Podsilece, an ?Early or mid-Ruscinian site in Poland (Kowalski, 1963; Michaux, 1971a); moreover

109

the abundant Late Ruscinian rodent fauna at Weze in Poland consists exclusively of microtines and microtine-like cricetids (Kowalski, 1960) unlike the more diverse Mediterranean faunas of similar age. On the other hand, the earliest known microtines in Spain are at younger levels than Skte (Layna, Villaroya) and are species which are more evolved than the ones which first invaded Central Europe (De Bruijn et al., 1971). It seems reasonable to connect this apparently slow, southward prochoresis with progressive climate deterioration during the Ruscinian. Evidence is cited elsewhere for worldwide temperature decline and perhaps early montane and polar glacial advances ca. 4-3 m.y. ago (see chronology of climate events).

Villa franchian (Pareto, 1865)

This was proposed as a stage corresponding to a sequence of continental sediments near Villafranca d’Asti in the Po Valley, and must be the oldest land-mammal-defined subdivision of the Neogene still in common use. It has been given wide (and unfortunate) lithostratigraphical applications in southern Europe and also in the Maghreb such that most conglomeratic formations associated with latest Neogene (“Alpine”) tectonism bear this name, but is has also been much expanded geographically and chronologically to be used as a biochronological subdivsion. In the following paragraphs we will confine ourselves mainly to the problem of identifying and correlating the beginning of the Villafranchian land mammal age, based on excellent summaries to be found in the works of Kurtdn (1968), West (1968), Azzaroli (1970), Tobien (1970a) and Cooke (1972).

It has been customary to identify the “typical” Villafranchian with the assemblage Mummuthus (“Elephus”), Leptobos, and Equus (E-L-E Group of Tobien, 1970a). Although it is true that Pareto mentions elephantids as well as mastodontids in his original description of the Villafranchian, Savage and Curtis (1970) and Azzaroli (1970) find no firm evidence that any of the E-L-E Group had ever been collected from the Villafranca d’Asti exposures. The earliest known elephantids in Europe, a primitive form of Mummuthus meridionalis which V. Maglio (written communication, 1972) considers nearly if not actually conspecific with M. africanuuus of the Maghreb, occurs in the Lower Valdarno Beds (Montopoli 1.f.) of Tuscany of supposed Early Villafranchian age (Azzaroli, 1970) and in Praetiglian strata of Holland (as “M. subplunifrons”) which most present authors consider to be somewhat younger than earliest Villafranchian (e.g., Azzaroli, 1970; Tobien, 1970a, Cooke, 1972). The presence of Equus in earliest Villafranchian levels is even more doubtful (Azzaroli, 1970), although Tobien (1970a) notes that Ruma- nian workers have identified two species of Equus in the “Early Villafran- chian” sites Beresti-Malusteni in which both Leptobos or elephantids are lacking. The large mammals of the Late Ruscinian and earliest Villafranchian appear to make up very similar assemblages in any case, in which only a few elements show evolutionary replacement (e.g., Parubos-Leptobos) (Kahlke, 1968).

110

The small-mammal biochronology during the Ruscinian-Villafranchian transition is potentially more precise, and recently the work of Michaux (1971 b) has demonstrated the close affinities between the level of Etou- aires-Perrier and the Arondelli small-mammal fauna at Villafranca d’ Asti (Berzi et al., 1970). Mirnornys species in these collections exhibit the advanced dental characteristics of Fejfar’s (1964) Group 11, and are thus distinctly younger than the forms of the S&te level (Late Ruscinian). Local faunas in the latest part of the Zone de Perpignan, e.g. Seynes, Layna, Villa- roya and others in the Midi and Spain, are tentatively correlated to Etouaires and thus to the Early Villafranchian by Michaux (1971a) because endemic Mirnornys species of this region also have Group I1 dentition, although the other rodents at Layna make up a characteristically Ruscinian assemblage (cf. Crusafont and Aguirre, 1971). Interestingly, Leptobos is also tentatively identified from Layna. Other transitional Ruscinian-Villafranchian local faunas with Group I1 microtines are Hajnhcka, without any of the E-L-E Group, and Rebielice (with Leptobos only) (Fejfar, 1964; Michaux, 1971a). It is unlikely that Group I1 microtines, radiating from the M . stehlini-M. gracilis immigrant stock, could have appeared simultaneously all over Europe but their development appears to have been sufficiently rapid, just prior to the type Villafranchian level, to be useful as a guide to the beginning of the Villafranchian.

Radiometric dating of the Villafranchian Age is based mainly on the work of Bout (Bout et al., 1966; Bout, 1969) and Savage and Curtis (1970). These workers agree reasonably well on determinations at Etouaires-Perrier, which indicate that the earliest Villafranchian levels are between 3.5 and 3.0 m.y. old. A date of 4.2-4.3 m.y. at Cava Simoni near Rome (Ambrosetti et al., 1968; Ambrosetti et al., 1972) has been correlated to the lacustrine basin of Poggio Mirteto, and is attributed to “Early Villafranchian” age apparently by default since the mammalian fauna found in the lake beds is very limited (Anancus aruernensis, Tapirus aruernensis) and could as easily be Ruscinian. The Vialette local fauna, dated at 3.8 m.y. by Savage and Curtis (1970) also does not contain any taxa restricted to post-Ruscinian levels (Heintz, 1969) and is here provisionally assigned to late Ruscinian age, on the grounds that the base of the Villafranchian Age is defined by the evolutionary level seen in the Villafranca d’Asti local fauna (Arondelli, Fornace RDB) and that this level is evidently the same as that of Etouaires-Perrier. This is an arbitrary decision, made in lieu of any specific argument to the contrary, which brings the base of the Villafranchian as close as legally possible to the first appearance of the E-L-E group in Europe.

The Villafranchian Age is extended in common usage to include a series of local faunal levels up to at least that of Le Coupet, dated ca. 1.9 m.y. (Sav- age and Curtis, 1970) and generally somewhat younger to levels dated 1.3 to 1.1 m.y. (Bout, 1969; Azzaroli, 1970; Ambrosetti et al., 1972). Subdivisions of Villafranchian time are much discussed, especially in relation to glacial events in Europe, but it is already clear that paleontologically constrained

111

and radiometrically calibrated studies of the geomagnetic polarity signature in rocks of this age will be the key to a consensus which is yet far from attained. The chronology of Late Neogene climate cycles is discussed in Chapter 8, and it must suffice here to point out that although the effects of colder climates are found in the appearance of glacially influenced sedimentation in the northern continents and oceans about 3 m.y. ago, the generally accepted date for the beginning of the Pleistocene is 1.8 m.y. ago and that only the post-Villafranchian land mammal faunas of Europe were associated with subpolar lowland ice-sheets. Accelerated average deposition of ice-rafted sediments is estimated to begin about 1.3 m.y. ago, and the age of the Nebraskan continental glaciation in North America is estimated to center at ca. 1.5 m.y. ago, so that it is probable that a sedimentological if not paleontological effect of lowered temperature can be made out in mid-Villa- franchian time at this approximate level, but so far no convincing link has been demonstrated between Villafranchian mammals and the early montane glacial advances in Europe.

Correlation of a detailed mammalian sequence of Villafranchian age to a well-calibrated marine microfaunal level remains also to be done. It has long been recognized that certain Early Villafranchian levels in Italy (Villafranca d’Asti, lower Valdarno) were in lateral or superpositional contact with yellow sands of the Astian Pliocene facies (Selli, 1967; Tobien, 1970b) and a relationship between the Late Villafranchian and the Calabrian series has been proposed (Azzaroli and Ambrosetti, 1970). In K-Ar chronology the base of the Villafranchian, at a minimum 3.0 m.y. old, is considerably older than the base of the Calabrian correlated to the base of the Olduvai Event at ca. 1.8 m.y. The assumption by Lona et al. (1974) that there is a simultaneous (Donau?) cold phase characterized by the extinction of Taxodium in pollen profiles of the Po Valley, at the base of the Villafranchian, and at the base of the Calabrian, is demonstrably wrong. Tobien (1970a) and Azzaroli (1970) have noted that the molluscan fauna of the Red Crag of England, which (like the Astian molluscan fauna) was originally put in the Pliocene by Lye11 in defining the epoch, is associated with a mammalian fauna which seems to be referable to the Middle Villafranchian. Recently, Beck et al. (1972) found that the Red Crag underlies the Ludhamian silts at the base of the “Cromer Forest Bed Series”, higher parts of which (Antian, Baventian, Pastonian stages: stages of West, 1968, p. 247) contain Late Villafranchian mammals. The Ludhamian is said to equal the Tiglian beds of Holland, whereas the lower part of the Tiglian series and the upper part of the Red Crag equivalent have normal paleomagnetic sign (see also Van Montfrans, 1971) which the authors tentatively linked to the Olduvai Event (1.8-1.6 m.y.) without any calibration support; based on other mid-Villa- franchian dates this seems to be a reasonable estimate to use (Savage and Curtis, 1970; Ambrosetti et al., 1972).

112

Pliocene mammals of Africa

According to the current calibration of Mediterranean Basin biostratigraphy the Pliocene covers a time from approximately 5 m.y. to 1.8 m.y. ago, and thus includes all but earliest Ruscinian and latest Villafrahchian land mammal ages. In East Africa the search for hominid fossils has stimulated intensive biostratigraphic and radiometric investigations in deposits of this general age, notably in the Lake Rudolf and Lake Baringo basins. A consensus on a detailed radiometrically calibrated sequence of large-mammal assemblages is emerging (Cooke and Maglio, 1972; Coppens, 1972; Maglio, 1972; H.B.S. Cooke, written communication, 1973) which allows close comparison with the very dissimilar European Pliocene fauna and also offers some insight into the correlation of the North African “Villafranchian”.

The Lothagam faunal level, ca. 6 m.y., is close to the age of the Turolian- Ruscinian transition. According to Cooke and Maglio (1972) the proboscidean fossils from this level include the earliest true elephantids, which had evolved in Africa from a gomphotheriid ancestor. (Even earlier elephantids may be known from Mpesida, ca. 7 m.y., according t o Coppens, 1972.) The first of the elephantids to reach Europe, however, was Mammuthus africanauus (= primitive M. meridionalis of Montopoli according to V. Maglio, written communication, 1972). In the East African sequence the africanauus evolutionary stage is limited to the upper earlier Shungura Fauna, which has an age span from ca. 3.5 to 2.5 m.y. (H.B.S. Cooke, written communication, 1973). Equus is first recorded from the later Shungura Fauna, ca. 2.5 m.y. ago, just above the last record of M. africanauus there. On the other hand, the mastodont genus Anancus in East Africa appears to be restricted to the Kanapoi Fauna interval, ca. 4.5 to 3.8 m.y. ago, where it is associated with a primitive form of Mammuthus, M. subplanifrons.

As there is little question that in Europe, Anancus (A. aruernensis) persists into the Villafranchian, where it is associated with the relatively advanced M . meridionalis at mid-Villafranchian sites such as Le Coupet at 1.8 m.y. (Bout, 1970), the problem is whether the association in North Africa of Anancus osiris and Mammuthus africanauus parallels the situation in Southern Europe (in which case it should indicate an Early Villafranchian age as indicated by Arambourg, 1970, and Coppens, 1971) or whether it is a feature of an older, more East African-like fauna. In support of the first alternative Arambourg (1970) reported that species of Equus had been found with Anancus and Mammuthus at Ain Boucherit (Algeria), Lake Rudolf Basin, Omo (South Ethiopia), Kaiso (Uganda), Kanam (Kenya) and Laetolil (Tanzania) in a characteristically Early Villafranchian assemblage. Coppens (1971) also regarded the occurrence of a progressive form of the suid Nyanzachoerus together with these two proboscideans in the Lower Hamada Damous beds (Tunisia) as confirmation of an Early Villafranchian age because remains of Nyanzachoerus are dated as young as 2.5 m.y. in East Africa. Nevertheless, the first Equus seems to be younger than the last Anancus in East Africa.

113

Arambourg greatly oversimplified the biostratigraphy at Laetolil, Kaiso, and Omo (cf. Maglio, 1969; Arambourg, 1970; Cooke and Coryndon, 1970), and the deposits at Kanam are geologically complex with a poor record of documentation for fossils found there. Most of the Kanam fauna appears to be well correlated to the ca. 4-m.y. Kanapoi Fauna level (cf. Saggerson, 1952; Maglio, 1970b) so that the record of Equus oldowayensis from this site presents a real anomaly. Cooke and Maglio (1972) also indicate that Nyanza- choerus hanamensis, the East African species to which Coppens (1971) compared the Lower Hamada Damous species N. jaegeri, ranges back to approximately 4 m.y.; furthermore Equus is known only from the Upper Hamada Damous beds. Controversy over the taxonomy of fossil proboscideans undoubtedly contributes to this problem but it may be that the form ascribed to M. africanauus in East Africa is more advanced than the originally defined form associated with Anancus in North Africa (cf. Coppens, 1971, footnote p. 54).

The only well-studied small-mammal fauna of this age in all of Africa is found at Lac Ichkeul (Tunisia), together with a MammuthusAnancus large-mammal assemblage (without Equus) (Coppens, 1971). Jaeger (1971) concluded that these small mammals compared directly with the assemblage from the Maritsa fissure-fillings on Rhodes, which De Bruijn et al. (1970) in turn had previously correlated to the level of Rousillon, Podlesice and other mid-Ruscinian sites.

A t Ain Brimba (Tunisia) is a “Middle Villafranchian” site, younger than Lac Ichkeul or Lower Hamada Damous (Coppens, 1971) with an interesting resemblance to the upper assemblages of the ca. 3.5-m.y. earlier Shungura Fauna of East Africa in that it contains M. africanauus but neither Anancus nor Equus. Whether it actually correlates to this level in East Africa, and thus to a radiometric date close to the age of the base Villafranchian in Europe, is another matter, but it agrees with the other evidence that the North African “Moghrebian” MammuthusAnancus fauna (Arambourg, 1970), like the Kanapoi Fauna of East Africa, is not Early Villafranchian but rather Middle to Late Ruscinian in age.

Marine Pliocene biostratigraphy in North Africa depends mainly on speculations as to the significance of marine terrace elevations and highly endemic molluscan faunas (Biberson, 1970). Arambourg (1970) refers to the stratigraphy at Fouarat (Morocco) where elements of the Mammuthus-Anancus mammal fauna are associated with littoral marine containing “Astian” bi- valves; near Oran, a less time-restricted Pliocene mammal assemblage (Stylo- hipparion, Libytherium i.al.) is said to occur with mollusca of the “Calabrian marine cycle. ” Conclusions based on such evidence must necessarily be relatively general, but it is safe to say that the Pliocene land mammal fauna as well as the shallow-water marine fauna of Africa was relatively very different from that of Europe (cf. Maglio, 1972, and Coppens, 1971, vs Kurt& 1968). Faunal interchanges between the two areas would thus seem to have been few and indirect, probably as a consequence of a wide-open Betic strait together with a strongly differentiated climate zonation.

114

Biharian (Kretzoi, 1941)

This period is founded on an extensively documented sequence of abundant small-mammal remains collected from karst-fillings in the central Car- pathian region, and has been correlated to numerous local faunas elsewhere in Europe including many large-mammal local faunas for which other age- names have been proposed (e.g., Cromer Forest Beds, Mosbach, Tiraspol). We have adopted Kretzoi’s definition in preference to the others because of the usefulness of the small-mammal chronology in the short Pleistocene time-intervals and because it comes close in concept and development to the Regional Land Mammal Age ideal (Tedford, 1970).

The subdivisions of the Biharian, seven in all, begin with the distinctive Betfia-phase associated with the waning of a pronounced cold-climate interval; the latest Villafranchian fossils appear to have survived until a cold- climate period (Bout, 1970; Ambrosetti et al., 1972) which may be the same one. A second cold-climate minimum with only moderate effects on the fauna marks the middle of the Biharian (Kretzoi, 1965) and it ends with the onset of the first major continental glaciation of Europe. Three (or four) major adglacial periods have long been identified in the Pleistocene stratigraphy of the Alpine region by the terms Gunz, Mindel, and Riss-Wurm. More recently two (or more) pre-Gunz montane glacial advances have been recognized under the names Biber and Donau (cf. Cooke, 1972, for references). The current consensus among Dutch and German workers on the correlation of palyno-stratigraphy in the Alpine river floodplains to the paleoclimatological sequence in Europe (Van Eysinga, 1972) is referred to in Fig. 11. According to this correlation the “Cromerian” interval of The Netherlands follows the Menapian cold phase, which is equivalent to the Gunz glaciation, and ends with the Elsterian cold phase, equivalent to the Mindel. As the type Cromer vertebrates of East Anglia seem to be mainly Biharian in age (see below) and the Mindel is generally agreed to be coeval with the earliest continental till-sheets in Europe, the Biharian age should correspond to the Gunz-Mindel interglacial period. It may, in fact, do just that but the presence of numerous climatic cycles rather than a few broad simple oscillations is clearly indicated by the pollen spectra of pre-continental glaciation strata (West, 1968; Zagwijn et al., 1971; Lona et al., 1974) just as in deep-sea cores of the same age, and this makes the direct relationship of the beginning (or end) of the Biharian to the mid-Pleistocene cold-climate maxima correspondingly more difficult to confirm.

Nevertheless, the Biharian age is relatively well defined and its calibration can be based on a number of K-Ar and paleomagnetic measurements. Re- markably few Villafranchian species persist into the Biharian age; Azzaroli and Ambrosetti (1970, p. 109) call this transition “the greatest revolution in the continental faunas of the whole Pleistocene period”. The transition is dramatic because the very youngest Villafranchian levels still contain most of the large mammals typical of the age, such as M. rneridionalis, Dicerorhinus

115

etruscus, Leptobos, Equus stenonis, Hyaena perrieri, and Hyaena breuiros- tris. The rapid change from faunas dominated by these elements to those of the Biharian type is recorded in the famous Cromer Forest Beds succession of East Anglia - although most of the collected specimens have poor stratigraphic documentation (West, 1968) -- the Mosbach sands of Germany (Kahlke, 1961), and calibrated sequences of local faunas in the Massif Cen- tral of France (Bout, 1970) and the Roman volcanic province of central western Italy (Ambrosetti et al., 1972). According t o West (1968) the younger levels of the Cromer Forest Beds contain mostly (if not exclusively) post-Villafranchian assemblages, which can be separated into a cold-climate ecofacies characterized by mammoths (M. trogontherii, M . primigenius), giant elk (Megaceros), and bison (Bison cf. priscus) and a forest-adapted warmer-climate ecofacies characterized by Elephas antiquus, Rhinoceros kirchbergensis (syn. Dicerorhinus merckii) and Hippopotamus amphibius *. In younger Pleistocene strata these forms are associated with climate-definitive pollen floras as well as other mammals not present in the Cromerian which confirm the indicated habitat preferences.

are distinctively characterized by a microtine rodent assemblage dominated by advanced species of Mimomys with Type I1 and I11 dental characters (Fejfar, 1964) ; the time-restricted genus Allophaiomys; and Pitymys and Lugurus, the first of the extant microtines to appear in the geological record. In the early warm phase of the Biharian that follows the Betfia-phase appear more of the extant microtine genera, e.g. Microtus and Lemmus. In the later Biharian, after an early-medial cold-climate phase, certain Mimomys species give rise to forms included in the modern genus Arvicola (Fejfar, 1964; Kretzoi, 1965). According to West (1968) a few specimens of Mimomys are known from the Hoxnian (Holsteinian) interglacial but for the most part the record of this genus ends with the Mindel glaciation and the end of the Biharian age. The early Biharian small-mammal assemblages are linked to post-Villafranchian large mammals most notably in the southwestern U.S.S.R. (Kretzoi, 1965; Nikiforova et al., 1970), the Cromer Forest Beds, and to a certain extent in the Betfia fissure-fillings themselves (Kretzoi, 1965). The weight of this evidence is clearly to equate the Betfia-phase, defined by Kretzoi, with the first distinctively post-Villafranchian large- mammal assemblages.

equivalent t o the latter part of the Matuyama and early part of the Brunhes paleomagnetic epochs. A reversely magnetized basalt associated with Late Villafranchian (advanced Mummuthus rneridionalis) fauna of La Malouteyre (Auvergne) suggests an age older than 0.7 m.y. (BrunheslMatuyama boundary) for the base of the Biharian, inasmuch as the oldest Biharian (Elephas antiquus) fauna in the Auvergne, a t Solilhac, is in beds which also contain a pre-Brunhes, reversely magnetized basalt (Bout, 1970). In the Roman volcanic province, research summarized by Ambrosetti et al. (1972) shows that a

Small-mammal faunas of the Betfia-phase, with which the Biharian begins,

Radiometric and paleomagnetic evidence indicates that the Biharian Age is


116

cold-climate pollen flora and the latest known Villafranchian large-mammal fauna are associated with volcanics dated ca. 1.1 m.y., at Scoppino and the medieval abbey of Farneta respectively. The Farneta fauna is thought (Azza- roli and Ambrosetti, 1970) to be somewhat younger than that of Imola, in yellow sands on the southern margin of the Po Valley (Azzaroli and Berzi, 1970), first placed in the Calabrian Stage by Gignoux (1913) and later in the Milazzian Stage by Selli (1962). Radiometrically younger in age is a warm- climate pollen flora (Taxodium-type) ca. 1.0 m.y., and at Mt. Oliveto the earliest known Italian Biharian large-mammal assemblage dated ca. 0.95 m.y. The date on the pollen flora casts some further doubt on Lona et al.’s (1974) correlation of the extinction of the Taxodium flora to the base of the Plei- stocene, the base of the Villafranchian, the base of the Calabrian, and the beginning of the Donau glaciation. The first major glacial event of the Appe- nine region, the Flaminian glacial, is dated at ca. 0.7 m.y. This appears to be too early for the Mindel, and as we shall see below and in Chapter 7 on the chronology of Late Neogene climate events it very probably fits with the age of the Cromer Till and the end of the Menapian-Gunz cold phase, in medial Biharian time. Unfortunately no paleomagnetic analyses of the Roman volcanics is known to us.

A paleomagnetic reversal is ascribed to the BrunheslMatuyama boundary in the lowest of two (or possibly three) warm-climate maxima of the “Cro- merian” interglacial stage of The Netherlands (Van Montfrans and Hospers, 1969; Van Montfrans, 1971a,b; Zagwijn et al., 1971). The uppermost Waal- ian beds beneath the Menapian appear to have normal polarity ascribed to the Jaramillo Event (Van Montfrans, 1971b), and Waalian vertebrates are ascribed by Azzaroli (1970) to latest Villafranchian age. The Menapian cold-climate phase thus appears to cover the later part of the Matuyama epoch from ca. 0.9 to 0.7 m.y., and to overlap dates given to the earliest Biharian age. From this it appears that the last Villafranchian faunas which extend up to a cold-climate phase dated at 1.1 m.y. in Italy are also at least as young as the Jaramillo Event (ca. 0.91 m.y. midpoint, fide Dalrymple, 1972) below Menapian cold phase sediments in The Netherlands. The approximately 0.2-m.y. difference may or may not be real.

equivalent, cf. Van Eysinga, 1972) beds yield poorly diagnostic but clearly post-Villafranchian mammals (Kretzoi, 1965, p. 638), but the presence of the BrunheslMatuyama boundary in the Biharian Age is more clearly documented not only by the Auvergne dates but also by paleomagnetic stratigraphy in Eastern Europe. Nikiforova et al. (1970) have linked the main part of the Tiraspol mammalian sequence to deposits of Terrace V on the Dniester River in which Pevzner (1970) reports a paleomagnetic polarity change attributed to the 0.7-m.y. BrunheslMatuyama geomagnetic field reversal. Kretzoi (1965) indicates that the Tiraspol faunal complex includes levels ranging from earliest (Betfia-phase) to latest Biharian age, but would agree with Nikiforova et al. (1970) in correlating the main levels (associated with

So-called Taxandrian (“Cromerian” equivalent) and Needian (Holsteinian

117

the Bmnhes/Matuyama boundary) to Sussenborn and other sites which follow the cold-climate interval in the median Biharian - not at the base. Likewise, Kukla (1970) has shown that a rich Biharian fauna occurs in soils which lie virtually at the BrunheslMatuyama boundary. The first of Kukla’s (1970) seven loess cycles (Cycle J) and by inference the earliest glacial climate cycle of central Europe lies just below the paleomagnetic boundary, and can probably be correlated to the Menapian cold-climate interval. (It has been the custom in the Soviet Union to refer the pre-Tiraspolian, pre- Brunhes cold-climate phase to the Mindel glaciation and also to the base Pleistocene (cf. Nikiforova et al., 1970), but there is little to support either assumption.)

As to the true Cromerian of East Anglia, the biostratigraphy of this ex- ceedingly important sequence is only recently emerging from a dark age of antiquarianism. Apart from the large and obviously mixed collection of mammals from undocumented “Forest Bed” sites which makes up the bulk of the Cromerian material, Kretzoi (1965) nates that a typical Late Villa- franchian M i m o m y s assemblage comes from the so-called Lower Freshwater Beds and from the Weybourne Crag. According to the improved stratigraphy of West and Wilson (1966; see also West, 1968) this interval belongs in part to the Baventian cold phase and in part to the succeeding Pastonian warm phase, but the M i m o m y s fauna is not yet located more precisely than these limits. Van Montfrans (1971) indicates that the same interval probably has normal geomagnetic polarity, but its calibration is unknown. The Shelly Crag of West Runton (Pastonian warm phase?, cf. West, 1968) has a possibly more advanced Mirnomys assemblage which could be either Villafranchian or Early Biharian. The somewhat more intense Beestonian cold phase following the Pastonian has not been characterized, but the East Runton Upper Fresh- water Beds of the next following Cromerian warm phase (s.s.) has a Mimo- rnys-Microtus-Pitymys microtine assemblage quite clearly indicative of the Early Biharian (post-Betfia) fauna. The typical Cromerian beds are overlain by the Cromer Till (the earliest glacial deposits in England), followed by Corton Sand beds and then the thick, extensive Lowestoft Till. An isolated, sea-cliff exposure known as the Bacton Forest Bed is found at a single locality under Lowestoft Till; the absence of Cromer Till is supposed to be accidental and the Bacton Forest Bed is equated to the typical Cromerian (West and Banham, 1968), but two species of Arvicola, indicating late Biharian age at the maximum, have been collected from the Bacton Bed but not from any other Forest Bed locality (Kretzoi, 1965).

By analogy with the continental succession it would thus seem that the Cromer Till may be equivalent to the mid-Biharian cold phase which pre- cedes the appearance of Arvicola; this cold phase, which has an extrapolated age of ca. 0.7 m.y. and which appears to correspond with the ca. 0.7- 0.8-m.y. Flaminian glaciation in its relation to the Biharian mammal sequence, apparently correlates with an ice-rafting maximum of ca. 0.75 m.y. ago (see Late Neogene climate chronology, Chapter 7) and with the

118

terminal Menapian cold-climate phase just prior to the Brunhes/Matuyama polarity reversal. The typical Cromerian beds, in this model, are Early Bihar- ian and correspond to a Menapian interstadial from ca. 0.9 to 0.8 m.y. ago, and only the Bacton Forest Bed-Corton Sand interval correlates with The Netherlands “Cromerian”.

Oldenburgian (Kretzoi)

This age refers to Eurasian mammalian faunas of post-Biharian age, beginning with the Mindel glaciation about 0.5 m.y. ago, and includes the faunal levels observed in the Holsteinian interglacial period, as well as correlating t o the coeval post-Illinoian fauna of North America. The controversies associated with continental stratigraphy and mammalian paleontology of the last 0.5 m.y. are, unfortunately, beyond the scope of the authors’ experience and time prevents a fruitful review of the literature. On the scale of events to which we have been addressing ourselves there seems to be little to dispute the observation that post-0.5 m.y. faunas, K-Ar dates, and paleornagnetism all show little or no difference from those of the present day.

LATE NEOGENE IN ASIA

Mammalian biochronology in Asia during the last 14 m.y. is based on a European-like succession north of the Alpine-Himalayan orogenic belt and an African-like succession to the south. For general reviews of the Neogene faunas of this reeon the reader is referred to Colbert (1935), Pilgrim (1938), Kurt& (1956), Von Koenigswald (1956), Thenius (1959) and Borisiak (transl. 1962). Except for Japan, where mammalian information is scanty, no K - Ar dates have been applied to Late Neogene biostratigraphy in this entire region (cf. Ikebe et al., 1969), and paleomagnetic studies have been limited to Japan and a very small beginning in the Russian Pliocene on the Dniester River (Nikiforova et al., 1970).

As for marine correlations, again the major information from this great area comes from Japan (Ikebe et al., 1969, 1972), but this is partly due to the fact that after ca. 14 m.y. ago the Asian mainland was an area of plains and shallow, brackish seas; the marine formations that do potentially inter- finger with continental deposits are mainly on the Arctic shelf and in the little-studied coastline areas of East Asia.

The main problem in the Siwaliks Series of India and Pakistan is the correlation of the abundant collection of mammals of Late Neogene age to the world sequence, and this hinges on the controversy over the Chinji and Nagri stage boundaries. Owing t o the unfortunate habit of some of the Siwaliks collectors to purchase their specimens without closely ascertaining the place of origin (it may have seemed practical in the eyes of the sellers, furthermore, to keep a good spot under wraps) and to the confusing fact that the tillage of Chinji (not the Chinji Rest House for which the formation

119

was named) sits on an outcrop of Nagri beds, a re-study of the collections is not a simple matter.

The lithological boundary between the Chinji Formation and the Nagri Formation is generally taken as equivalent to a mammalian “stage” boundary, and in some interpretations an unconformity is placed between the formations as well (Hussain, 1971). Pilgrim (1938) reviewed the work of Matthew (1929), Colbert (1935) and Lewis (1937), who argued that the reported presence of Hipparion remains in the Chinji was of significance in correlating the Chinji level to early Hipparion faunas of North America and Western Europe, and (fide Matthew, 1929) to the Early Pliocene because this mammal first appears in the South Russian sequence in Pontian brackish- marine deposits, the molluscs of which had a “Pliocene” extinction percentage (see Van Couvering and Miller, 1971). In the face of this, Pilgrim (1938) maintained his earlier stand, that the Chinji had an inescapably “Middle Miocene” (i.e., pre-Vallesian) character which was out of place with the New World Hipparion correlation, and reiterated his theory that the Chinji Hip- parion was an instance of indigenous, parallel evolution. The likeness of the Chinji fauna t o that of pre-Vallesian levels in Africa (Fort Ternan) and Europe (La Grive, St. Gaudens), insofar as the data are not compromised by bad collecting, is re-emphasized by the recent comparison of the Fort Ternan Bovidae shown by Gentry (1970), and Hussain (1971) concludes that the first abundant Hipparion remains, ca. 300 m above the base of the Nagri, may in fact represent the first valid occurrence of the genus. As far as the controversy touches upon Ramapithecus, the earliest recognized member of Hominidae (Simons and Pilbeam, 1965) which occurs both at Fort Ternan (Kenya) and in the Siwalik Series, its attribution t o the Nagri Formation (’!-age) seems t o be the correct one (Pilbeam, 1972; G.E. Lewis, personal communication, 1973) so that the Kenya specimen is significantly older, whether Chinji = Fort Ternan or not.

According to Hussain’s interpretation, the Siwaliks can be calibrated second hand as follows: the 12.5-m.y. Hipparion Datum, in a Vallesian context, appears t o be located in the Nagri Formation faunal succession and significantly above its base. The Dhok Pathan fauna, next above, is characterized by the differentiation of Hipparion into gracile and massive species (Hussain, 1971) and the influx of steppic elements which marks the Turolian in Europe (Colbert, 1935; Thenius, 1959). The Tatrot and higher levels have not been studied for small mammals but the large mammals indicate a general parallelism t o the faunas of the South European Ruscinian and Villa- franchian.

In the Russian literature, the mammal ages and marine-brackish molluscan stages are regarded as indivisible, which sometimes makes it difficult t o understand the means used to correlate the faunal levels. The first appearance of Hipparion is ascribed t o the medial or lower “Sarmatian”, according to the provincial interpretation of Suess’ original term (in the Vienna Basin, the Sarmatian mammal assemblages end with the appearance of Hipparion but

120

there is some doubt as to whether this represents the earliest Hipparion in Europe; Thenius, 1959) and is understood to lie in the Miocene. Vertebrate paleontologists in the Soviet Union are accustomed to place the beginning of the Pliocene at the base of the “Maeotian” mammal age, with the development of the Turolian-like steppic fauna; this level has been correlated to the 10-m.y. old base of the Hemphillian in North America (Gabunia and Ruben- stein, 1968; Wilson, 1968), and in this our estimate of ca. 10 m.y. for the base of the Turolian is independently supported. It is however quite inconsistent with the probable age of the base of the Pliocene (see Time Scale, Fig. 1).

The Kvabebi fauna, of middle Akchagylian age in eastern Georgia, was discussed by Gabunia and Vekna (1968), who pointed out that numerous elements of the Rousillon faunal type are recorded in this collection. The authors correlated the Akchagylian to the Astian Stage, and thus t o the Villafranchian mammal age in a broad sense, but if the Rousillon level is actually present at Kvabebi it is of Middle Ruscinian age at the youngest and the Akchagylian is mostly pre-Villafranchian. The authors, however, included the Cimmerian Stage in their concept of Akchagylian and we have compromised, in our ignorance, to separate a Ruscinian = Cimmerian and Akchagylian S.S. = Early Villafranchian.

Nikiforova (1970) on the other hand proposed a correlation between the “Sarmatian” + Maeotian stages to the Messinian. Numerous K-Ar dates on the “Sarmatian” (Soviet sense) of Slovakia and the Caucasus (Konecn? et al., 1969; Bagdasarjan et al., 1971) indicate that the “Sarmatian” meant by this author began approximately 14 m.y. ago, whereas the oldest known date on the Messinian (disregarding the glass-shard age of 21.3 m.y. on middle Mes- sinian oeds reported by Selli, 1970, as too anoma ously old to be acceptable) is ca. 5 m.y. (Choubert e t al., 1968). Furthermore , as we have seen, the “Maeotian” mammals (if not molluscs) correlate quite confidently to 10 m.y. old levels in North America and Western Europe. Probably, this author was prey to the long-lasting assumption (cf. Papp, 1969) that the recession of the marine conditions in the Paratethys which marks the beginning of the Sarmatian is equivalent in time to the regression of the evaporat- ing Paleo-Mediterranean which marks the beginning of the Messinian, but the plate-tectonic motions which destroyed the Central Tethys between Africa and Asia were not coincident with the accident that closed off the Betic straits. Nikiforova (1970) shows a characteristic Villafranchian fauna, with Equus preceding the first elephantid, in Lower Akchagylian, and an unchar- acterized Cimmerian below. From her charts, the author indicates that the Cimmerian falls into the faunal interval between a representative Turolian assemblage (here “Pontian”) and the roughly Early Villafranchian (Lower Akchagylian), again suggesting that “Cimmerian” mammals would be of Ruscinian age.

In 1970, a group of Soviet paleontologists (Gromov, 1970; Nikiforova et al., 1970; Vangengeim and Sher, 1970) made a concerted effort to establish a “type locality” for the (continental) Pleistocene of Eastern Europe in the

EUROPE- :UROPEAN YEDITERRANEAN- MARINE PLANKTONIC CARIBBEAN

1,~(6

FAR

LARGER FORAMINIFERAL DATUM LEb'ELS AGE

STAGE FORAMlNlFl

CnATTIAN I I ='-I

4 - First oppeoronce

NORTH

EVENTS

OLD U VA I

HOH I WID5

P.ObOlEld.0

IEYINGFORD- ' I

IAN 1 IRIKAREEAN 7 VHITNEYAN

ORELLAN 11 CHADRONIAN I

Fig. 12. Biochronologic relationship between Late Paleogene and Neogene marine (foraminiferal) and non-marine (terrestrial vertebrates) and geologic events.

UROPEAN b4AMMAL

VILLA- 'RANCHIAN

USClNlAN

TUROLIAN

IURDIGALIM

IOUITANIAN

CHATTIAN

_ _ _ _ _

RUPELIAN

_ _ - _ -

YNNDlS lA l

122

Tiraspol alluvial series on the Dniester River, although there is no way that such a proposal could be accepted under present rules of stratigraphy (the authors did not refer to a comparative “type locality” for the Pleistocene of Western Europe, for instance, because there is none; also the Tiraspol complex represents only a very limited span of the Pleistocene time). Never- theless, the Tiraspol fauna remains the best-studied and best-dated Pleisto- cene fauna east of the Carpathians and is well suited as the basis for a Tiras- polian mammal age. Kretzoi (1965) has called attention to the fact that its sequence of fossiliferous levels appears to correspond to the complete Bihar- ian faunal span with few major lacunae. Furthermore, Pevzner (1970) has positioned a paleomagnetic reversal attributed to the Brunhes/Matuyama boundary (0.7 m.y.) in the main level (mid-Biharian faunal age, cf. Kretzoi, 1965). Thus the Tiraspolian is probably the same age as the Biharian and is shown to range from 0.9 to 0.6 m.y. on our chart (Fig. 11). Like the Bihar- ian of Western Europe, and Irvingtonian (Yarmouthian) of the North Ameri- can continent, it is terminated by the maxilhum Pleistocene glaciation, the Mindel-Illinoian, The advanced condition of some of the elephant molars from Tiraspol (high numbers of enamel lamellae) is said to be evidence for post-Mindel age (Nikiforova et al., 1970), but this is certainly a condition which could be expected to vary between extremes in a large sample. The material is referred to Murnrnu thus trogontherii by Gromov (1970) whereas its descendant, M. prirnigenius, characterizes the undoubted post-Mindel faunas of the Holsteinian-Hoxnian interglacial (Kurt&, 1968; West, 1968) *.

The stratigraphy of the terrace deposits of the Don Basin, Dneister River, and the estuary of the Danube has been related to that of the Tiraspol faunal complex and to the sequence developed in the Caspian Sea region (Chepa- lyga, 1967). The Tiraspolian, in turn, has been correlated by Soviet geologists (Nikiforova et al., 1970; Popov, 1970) with the Bakunian Stage of the Caspian Sea region in which a rich molluscan fauna has been described (Popov, 1955; Ivanova and Popov, 1961; Fedorov, 1957).

hominoid and human culture in the “Quaternary”, the base of which they believe to coincide with the base of the marine Astian = base Akchagylian (= - 3.3 m.y.) as a basis for replacing the name Quaternary with the term Anthropogene. However, this proposal has received little support outside the Soviet Union and is seen as an inaccurate means of subdividing geologic time. The problems of defining what constitutes earliest hominids (Pilbeam, 1972), and earliest evidence of “human culture” (Isaac, 1972) are quite obvious, to say nothing of the fact that the basic subdivisions of geologic time are traditionally based upon faunal changes recorded in the marine sedimentary rock record.

Fig. 1 2 shows the relationship of Late Neogene marine planktonic and continental mammalian biochronology and the geological events connected with the evolution of the Tethys Sea into the present-day Mediterranean Sea within the framework of global Neogene biochronology.

Gromov and Nikiforova (1968) stress the significance of the appearance of


CHAPTER 7

LATE NEOGENE MARINE PALEOCLIMATOLOGY

Climatic change is apparently the result of variables in the celestial me- chanics of the solar system. The total heat budget of the earth is affected by variations in the amount of incoming solar radiation and the seasonal and latitudinal distribution of incoming solar radiation over the earth’s surface (Milankovitch mechanism). The major cold periods of the Late Pleistocene have been found to coincide with intervals of decreasing winter insolation in the Northern Hemisphere and, in fact, gross climatic changes apparently origmate in winter-regimes on the continents of the Northern Hemisphere (Kukla, 1972).

The normal climatic condition of the earth during the past has been one which is free from polar ice cover. However, during the last ten million years the earth has entered a phase characterized by definite climatic deterioration, climaxed by the development of bi-polar sea-level ice sheets about three million years ago. It is the climatic history of these last ten million years, as recorded primarily in the deep sea, which we shall attempt to delineate below.

Late Neogene paleoclimatology has been treated in detail by Bandy and his colleagues (Bandy and Wade, 1967; Bandy, 1968a,b, 1969; Bandy and Casey, 1969a; Bandy et al., 1969a, 1971a). On the basis of integrated studies on planktonic foraminifera and radiolaria these workers outlined several major contributions to Late Neogene paleoclimatology.

was developed in which minor incursions of polar faunas into temperate regions were shown to correspond with minor incursions of temperate or sub-tropical planktonic faunas into tropical areas. The first observed expansion of the polar fauna was held to indicate polar cooling coincident with the formulation of a continental ice cap on Antarctica about 10 m.y. ago (Ban- dy, 1966, 1967b). In Bandy’s model, subsequent prominent expansions of the polar planktonic fauna, resulting from expansions of polar ice-forming climate, occurred during the Late Miocene and Pliocene in addition to the well-documented expansions of the Pleistocene. It was suggested that the pre-Pleistocene polar fauna expansions were accompanied by extension of the Greenland and Antarctic ice caps and perhaps in other areas of the Northern Hemisphere, but sub-polar glaciation in the Northern Hemisphere probably did not begin until about 3 m.y. (Berggren, 1972b), and Arctic ice sheets, if they existed, may not have formed down to sea level before that

(1) A model of the Late Neogene paleoclimatic history in the ocean basins

124

time. An Antarctic ice cap probably existed as long ago as Late Eocene, 40 m.y., (Le Masurier, 1972a,b) and may have extended to sea level earlier because of the refrigerative effect of the polar continental mass; 7--8 m.y. according to evidence cited by Rutford et al. (1970) and Denton et al. (1971), or some time around 4 m.y. according to Mercer (1973) and Margo- lis and Kennett (1970,1971).

(2) A model was proposed describing major sea-level changes in the Late Neogene in which the expansions of polar planktonic faunas were shown to correlate with periods of lowered sea level during the Late Miocene and Middle Pliocene and within the Pleistocene caused by growth of polar ice sheets. It was suggested that reductions in sea level during the Late Miocene and Middle Pliocene may not have been as great as those during the Pleisto- cene, while conversely the intervening warmer periods might have represented times of higher sea level and marine transgression. It was further suggested that the lowered sea levels of the Late Miocene may have been correlative with the restriction of circulation in the Mediterranean Sea during the Messinian Stage, resulting in closed marginal basins and deposition of evaporites. (A similar suggestion was made independently by Kennett, 196713, and Berggren, 1972a.) Indeed the suggestion of a causal relationship between evaporation of the Mediterranean Sea during the Late Miocene due to lowered sea level as a result of major glaciation in the Antarctic has recently received support from two lines of evidence: (a) the recent results of Leg 13 of the Deep Sea Drilling Project in the Mediterranean Sea; and (b) dating of the base of the Messinian at about 6.6 m.y. by integrated paleontological and paleomagnetic studies (see discussion elsewhere in this study). The date of - 6.6 m.y. for the Tortonian/Messinian boundary, that is the base of the evaporative phase in the Mediterranean Sea, is seen to correspond well with one set of dates for the expansion of the Antarctic ice sheet mentioned above. However, it should not be overlooked that sea-level lowering could not have isolated the Mediterranean basin until tectonism had previously reduced the openings to relatively constricted size, and that this might have eventually closed the connection completely.

(3) Variations from polar-type to tropical-type planktonic foraminifera1 faunas during the Late Neogene suggest surface summer water temperature changes on the order of 12°C (from less than 6°C to more than 18°C).

Although there appears to be general support for the general conclusions regarding paleoclimatic fluctuations during the Late Neogene by Bandy and his colleagues, there has been somewhat of a problem regarding the timing of these events. A discussion and comparison of the time framework used by Bandy and his colleagues in delineating Late Neogene biostratigraphy and paleoclimatology is presented elsewhere in this paper.

Two emergent shorelines in the southeastern United States may offer significant evidence in our description of Late Neogene glacial history. The Orangeburg scarp at an elevation of 200-250 f t . is the oldest and highest emergent marine shoreline in the Southeastern United States. It has been

125

suggested that it is of late Miocene age (Colquhoun and Duncan, 1964; John- son and DuBar, 1964; Alt and Brooks, 1965). According to J.E. Hazel (written communication, 1973): “It was probably cut during middle Miocene (Calvert) time or during the Pliocene. The late Miocene age is the result of misdating. Duplin Formation deposits occur under the terrace in front of the Orangeburg escarpment at elevations perhaps as high as 180 feet. Traditional- ly, the Duplin has been called upper Miocene, but it correlates without question with rocks containing Zone N12 planktonics.”

90-100 ft . It is generally considered to be of Pliocene age based on the occurrence of both land vertebrates and marine sediments underlying the terrace associated with the Surry scarp. These sediments contain marine faunas regarded as Pliocene in age (Mansfield, 1932; Cooke, 1945).

Alt and Brooks (1965) suggested that the abandonment of the Orangeburg scarp may have been caused by glacially controlled eustatic lowering connected with initiation of glaciation in Antarctica. Evidence cited elsewhere in this text suggests that glaciation in Antarctica reached sea level in a major advance about 6--7 m.y. ago and that the resultant lowering of sea level may have triggered the isolation and reduction of the western remnant of the Tethyan Sea. On the other hand, if the Surry scarp was abandoned in the Pliocene it may be connected with the initiation of sea-level glaciation in the Northern Hemisphere at 3 m.y. (see discussion below on paleoclimatology ) and a marine regression which J.E. Hazel (written communication, 1973) has recognized elsewhere in the Atlantic Coastal Plain at the same time. The age of the abandonment of the Surry scarp might then be correlated with the initiation of major sub-polar glaciation in the Northern Hemisphere, at about 1.5 m.y. (? the Nebraskan; see discussion below on paleoclimatology).

The next highest emergent shoreline is the Surry scarp at an elevation of

ATLANTIC OCEAN

Recent coring in the North Atlantic (Laughton, Berggren et al., 1970, 1972; Berggren, 1972b) has shown that the continental glacial history of the Northern Hemisphere can be extended back to three million years. And yet, in comparison with other parts of the world, our information on the stratigraphic record in this region is comparatively incomplete within this interval. This is due, in large part, to the relatively great thickness of terrigenous ice-rafted detritus in the late Pliocene-Pleistocene sediments.

The pioneering investigations on the Pleistocene paleoclimatic history of the Atlantic Ocean were made by Ericson and his colleagues at the Lamont- Doherty Geological Observatory (Ericson et al., 1956,1963, 1964a,b; Eric- son and Wollin, 1956a,b, 1968). As a result of detailed and laborious investigations on numerous deep-sea cores, a letter zonation of the Pleistocene was formulated (Zones Q through Z) based on the presence or absence of Globo- rotalia menardii. Those intervals in which menardii were absent were interpreted as “cold periods” and correlated with the classic continental glacia-

126

tions of Northern Europe, a procedure which had little justification at the time and which in retrospect has been extremely misleading. The paleoclimatic cycles thus derived were placed within a time framework originally based on a sedimentation rate extrapolated from l4 C and U/Th age measurements in the upper parts of various deep-sea sediment cores. This time framework was subsequently replaced by another one based on paleomagnetic stratigraphy by Glass et al. (1967) and Briskin and Berggren (1974; see Fig. 13).

The Milankovitch theory assumes that climatic fluctuation is a result of perturbations in the earth’s orbit and its axis of rotation. A general outline of the theory is presented in Broecker and Van Donk (1970), based on research of Broecker et al. (1968) and Mesolella et al. (1969) who showed a correlation between times of eccentricity maxima and reef growth on Bar- bados at 122,000, 103,000, and 82,000 years ago, within Ericson’s (warm) Z Zone.

On the other hand, Mathews and Imbrie (1971) and Mathews (1973) have pointed to some apparent discrepancies in the correlation of climatic history over the last 0.4 m.y. They explain this as due to the fact that some high stands of sea level reflect a warm Caribbean whereas others reflect a cool Caribbean. They indicate a general agreement between high sea level, coral reef development on Barbados, warm-water foraminifera (level 4), maximum summer insolation in the Caribbean and Northern Hemisphere and Termination I (of Broecker and Van Donk, 1970). Similar agreement was shown for terminations 11,111, and V. Terminations IV and VI do not fit this scheme, however. The foraminifera1 data suggest cooler Caribbean waters (level 3) at Termination IV, the Barbados data indicates an absence of coral reef development and the event occurs during subdued variation in Northern Hemi- sphere insolation. The authors explain this observation in the following manner: the highly negative oxygen isotope values associated with these terminations may reflect primarily the dissipation of continental glaciers independent of Caribbean sea-water temperatures (cf. Shackleton, 1967; Dansgaard and Tauber, 1969). In their view, climatic variation in the tropics may be strongly influenced by orbital eccentricity and precession, whereas at higher latitudes ice-melting may be influenced more directly by variations in tilt of the earth’s axis of rotation which has little effect on climate in the tropics. This model is furthermore seen to be consistent with the Milanko- vitch hypothesis. On the other hand, it has been recently suggested that variations in the earth’s magnetic intensity may modulate climate (Wollin et al., 1971).

Temperature variations in the Pleistocene, as reflected in the oxygen isotope ratios in fossil planktonic foraminifera, have been examined in detail by Emiliani (1955, 1961, 1964, 1966a,b) and Emiliani et al. (1961). This research demonstrates cyclical climatic variations within deep-sea cores having a periodicity of about 100,000 years. These cycles are numbered so that each warm half-cycle is odd and each cold half-cycle is even. The Emilia-

127

ni climatic cycles were placed within a time framework based on absolute ages obtained by the 231Pa/230Th method (Rosholt et al., 1961) which extended back to 150,000 years, subsequently extended by Emiliani (1966a) to about 425,000 years. More recently Broecker and Van Donk (1970) have reviewed evidence from various sources and have suggested that beyond the range of I4C dating the time-scale of Emiliani should be expanded by about 25%; for instance, they indicate that the age of 295,000 years Emiliani shows for the paleontological U/V boundary of Ericson et al. (1961) should be increased to about 400,000 years. Briskin and Berggren (1974) date this boundary at 430,000 years (see Fig. 13), and oxygen isotope investigations on a Pacific Ocean core substantiate this calibration (Shackleton and Op- dyke, 1973; N.J. Shackleton, personal communication, 1973).

I t may be expected that this error would increase down into the Pleisto- cene and indeed Hays and Berggren (1971) have presented evidence, based on paleontological cross correlation between cores and paleomagnetic stratigraphy, that Emiliani’s Pleistocene chronology may be low by a factor of 2 below the Brunhes/Matuyama boundary.

Olausson (1965), Shackleton (1967), and Dansgaard and Tauber (1969) have suggested that oxygen isotope measurements are a measure of the extraction of water from the oceans during periods of glaciation and the recirculation of this water during interglacial periods, rather than a true reflection of sea-water temperature. The oxygen isotope curve then would be an ice-volume, rather than a paleotemperature, curve.

climatic “stages” of Emiliani (1966a) beyond 16 to 22 in an equatorial Pacific core, Vema V28-238. A direct one-to-one correlation was shown between the carbonate minima (Arrhenius, 1952; Hays et al., 1969) and the stages of Emiliani for the past 0.9 m.y. The age of each stage boundary was calculated using an extrapolated uniform sedimentation rate between the top of the core and the Brunhes/Matuyama boundary at 12 m in the 16-m core. These isotope “stages” were suggested to serve as a standard for a time- stratigraphic subdivision of the latter half of the Pleistocene Epoch, inasmuch as it is based upon a non-reversible, synchronous process.

Broecker and Van Donk (1970) have pointed to the general asymmetrical (“sawtoothed”) shape of the major climatic oscillations in which a gradual cooling trend is terminated abruptly by a relatively rapid warming (the rapid warmings were called “0 “terminations”). Similar observations were made earlier by Deuser and Degens (1969) in which - 20,000 years evaporation cycles in the Red Sea were abruptly terminated by rapid (- 2,000 year) incursions of normal sea water from the Indian Ocean. These cycles were interpreted as being related to fluctuations in glaciation (Deuser and Degens, 1969; Berggren, 1969d; Berggren and Boersma, 1969).

Emiliani (1971) has recently presented a summary of data from which he has estimated glacial/interglacial temperature ranges in various parts of the world oceans and on the continents as summarized below:

Shackleton and Opdyke (1973) have extended the oxygen isotope marine

128

(1) In the seas, the maximum glacial/interglacial temperature range was 4" in the Atlantic and 2" in the Pacific. The amplitude of glacial/interglacial surface-water temperature range was 7-8" C in the Caribbean, 5-6" C in the equatorial Atlantic, and 3-4°C in the equatorial Pacific.

(2) In middle latitudes on the continents, the glacial/interglacial temperature range was on the order of 6-10°C (yearly mean) based on lowering of nivation sculptures in both the Alps and the Rocky Mountains, the southern extension of frozen soils and on pollen evidence.

(3) A t lower latitudes on the continents the glacial/interglacial range was on the order of 5-10", based upon the 10" latitudinal displacement toward the equator of the trade wind belt in Africa north of the equator, the lowering of nivation sculptures on Mount Kenya and New Guinea, forest biogeography in East Africa and pollen work in both South America and Southeast Asia.

Emiliani (1971) also obtained similar, but somewhat lower, figures for the Caribbean and Atlantic based on the distribution of the planktonic foraminiferal species Sphaeroidinella dehiscens and Pulleniatina obliquiloculata.

McIntyre (1967) constructed a paleoisotherm map of glacial and postglacial sediments of the North Atlantic based upon present latitudinal distribution and temperature ranges and optima of various coccolithophorid species (see also McIntyre and Bi., 1967). Glacial isotherms, it was suggested, were displaced 15-20" of latitude toward the equator, corresponding to a temperature decrease of 5--6"C below the February mean at low latitudes or about 10°C below the August mean or about 7.5"C below the yearly mean. More recently McIntyre et al. (197213) have demonstrated that in the eastern North Atlantic southward penetration of polar waters occurred at least six times in the past 225,000 years, the two most severe ones having occurred 165,000-135,000 years ago and 30,000-15,000 years ago (Fig. 7). At the same time a marked northward incursion of warmer subtropical faunal and floral elements occurred at least six times, the most pronounced being at 175,000 years, 125,000 years, and 8,000 years, respectively. The southern limits of the Pleistocene polar-water incursion was about latitude 42-45" N.

Pleistocene (the last 1.8-m.y.) faunal paleoclimatology in the eastern equatorial Atlantic has been described by Ruddiman (1971). Significant conclusions reached by Ruddiman include the following points:

(1) Two large-scale climatic shifts were shown to have occurred at 1.3 m.y. and at 0.9 m.y. In the case of the former the mean climatic condition deteriorated and short severe cold pulses punctuated the previously moderate warmth of the late Matuyama. In the case of the latter, the duration of cold intervals increased. Prior to the Jaramillo it was observed that no cold pulses exceeded 30,000 years, whereas three post-Jaramillo cold intervals ranged from 50,000 to 150,000 years in duration. The shortest and most recent of these corresponds to the Wisconsin glaciation.

(2) Although the absolute input rate of pelagic carbonate to sediments increases during cold intervals, the glacial carbonate percentage tends to

129

decrease during these times due to dilution by greater influxes of terrigenous detrital material. Carbonate percentages are depressed to very low values begmning in the Jaramillo, whereas pre-Jaramillo sections are generally calcareous oozes.

Ruddiqan (1971, p. 297) points out that the noteworthy factor in the Antarctic during the Brunhes is the prominent warming episodes relative to the uniformly cold latter portion of the Matuyama. By way of contrast the Northern Hemisphere data point to a nearly coincident climatic deterioration. The similarity of the paleoclimatic curves derived by Emiliani, McIn- tyre and Jantzen, and Ruddiman in the Atlantic and Caribbean during the Brunhes is substantial (Ruddiman, 1971, p. 298).

A major discrepamy between these Atlantic paleoclimhtic interpretations occurs in the lower Brunhes. McIntyre and Jantzen (1969) found the strong- est, most prominent North Atlantic carbonate maximum (inferred to reflect warm climates) at about 450,000 years. Studies by Imbrie and Kipp (1971) and Ruddiman (1971) found that a significant cold interval brackets this point and that this occurs within the U Zone of Ericson. Emiliani (1966a,b) had found a long, warm peak at a correlative level in Caribbean cores utilizing oxygen isotopes. Ruddiman (1971, p. 297) indicates that this anomaly requires one of two alternatives: (1) otherwise climatically indicative foraminiferal species in the equatorial Atlantic and Caribbean have registered a false cold period; or (2) an influx of 160-rich water into the Caribbean coincided with a temporarily reversed phasing of the usual carbonate- minimum/glacial-maximum pairing in the equatorial and North Atlantic. He points out that the dominant Antarctic warm-climatic pulse indicated by Kennett (1970) and Hays (1965) may be relevant in this connection. This very warm Antarctic fluctuation may have intruded upon and influenced the Northern Hemisphere cycle according to Ruddiman.

Ruddiman (1971, p. 299) points out that none of the long-term Pleisto- cene climatic curves can be considered definitive. Faunal curves are subject to non-thermal determining factors which were acting on specific elements in the fauna, and the carbonate curves were subject, in turn, to the intricate interplay of parameters which determine percent carbonate. It is gtill uncertain whether the detailed discrepancies among equatorial and Northern Hemisphere lithologic/paleontological studies are due to inadequate correlations between the paleomagnetically dated horizons, climatic phase differences from one area to another, or to collapse of essential assumptions of one or more of the methods, and he concludes, “that an exact chronology of Pleistocene glacial episodes and a regionally unified interpretation of that sequence remain unavailable”. The opposed phasing of carbonate minima and glacial maxima in the eastern equatorial Atlantic is primarily a consequence of influxes of terrigenous lutite from Africa which mask variations in net carbonate deposition. Terrigenous dilution may also explain the opposite timing of carbonate cycles between the Atlantic and Pacific oceans and obviate the necessity of ascribing it to a massive net transport of carbonate from one ocean to the other (Olausson, 1967).

130

Berggren (1968) made some semi-quantitative estimates on climatic cycles in the Pleistocene of the south-central North Atlantic (CH 61-171 core). Based on the relative abundance of Pulleniatina obliquiloculata and Sphaeroidinella dehiscens (warm indicators) to Globorotalia inflata + G. hirsuta (cold indicators) several climatic cycles were recognized over the last 1.8 m.y. If the paleomagnetic stratigraphy is re-interpreted and the dates on the Olduvai Event are changed from 1.8-1.95 to 1.61-1.79 then the estimates in Ta- ble XI1 may be made.

correspond favorably to the zonal scheme of Ericson et al. (1964b), Ericson and Wollin (1968), and Ruddiman (1971). A cool interval between - 1.7 and 1.5 m.y. is seen to correspond to the generally cool Q zone; a cooler interval between - 1.43 and 1.15 m.y. corresponds to the upper part of the R and lower part of the S (cool) zone and to include the significant temperature minimum recently noted at about 1.2-1.3 m.y. (Ruddiman, 1971; Briskin and Berggren, 1974).

These re-evaluated dates are only approximations but they may be seen to

TABLE XI1

Age estimates of climatic cycles in the last 1.8 m.y. I

hain 61-171 1 :;?on Zone

Suggested climatic variations (Berggren, 1968) I

1- 1.15

T

FI 1- 1.50 I / / / ////////// a - 1.56 V/LQY/h -

5 - 1.63 P I/ / / / / / / / / / / / / c) - 1.71

- 1.75 warm

I I I

White =warm; hatched = cool. * Reinterpreted by Briskin and Berggren (1974)

131

Quantitative paleoclimatologic investigations have recently led to more- precise estimates of late Pleistocene climatic conditions in the North Atlantic (Imbrie and Kipp, 1971; Hays et al., 1972a; Imbrie, 1972; Kellogg, 1972; McIntyre et al., 1972a). Transfer function treatment in which faunal and floral characteristics have been related to oceanographic parameters produced nine factors which yield winter/summer surface-water temperature, salinity and insolation from which paleoisotherm and oceanographic maps were drawn for an isochronous surface 17,000 years ago in the North Atlan- tic (McIntyre et al., 1972). The Polar Front which currently extends oblique- ly from Labrador to the north of Norway described a gentle curve nearly parallel to the latitude from Cape Hatteras to Spain 17,000 years ago. This represents a southward displacement of 20" lat. in the west to 30" in the east relative to today.

ago, referred to by many geologists as the Pleistocene/Holocene boundary) denote the division of the glacial North Atlantic by the Polar Front. The difference between today and the 17,000-year B.P. winter/summer temperatures for 50"N 30" W is 7.2-12.7"C to 1.2-6.6" C respectively, an average of 6°C colder than at present. South of the Polar Front the temperature 17,000 years ago was only about 3°C colder (McIntyre et al., 1972).

A quantitative planktonic foraminifera1 faunal analysis in the tropical North Atlantic by Briskin and Berggren (1974) has extended our understanding of climatic fluctuations back to two millions years. Winter changes are estimated to have fluctuated within a range of 4°C and summer temperatures within a range of 2" C. Pleistocene seasonal differences (T,-T, ) are estimated to have ranged between 1.5 and 4°C. Seasonality was greatest in the period between the Olduvai and late Jaramillo, ranging from a minimum of 2.5"C to a maximum of 4°C. During the latest Matuyama and Brunhes time seasonality ranged from 1.5"C to 3.3"C. Two general climatic regimes were recognized (see Fig. 7):

(1) From the Olduvai to late Jaramillo the amplitude of winter temperature contrasts (severe to mild) were subdued and on the average cooler (22.3-24.2"C) than during the Brunhes.

(2) During the late Jaramillo and the Brunhes, winter contrasts were greater and on the average warmer (23°C to 23.5"C).

Four roughly symmetrical major climatic cycles with periodicities of 500,000 years were recognized within the Pleistocene (Briskin and Berggren, 1974). These cycles are more clearly expressed in the winter estimate (T,) than in the summer estimate (T,). The coldest winters occurred at the following times (see Fig. 7):

The temperatures at the termination of the last glaciation (- 10,000 years

(1) in the lower R Zone (23°C) at 1.5 m.y.; (2) in the lower T zone (20.1"C) at the base of the Jaramillo, 960,000

(3) in the Brunhes (23"C), 610,000 years ago; (4) in the W and Y zones (23.6") 150,000 and 50,000 years ago.

years ago;

132

A climatic fluctuation with periodicity averaging 91,000 years becomes conspicuous in the interval from early V time (- 0.4 m.y.).

A comparison of “0 and the faunal index T, led the authors to suggest that the climatic shifts in the early and mid-Matuyama were associated with smaller ice-volume changes than those in the late Matuyama and the Brunhes, where ice volumes were of greater magnitude.

During the Pleistocene two types of cold interval were identified by the systematic change which occurs in the composition of typically cold-water species. In the Matuyama, cold episodes are characterized by the cold, high- latitude dextrally coiled Globigerina pachyderrna, whereas in the Brunhes cold episodes are characterized by temperate-cool Globorotalia inflata. The following circulation model was suggested. During warm intervals south equatorial waters penetrated the North Atlantic system. During cold intervals the Canaries Current was more vigorous and was displaced southward. Maximum displacement of the Canaries Current occurred during the Matuya- ma. In the Brunhes a moderate displacement of the Canaries Current brought temperate waters into the tropical province. Maximum penetration of south equatorial water occurred during V Zone time. During the Pleistocene this circulation pattern was repeated four times, once every 500,000 years.

ARCTIC OCEAN

It is over seventeen years since the Ewing-Donn theory (1956) was first promulgated, in which it was suggested that the Arctic Ocean remained ice-free during times of continental glaciation and ice-covered during interglacial times. Since that time there have been many papers dealing with the Late Neogene glacial history of the Arctic Ocean, the more important ones of which include Ericson and Wollin (1959), Ericson et al. (1964a,b), Steuer- wald et al. (1968), Herman (1969, 1970), Clark (1970, 1971), Hunkins et al. (1971), Steuerwald and Clark (1972).

A summary of previous studies based on material collected on the floating ice stations “Alpha” and “T3” is presented in Hunkins et al. (1971). Ericson et al. (1964a,b) found a good correlation between various peaks of foraminiferal abundance among cores in the Alpha cordillera. At that time however, it was assumed that abundance peaks correlated with glacial stages (open Arctic Ocean), while barren zones correlated with interglacial stages (frozen Arctic Ocean). Attempts were also made to correlate alternating cold and warm Arctic Ocean stages with continental glacial and interglacial stages, but as Clark (1971, p. 3317) points out these correlations have not been convincing, and indeed an alternative interpretation seems to have emerged from studies of the past decade.

The oldest erratics recorded in the Arctic Ocean have been dated within the Gauss Normal Epoch at a b m t 3 m.y. (Clark, 1971) although that author notes that the oldest sediments studied correlated with the Middle Pliocene at about 3.5 m.y. and that apparently the Arctic Ocean was frozen at least

133

by that time. In view of the evidence from the Labrador Sea in which the initiation of ice-rafting in the Northern Hemisphere has been dated at 3 m.y. it would seem that the assumption that the Arctic Ocean was frozen by 3.5 m.y. may be unwarranted. I t might be better to assume in the absence of other evidence to the contrary that the oldest erratics dated at about 3 m.y. in the Gauss in the Arctic Ocean do indeed represent the first evidence of glaciation in the vicinity.

Fluctuations in the abundance of planktonic foraminifera (primarily Glo bigerina pachyderrna) also suggest climatic variations within the Arctic Ocean although the interpretation of these results varies somewhat among the authors cited above. For instance Clark (1971) has found two abundance peaks in the Brunhes, one representing the Holocene and the other near the Brunhes/Matuyama boundary about 0.7 m.y. ago. Within the Matuyama two peaks were also observed, one within the middle part of the Matuyama and the other near the Olduvai Event. He has suggested that the intervals in which foraminifera are rare or absent represent environmental conditions different from those of today. Planktonic foraminifera have never been more abundant in the Arctic Ocean than today since the mid-Pliocene. He suggests that a thicker ice pack could have significantly affected the productivity of organisms whose economy is based on photosynthesis. Thus thicker ice conditions would correlate with periods in which planktonic foraminifera were absent or rare, whereas thinner ice conditions, similar to the present time, would have allowed a larger standing population of planktonic foraminifera to develop. He concludes that the surface of the Arctic Ocean has been frozen at least since the Middle Pliocene and that the most significant change in the Arctic ice cover has been its thickness.

Support for this idea was provided by Ku and Broecker (1967) who observed that sedimentation rates (0.2 cm/1,000 years) in the Arctic Ocean were relatively constant during the past 150,000 years and that the biologi. cal productivity rates have not exceeded those of the present-day ice-covered Arctic Ocean in the past 150,000 years and that these are much lower than in the Atlantic Ocean. The implications of these observations, according to Ku and Broecker (1967, p. 102), is that this does not favor theories of glaciation which call upon the influence of an open Arctic Ocean (cf. Ewing and Donn, 1956,1958; Donn and Shaw, 1967).

Hunkins et al. (1971) note that Globigerina pachyderma shows predomi- nantly left coiling during the Brunhes, whereas there is a right-coiling zone near the Jaramillo Event. This suggests warm conditions at that time, but this conflicts with the abundance data which shows a markedly lower percentage below the Brunhes/Matuyama boundary. The authors suggest that solution effects may have been more pronounced prior to a million years ago. The presence of manganese nodules and low foraminiferal counts in the Matuyama, at least to 1.5 m.y., suggest that climatic conditions were more uniform and without the wide fluctuations of the Brunhes epoch. Most of their cores show seven cycles in the Brunhes with periodicities of about

134

100,000 years. They conclude that the ice pack has been similar to that of the present day for the past 80,000 years but that prior to this time a number of cycles with conditions similar to today fluctuated with colder intervals during which the ice pack was extremely tight and during which there was no open water even in the summer.

A somewhat different interpretation has been presented by Herman (1969, 1970), in which it was concluded that the Arctic had been frozen only since the beginning of the Brunhes and that the preceeding several million years were times of warmer ice-free conditions in the Arctic Ocean. However, in the light of the recent results of Leg XI1 (North Atlantic) of the Deep Sea Drilling Project, in which it has been shown that major sea-level glaciation began 3 m.y. ago, indicating climate conditions similar to those of the present interglacial with seasonally (at least) frozen Arctic waters, this interpretation seems unlikely.

The conclusions of Clark (1971) are interesting in that they suggest that the ice cover of the Arctic Ocean remained relatively stable while continental glaciers expanded and retracted repeatedly in the Northern Hemisphere. If thus is true, the Arctic Ocean has not been a major factor in the growth or melting of continental glaciers. However, from a stratigraphic point of view one must object to the statement made in Steuenvald et al. (1968, p. 84):

“Our data along with those cited suggests that glaciation is evident in the polar seas at an earlier time than has been suggested on land. If this can be proved to correlate with Plio/Pleistocene boundaries elsewhere a greater time interval for the Pleistocene is evident.”

The Pliocene/Pleistocene boundary, as the senior author, among others, has been at pains to explain (Hays and Berggren, 1971; Briskin and Berggren, 1974), is a fixed boundary with a biostratigraphical not a paleoclimatological definition.

EQUATORIAL PACIFIC

The Late Neogene paleoclimatic history of the equatorial Pacific region has been described by Hays et al. (1969). Among the more significant conclusions reached by this group we may cite the following:

(1) Eight distinct carbonate cycles were recorded in the Brunhes (last 0.7 m.y.) with periodicities of about 75,000 years in the upper part to about 100,000 years in the lower part of the Brunhes. The carbonate peaks were correlated with glacial stages, whereas the troughs were correlated with interglacial states. This interpretation is supported by paleomagnetic and 14C dating of the last carbonate high which corresponds to the Wisconsin glaciation (80,000-11,500). The authors suggest therefore that eight major glacial fluctuations have occurred in the past 0.7 m.y.

135

(2) Good correlation is achieved between the carbonate cycles in the Pacific determined previously by Arrhenius and with those delineated by Ericson and Wollinj1968), and Emiliani (1966a,b).

( 3 ) Slightly lower carbonate content was observed in the interval between the Olduvai and the Brunhes, that is the late Matuyama. Average periodicities here were on the order of 100,000 years. The generally lower concentra- tion of calcium carbonate in the late Matuyama was interpreted as suggestive of warmer average climates during that time.

(4) A sharp rise in calcium carbonate content in the upper Gauss Normal series was thought to reflect mid-Pliocene cooling at about 3 m.y., which had been previously recorded in Iceland (Rutten and Wensink, 1960), in south- em Nevada at 2.7 m.y. (Curry, 1966), and in New Zealand older than 2.5 m.y. (Mathews and Curtis, 1966). Hays and Opdyke (1967) had previously reported pronounced faunal changes in Antarctic cores between 2.5 and 3 m.y. These observations have recently been verified by North Atlantic sea-floor evidence that sea-level glaciation was initiated in the Northem Hemisphere 3 m.y. ago (Berggren, 197213). Hays and his colleagues observed that no marked change in the carbonate content occurs during the Olduvai Event, although in Antarctic cores there is a lithologic change (see below) in the neighborhood of the Olduvai Event (Hays, 1965,1967; Opdyke et al., 1966) which may have been due to cooling during this interval. A general cooling trend across the Pliocene/Pleistocene boundary at Le Castella in Calabria, southern Italy, has been described by Emiliani et al. (1961) on the basis of oxygen isotope studies. However, whatever temperature changes may have occurred at the Pliocene/Pleistocene boundary were probably small compared with those observed at 3 m.y. and 0.7 m.y. ago.

ANTARCTIC-SUBANTARCTIC

The Late Neogene paleoclimatic history of the Southern Hemisphere (Antarctic-Subantarctic region in New Zealand) has been delineated in a series of papers by Kennett (1966a,c,d; 1968d; 1972), Margolis and Kennett (1970, 1971), Kennett et al. (1971), Watkins and Kennett (1971), Keany and Kennett (1973). Major climatic changes related to the expansion and contraction of the Antarctic ice sheet have been placed within a time-stratigraphic framework - the New Zealand stages - which have in turn been paleontologically correlated with the European stage units. Among the more significant results which have come out of this series of papers we may cite the following:

(1) Significant changes occur in planktonic foraminiferal faunal assemblages during the Late Miocene and Early Pliocene in New Zealand. The presence of abundant elements of cool-temperate planktonic foraminiferal faunas in the Kapitean Stage and their relative rarity in the Tongaporutuan Stage below suggests a significant cooling in the Late Miocene of this region (Kennett, 1968d). It was suggested (op. cit.) that a rapid cooling occurred

136

during the Late Miocene Tongaporutuan Age which culminated in the “uppermost Miocene” Kapitean Age*, which was followed in turn by warmer conditions during the Opoitian Age. The Kapitean faunas indicate oceanic temperatures cooler than those of the present-day New Zealand seas; apparently during Kapitean time Antarctic-Subantarctic water masses covered much of the New Zealand oceanic region. This has been corroborated by paleoecological data which indicates that there was a significant regression during the latest Tongaporutuan and succeeding Kapitean time and that a transgression occurred once more in the earliest part of the Opoitian Age (Kennett, 1967b). The lowering of sea level during the latest Miocene in New Zealand is attributed to glacio-eustatic rather than tectonic causes said to have coincided with late Miocene regressions elsewhere (California, the Caribbean region and the Mediterranean region).

been delineated in northern Antarctic and Subantarctic waters south of Australia and New Zealand, which are based on alternation of cold- and warmer-water planktonic foraminiferal faunas (Keany and Kennett, 1973). Eight climatic cycles over the last 1.3 m.y., six of which occur within the last 0.7 m.y., had been previously recorded in the study of Southern Ocean cores by Kennett (1970). The relative amplitude of the climatic warmings during the Matuyama is somewhat lower than those recorded in the Late Pleisto- cene (the last 0.7 m.y.) and this is taken to indicate more uniform Southern Ocean conditions during the Matuyama.

In contrast to most investigators who have claimed that the general pic- ture is one of declining climatic conditions during the Pleistocene, but with much colder climates during the Bmnhes (last 0.7 m.y.) than the late Matu- yama (- 1.8-0.7 m.y.), Keany and Kennett (1973) find evidence that Sub- antarctic climatic conditions were more stable and average temperatures lower during the late Matuyama than during the Brunhes, whereas the Brunhes, in general, was a time of erratic, high-intensity climatic cycles of generally warmer temperature range. One of the most convincing arguments for this view is the fact that Globorotalia inflata, which appears in the North Atlantic, the Mediterranean and New Zealand about 3 m.y. ago (Kennett et al., 1971; Berggren, 197213; Cita, 1973) appears for the first time in the Subantarctic region at the base of the Brunhes (0.7 m.y. ago). As Keany and Kennett (1973) point out, if Subantarctic-northern Antarctic conditions were temperate to subtropical during the Matuyama as suggested by Bandy et al. (1971a), G. inflata would be expected to occur throughout this interval as it does in New Zealand and equivalent latitudes. Evidence supporting Keany and Kennett (1973) was found in quantitative studies on planktonic foraminiferal faunas (Briskin and Berggren, 1974) in the North Atlantic, but

(2) Ten climatic cycles within the Matuyama (T = 2.43 to 0.7 m.y.) have

*It now seems more likely that the Kapitean Age straddles the Miocene/Pliocene boundary as defined in Western Europe.

137

is in conflict with paleoclimatic interpretations in the Southern Ocean based on distribution of radiolarian species by Bandy et al. (1971a) and Hays (1967) on the basis of variations in the abundance of temperate species of radiolarians. Bandy et al. (1971a) concluded that the Matuyama epoch was, in general, approximately 5" warmer than the Brunhes epoch. Hays (1967) concluded that there were two main periods of Late Cenozoic climatic cooling, one coinciding approximately with the Gauss/Matuyama boundary (2.5 m.y.) and the other coinciding approximately with the Jaramillo Event (0.9 m.y.). Although the paleoclimatic curve for the Southern Ocean (Ken- nett: 1970) during the last 1.3 m.y. is similar to that developed in the Atlan- tic-Caribbean area by Ericson and Wollin (1968), it was suggested that the individual paleoclimatic events were not necessarily synchronous in the two areas and that indeed the paleoclimatic sequence in the Southern Ocean has been out of phase or independent of Northern Hemisphere and equatorial climatic variations (Keany and Kennett, 1973). However, this latter conclusion would seem premature at this time inasmuch as the correlation of paleoclimatic events over large areas of the world is contingent upon the development of an extremely accurate time framework. The time framework in many areas is now provided by extrapolated sedimentation rates, and there is thus considerable room for error.

Fillon (1972) has recorded evidence of widespread submarine erosion in the Ross Sea sometime after late Gauss time (5 2.4. m.y.). A major change in sedimentation (calcareous benthonic foraminifera below, agglutinated foraminifera of late Brunhes age above) was shown to be due to a sharp decrease in the amount of debris delivered by the Ross Ice Shelf to the Ross Sea owing to a significant northward advance of the Ross Ice Shelf and thicker ice-pack formation in the Ross Sea. This, in turn, caused an increase in bottom scour. These conditions suggest climatic cooling which has also been suggested by Kennett et al. (1971), Watkins and Kennett (1971), and Keany and Kennett (1973). The latter have attributed increased erosion in the Southern Ocean and Tasman Sea to an increase in bottom-water produc- tion and velocities in post-Gilbert or post-Gauss time (- 3.0 to 2.4 m.y.).

(3) The Late Neogene paleoglacial history of Antarctica has been recorded in Subantarctic deep-sea cores (Margolis and Kennett, 1970, 1971) by the presence of ice-rafted grains and variations in planktonic foraminiferal diversity. Ice-rafted grains become locally abundant in the latest Miocene and Pliocene and are generally abundant within the Pleistocene, supporting the idea that the present Antarctic ice cap was initiated during the Late Miocene (approximately 6-7 m.y. ago) and grew to present-day proportions at least 4-5 m.y. ago.

On the other hand, geological and radiometric investigations on the exposed volcanic mountains of Marie Byrd Land by LeMasurier (1972b) indicate that a continental ice sheet was in existence on Antarctica as far back as Late Eocene time (40 m.y.). As this was a time when world climate was at a near optimum, and as a number of other volcanoes showing features indicat-

138

ing eruption under ice sheets of substantial thickness are dated from the Early and Middle Neogene of Marie Byrd Land and also West Antarctica (LeMasurier, 1972a; U.S. Geological Survey, 1970), the increase in glacially derived particles observed by Margolis and Kennett (1970, 1971) probably reflects a phenomenon other than the original initiation of Cenozoic glaciation in Antarctica, possibly the large-scale development of pack and shelf ice in the seas around Antarctica (see below).

Hays (1965, 1967) has observed that the distribution of radiolarians and various sediments in the Pleistocene in the Antarctic Ocean is strongly influenced by the northern limit of pack ice and the Antarctic Polar Front (Ant- arctic Convergence). The Pleistocene sediments between the Polar Front and the pack ice are primarily diatom oozes. At the boundary between radiolarian zones (a/X there is a lithologic change from clay below to diatom ooze above which has been interpreted (op. cit.) as having resulted from the initiation of large-scale freezing of sea ice around Antarctica which resulted in greater vertical circulation (upwelling) in the Antarctic Ocean and the development of high productivity of the Antarctic surface water which persists to the present day. Alternation of radiolarian and diatom-rich sediments with layers poor in siliceous microfossils north of the limit of modern pack ice as well as the alternation of warm- and cold-water radiolarian species were suggested to have been a reflection of northward expansion of the Antarctic pack ice.

Further back in time the evidence seems to indicate somewhat warmer conditions in the Antarctic region. At 4.5 m.y. various warm-water radiolarian species (Axoprunum stauraxonium and Heliodiscus astericus) have been recorded in Antarctic cores, and at 5.0-m.y. fragments of the diatom Ethmodiscus rex have been observed almost 20" farther south than the southernmost known occurrence of this species in Recent sediments (Hays and Opdyke, 1967, p. 1010). Bandy (1969) and Bandy et al. (1971a) indicate that subtropical conditions existed in the early part of the Gilbert epoch (< 4 m.y.) based on the presence of tropical collosphaerids in Subantarctic cores. Cold-water conditions were said to have developed in the late Gilbert, Gauss and early Matuyama (> 4- < 0.7 m.y. B.P.). Temperate conditions increased during most of the Matuyama indicated by the almost continuous presence in Antarctic cores of Pterocanium trilobum and Saturnulus planetes. A significant temperature reduction occurred at the Brunhes/Matuyama boundary as shown by the local disappearance of warm-water polycystines. Temperate conditions were re-introduced toward the end of the Brunhes (Holocene) as shown by the re-appearance of Saturnulus planetes and Theo- canus zancleus and other forms. Temperature cycles in the early Gilbert were said to have been on the order of 15-20°C (average summer), in the 5-15" range during the late Gilbert, Gauss and Matuyama, whereas 4 cooling cycles well below 5°C were said to have occurred during the Brunhes (Bandy et al., 1971a). The first appearance of ice-rafted detritus in Antarctic cores occurs at a point above the disappearance of the tropical collosphaerids (- 4 m.y.)

139

(Goodell and Watkins, 1966,1968; Goodell et al., 1968; Hays and Opdyke, 1967; Watkins and Goodell, 1967c; Bandy et al., 1971a).

Geitzenauer (1969, 1972) has investigated the Pleistocene calcareous nannoplankton biostratigraphy and paleoclimatology of the Subantarctic region. Several coccolith “stages” were recognized based on alternating high and low frequencies of Cyclococcolithus leptoporus, a eurythermal species characteristic of the Pliocene and Pleistocene (McIntyre et al., 1967). “Stages” I, 111, and V are considered “warm” and VI is the warmest, whereas “stages” I1 and IV are considered cold.

Within the last 400,000 years interglacial periods within the Subantarctic have been characterized by a relatively high frequency of Cyclococcolithus leptoporus and .Coccolithus pelagicus, whereas during glacial intervals these two forms were sharply reduced, the dominant form being the eurythermal Gephyrocapsa caribbeanica. A maximum peak of Gephyrocapsa Carib beanica occurred between 0.4 and 0.5 m.y. ago and resulted in a nearly monospecific coccolith ooze. This was correlated by Kennett (1970) with the climatic optimum which occurred during the V interglacial zone of Ericson and Wollin (1968) based on foraminiferal data. (See in this connection also Imbrie and Kipp, 1971, and Briskin and Berggren, 1974). The apparent contradiction between foraminiferal and coccolithophorid data within this specific interval (coccolith stage VI of Geitzenauer, 1972 = V interglacial zone of Ericson and Wollin, 1968) suggests that Gephyrocapsa caribbeanica exhibits an apparent bimodal distribution pattern with respect to temperature. These maxima occur at glacial stages and at the warmest interglacial. Geitzenauer (1972) cautions that a simple coccolithophorid-temperature relationship may be dangerous in Antarctic paleoclimatic interpretations; other factors such as salinity and nutrients may also have been significant determinants in the distribution of Pleistocene coccolith flora. Comparison of the coccolith “stages” (Geitzenauer, 1969,1972) of the Subantarctic region with the general climatic scheme of Ericson and Wollin (1968) shows a good correlation down to the V zone (to about 400,000 years) if one bears in mind the inherent margin of error in providing an accurate chronologic scale between these two regions.

of Coccolithus pelagicus and Cyclococcolithus leptoporus the similarity in frequency peaks in both of these species in the Upper Pleistocene indicates that the optimum conditions for both species were similar and that the climate of the “cool” interglacial stages of the Late Pleistocene were very similar to the present time (Geitzenauer, 1972).

climatic curves in the Atlantic, Caribbean, Pacific and Subantarctic during the Late Pliocene and Pleistocene. In Fig. 13 the climatic cycles of several investigators have been calibrated to the paleomagnetic time-scale and several of the more obvious correlations are shown, for instance the major cooling seen at about 1.2-1.3 m.y. in both the North Atlantic (Ruddiman, 1971; Briskin and Berggren, 1974), North Pacific (Kent et al., 1971) and the Sub-

On the basis of present-day temperature ranges governing the distribution

In summary it may be said that a substantial similarity exists between the

-19-

2.0-

-2.1 ~

-2 2-

-2 3-

s g

-24- E :

& . '

2 -

i' 3.4

Fig. 1.3. Correlation of some Late Pliocene and Pleistocene marine climatic events.

141

antarctic (Kennett, 1970; Keany and Kennett, 1973). Additional events which are detected over a larger area are the major Late Pleistocene cooling at about 0.6-0.4 m.y., and commonly attributed to the Mindel glaciation, as well as the cooling at about 150,000 years ago (Emiliani, 1966a,b; Ericson and Wollin, 1968; Broecker and Van Donk, 1970; Kennett, 1970; Kent e t al., 1971; McIntyre et al., 1972a; Keany and Kennett, 1973; Briskin and Berggren, 1974) which is probably attributable to the late (Haupt) Riss glaciation (see discussion in next chapter).

CHAPTER 8

A CHRONOLOGY OF LATE NEOGENE CLIMATE EVENTS

The framework of time is the sine qua non of understanding geological phenomena, whether it be rates of evolution, sedimentation, rise and fall of sea level, periods of orogeny, and so forth. This is no less true of glaciation. The evidence of glaciation is manifold and is reflected in both marine and non-marine sediments. Because of the nature of the evidence, however, the interpretation of the record has varied considerably, and this is reflected in the attempts to erect a chronology for the glacial record. In this section we shall present a basic chronologic framework for the Late Neogene climatological record, based upon an integration of the data recently obtained in the North Atlantic (Laughton, Berggren et al., 1970,1972) and the published literature dealing with marine (deep-sea cores) and non-marine sequences (see Fig. 14).

All too often the assumption is made that the “Glacial Period” is synonymous with the Quaternary. “Date the base of the earliest glaciation and you have dated the base of the Quaternary”, or from the other point of view “date the base of the Quaternary and you have dated the earliest glaciation”. This is bad stratigraphy, to say the least, if one is familiar with the Strati- graphic Code. On the contrary, the base of the Pleistocene is referred to a clearly defined biostratigraphic horizon and it is upon this base that the concept of a Pliocene/Pleistocene boundary must be founded. This boundary can be recognized by paleontologic means and has been dated by paleomagnetic stratigraphy at about two million years (Berggren et al., 1967; Glass et al., 1967; Phillips et al., 1968). Current discussion about the identification of one or more polarity events between 1.6 and 2.0 m.y. and their terminology are of secondary importance to our discussion. Recent work in the laboratory at Woods Hole suggests that the extinction level of discoasters occurs near the top of the Olduvai-Gilsa at about 1.6 m.y. and that the G. tosaensis-G. truncatulinoides transition appears to occur near the base of the Olduvai, between 1.8 and 1.75 m.y. The extinction of Globorotalia miocenica and G. exilis appears to occur at about 2.25 and 2.0 m.y., respectively. Referring these events to the typical Pleistocene of Italy, we prefer to use the base of the Olduvai in drawing the Pliocene/Pleistocene boundary and accordingly we place the Pliocene/Pleistocene boundary at about 1.8 m.y. The point is that this boundary has been narrowed down to a small time-span from estimates which formerly ranged from less than one million years to over three million years. Basic concepts relating to the Pliocene/

\I 31Wl

AlV

3

31Vl A

lW3

145

Pleistocene boundary have been discussed more fully by Hays and Berggren (1971).

Climatic change has been the most fundamental and distinctive charactqr- istic of the Pleistocene. However, it cannot be used as a criterion in determining or defining its limits. This idea has been held by most Quaternary specialists and persists in recent research (see Morrison, 1968). The most obvious flaw in this belief is the failure of “climatic deterioration” or “climatic change” or “climatic cooling” to provide a basis for recognizing the boundary between the Pliocene and Pleistocene epochs. Rather, it was the recognition of paleontologic criteria which could be correlated with the Italian “stratotype” section, and subsequently dated by paleomagnetic methods (Berggren et al., 1967; Phillips et al., 1968), which placed the Pleistocene within the proper time perspective for the first time. Within this time framework climatic cycles could be, and are now, properly recognized and positioned. Indeed, the base of the Pleistocene (as typified in Italy) is not characterized by a significant cooling.

The concept of equating the Quaternary with the ice age exists to the present day in the publications of Quaternary specialist and stratigraphic generalist alike. But there is ample evidence now that, although climatic deterioration occurred in the world during the Cenozoic, in the Northern Hemisphere this was accentuated in the Late Neogene and led to the extension of major ice sheets to sea level with consequent berg-calving and ice-rafting during the Pliocene some three million years ago. In the Southern Hemi- sphere definite evidence of ice-rafting has been encountered at levels older than 4 m.y. (Preliminary Reports of the Deep Sea Drilling Project, Leg 29) and glacial advances in the southern Andes are dated at least older than 2 m.y. (Mercer et al., 1972).

In general it has been assumed by Quaternary specialists that the four classic Alpine glacial stages (Gunz, Mindel, Riss, Wurm) can be correlated with those of North America (Nebraskan, Kansan, Illinoian, Wisconsinian). The older Donau and Biber have only recently been correlated with the American glacial sequence (Repenning, 1967; Richmond, 1970; Cooke, 1972) with indications that the conventional correlation is oversimplified. At the same time two alternative schools of thought have arisen regarding the chronology of the Late Neogene glaciations. These are the “short chronology” school (Emiliani, 1964, 1966a,b; Selli, 1967; Hays and Berggren, 1971;

Fig. 14. A suggested chronology for Late Pliocene-Pleistocene glaciation. Three significant glacial events are shown on the right side. They are: ( I ) expansion of polar ice-caps and sea-level ice at 3 m.y. in the Northern Hemisphere; (2) major glacial event in Early Pleistocene about 1.5 m.y.; (3 ) major Late Pleistocene glaciation at about 0.4 to 0.5 m.y.

The data listed in the columns under Paleoclimatic Cycles came from the following sources: Column 1 (Loess): Kukla (1970). Column 2 (Terminations): Broecker and van Donk (1970). Column 3(CaC03): Hays et al. (1969). Column 4 (Caribbean): Emiliani (1961, 1964, 1966a,b). Column 5 (Zones): Ericson et al. (1963), Ericson and Wollin (1968). Column 6 (Zones): Banner and Blow (1965b).

146

and most European Quaternary specialists) in which four major continental glaciations are compressed within the last 0.7 m.y. and in the case of Emilia- ni within the last 0.4 m.y.; and the “long chronology” school (Ericson et al., 1963, 1964b; Ericson and Wollin, 1968; Beard, 1969; and some Quaternary specialists) in which four or five major glaciations in the Northern Hemi- sphere are essentially equated with the Pleistocene Epoch and placed within a framework ranging back to dates from 1.5 to 2.5 m.y. ago.

There have been several attempts to relate the deep-sea climatic record to the classic glacial periods (Emiliani, 1961, l964,1966a,b; Ericson et al., 196413; Ericson and Wollin, 1968; Beard, 1969; Ruddiman, 1971). Emiliani’s (1961) earlier estimates placed the base of the glacial Pleistocene at about 0.3 m.y. and later (1966a,b) at 0.42 m.y., and the base of the Pleistocene Epoch at 0.6-0.8 m.y. (l961,1964,1966a,b). Ericson et al. (1964b) estimated the base of the Pleistocene (and onset of continental glaciation) at 1.5 m.y. Following the establishment of a paleomagnetic time-scale, they extended the Pleistocene glacial chronology to 2 m.y. (Ericson and Wollin, 1968). Beard (1969) subsequently extended the glacial chronology to 2.8 m.y., in suggesting that it was the Nebraskan glaciation and associated eustatic sea-level lowering that brought about (1) terrace incision and other signs of lowered base level in Gulf Coast deposits, and (2) concurrent indications of climatic deterioration in the microfossil record at that age level. In this respect the reader is referred to Kukla and Opdyke (1972) who indicate that the early Mississippi Delta terraces are of tectonic origin and are considerably older than the classic North American continental glaciations with which they have generally been correlated, and to Poag (1972) who demonstrates the presence of several pre-Pleistocene biostratigraphic units in the supposedly simple “Aftonian” (post-“Nebraskan”) Gulf Coast unit which had earlier been correlated with the type Pleistocene.

indicates unprecedented cooling trends and montane glacial advances, if not the formation of continental lowland ice sheets, beginning approximately 3 m.y. ago (Savage and Curtis, 1970). The oldest tillites in the Sierra Nevada, the Deadman Pass beds, are bracketed by 2.7 and 3.1 m.y. dates (Curry, 1966), and the summaries of Birkeland et al. (1971), Dalrymple (1972) and Cooke (1972) agree in general, if not in detail, in placing other glacial events between 3 and 1.5 m.y. Also in the southern Andes several montane glacial advances in a conformable sequence of tills and lavas have been dated prior to ca. 1.0 m.y. by Fleck et al. (1972) with the oldest tillite older than 2 m.y. In Europe, the Praetiglian cold-climate stage is marked by the first of numerous subarctic paleofloras seen in Late Neogene pollen suites (Van Montfrans, 1971b, p. 233; Zagwijn et al., 1971), and it is generally considered by Dutch geologists to be the earliest part of the Pleistocene in North Sea Basin sediments where its base is correlated with the Elphidium oregonense cold-water microfaunal zone (Van Voorthuysen et al., 1972). Vertebrate fossils, principally primitive Mammuthus “subplanifrons” (= meridionalis), indicate

In the continental sequences of the temperate regions, the evidence clearly

147

however, that the Praetiglian beds are equivalent in age or slightly older than the Middle Villafranchian, e.g. Roccaneyra local fauna ca. 2.5 m.y. (Azzaroli, 1970; V.J. Maglio, written communication, 1973) and paleomagnetic analysis agrees in placing the base of this stage “between 2.0 and 3.0, perhaps 2.3 m.y. old” (Van Montfrans, 1971b, p. 233). The equivalence of the base Villafranchian and the base Calabrian, which was assumed when the two Stages/Ages were both set equal to the beginning of the Pleistocene (Int. Geol. Congr., London, 1948) is clearly wrong; the Calabrian base is evidently more than 1 m.y. younger, but the effects of this mistaken assumption linger on.

The continental record, discussed in more detail elsewhere in this study, also seems clearly to indicate that the period of progressively more intense montane and high-latitude glacial activity which preceded major continental glaciation ended in North America with the Nebraskan glaciation, tentatively centered at 1.5 m.y., but in Europe (south of England) not until the Min- del-Elsterian glaciation dated roughly 0.6 m.y. Several authors (Richmond, 1970; Cooke, 1972 i. al.) have correlated the Donau-Eburonian cold-climate phase of Europe with the Nebraskan, and Repenning (1967) reached a similar conclusion based on mammalian correlations before the geophysical information was well developed. We have further concluded that the Kansan glaciation, which is apparently in the Matuyama paleomagnetic age (pre- 0.7 m.y) cannot correlate to the post-Matuyama deposits of the Elsterian (Van Montfrans, 1971b; Zagwijn et al., 1971), and that the Yarmouthian (post-Kansan) interglacial is characterized by Late Irvingtonian mammal faunas comparable in evolutionary grade and geophysical age to those of the Biharian mammal faunas of Europe which lived during the pre-Mindel interglacial. The Biharian/Villafranchian transition in Europe is marked by a cold-climate period, possibly a close-set series of stades and interstades, beginning close to 0.9 m.y., which broadly agrees with estimates of the age of major glacial activity in the North American mountains (cf. Cooke, 1972) and falls within the limits of 1.2 and 0.7 m.y. established for the Kansan glaciation. Based on mammalian biochronology the assumed correlation of continental paleoclimatic stages in the Northern Hemisphere is as follows:

Wisconsinian = Riss-Wurm = Saal-Wechsel; begins ca. 0.35 m.y. (includes Alt Riss) (Sangamonian = Holsteinian interglacial) Illinoian = Mindel = Elster; begins ca. 0.6 m.y. (Yarmouthian = “Cromerian” interglacial) (but not type Cromerian; see below) Kansan = Gunz = Menapian; begins ca. 0.9 m.y. (Aftonian = Waalian interglacial) Nebraskan = (?later part of) Donau = Eburonian; begins ca. ?1.6 m.y. (Blancan warm-climate phase = Tiglian “interglacial”) Early Blancan cold-climate phase = ?Biber = ?Praetiglian; begins ca. ?2.5-3 m.y.

148

As we shall see, the marine record contains features which cannot be expressed in such a simple chronology, and which suggest important refinements. Tb date there has been no general agreement on the correlation of the various glacial episodes on the continents with paleoclimatic cycles reflected in the deep-sea sediments and fossils, and it has all too often been the practice of oceanographers to identify a continental glacial event in marine cores according to some poorly tested paleoclimatological analogy. The difficulty has largely been with the nature of the data. The record on land, with the exception of loess deposits and certain lacustrine sequences beyond the glaciated regions, is essentially discontinuous because of the repeated passage of glaciers and the later erosion of their deposits. The preserved remains of a given “glacial stage” in most places represents but a small fraction of the time during which given climatological conditions prevailed and much of the history of events during that time is unavailable. On the other hand, in the deep-sea record one sees evidence of relatively continuous paleoclimatic changes on a relatively fine scale and for this reason marine paleoclimatological cycles may prove eventually to be the best means of providing a chronology of the Late Neogene glacial record. It is still premature, despite the greatly improved calibration of the continental record, to attempt more than a tentative correlation of the major and more general paleoclimatic trends in continental and deep-sea deposits, and the purpose of the present discussion is mainly to organize the available data within a reasonable chronological framework.

As criteria for “reasonableness” we are assuming the following: (1) the basic adglacial-deglacial climate curve has a sawtoothed character, expressed by a series of sub-cycles in which temperatures drop gradually (ice accumu- lates slowly) over periods averaging ca. 90,000 years, and then rise sharply (ice melts rapidly) in less than a tenth of the time (Broecker and Van Donk, 1970) and in which the base curve is an average of great numbers of these irregular secondary oscillations (Ericson et al., 1961,1963; Emiliani, 1955, 1961, 1964, 1966a,b; Ericson and Wollin, 1968; Imbrie and Kipp, 1971; Zagwijn et al., 1971); (2) climatological changes in the Northern Hemisphere (at least) were synchronous and parallel in sign, as shown by the general and growing agreement in the broad calibration of pre-Wisconsinian events discussed in this work and also by the much more refined 14C chronology of the later Wisconsinian (Weichselian, Wurm) and Holocene climate cycle in which even relatively insignificant events can be correlated between North America, Scandinavia and Siberia with temporal uncertainties on the order of only a few hundred years (Flint, 1971); and (3) past climates themselves are beyond our grasp, and climatic effects in the geological record may have properties which do not correspond quantitatively to the intensity of the causative climate cycle or to properties of other effects of the same cycle. Plainly put, this means that in some instances the climate effect in one set of data may be interpreted as a “mild cold phase” but will look like a “major glacial advance” in another, coeval set.

149

Our point of departure is the evidence of widespread glaciation in both the Northern and Southern Hemispheres by at least three million years ago. In the Southern Hemisphere Opdyke et al. (1966) found that in cores obtained from about latitude 60" S the oldest ice-rafted detritus was deposited at or near the Matuyama/Gauss boundary (2.4 m.y.). Subsequently Hays and Opdyke (1967) recognized ice-rafted detritus in a core from 65"s lat. at a level estimated to be about 4 m.y. old, and Goodell et al. (1968) presented evidence that ice-rafted detritus was present in Subantarctic cores at levels estimated from paleomagnetic stratigraphy to be greater than 5 m.y. old. Denton et al. (1971, p. 280) suggested that an extensive ice sheet was present in West Antarctica from 7 to 10 m.y. ago, and LeMasurier (1972a,b) found evidence for an Antarctic ice cap going back to 40 m.y. ago. However, Mercer (1973) suggested that (1) the formation of the Antarctic ice cap was associated with major cooling between 4 and 6 m.y. ago, (2) that the evidence is not conclusive for extensive glaciation in the Antarctic region before 7 m.y. ago, and (3) that glaciation dated in the Wrangell Mountains of Alaska by Denton and Armstrong (1969) at approximately 9-10 m.y. ago was probably montane and not correlative to continental (polar) ice sheet formation. In Marie Byrd Land, nevertheless, a number of volcanic mountains were formed underneath ice sheets more than 2,000 m thick at approximately the same 10-m.y. time (LeMasurier, 1972a,b). It seems probable therefore that a massive temporary extension of Antarctic shelf ice ca. 6 m.y. ago, as reported by very recent D.S.D.P. news releases, can be correlated with the cooling period cited by Mercer but does not necessarily indicate the initial formation of the Antarctic ice cap which may have been much earlier. While a major increase in Antarctic ice ca. 6 m.y. ago may not have had the same influence on sea levels as that calculated by Mercer (1973) for initial formation of the ice cap, its effects should have been noticeable in near-shore marine environments elsewhere if the build-up were rapid (ca. 1-2 m.y.) and might be dated in this way. One such example may be the sea-level lowering noted at the base of the Late Miocene Kapitean Stage of New Zealand, the base of which is probably correlative with the Messinian Stage of the Mediterranean. The initiation of Messinian evaporation in the Mediterranean Basin has been linked with eustatic sea-level change and Miocene glaciation (e.g. Bandy, 1969), and our present estimate of ca. 6.5-7 m.y. for the base Messinian suggests that the major expansion of Antarctic shelf ice and cold-climate conditions at 6-5 m.y. could have been responsible. On the other hand, the recent drilling in the Mediterranean by the D.S.D.P. (Ryan, Hsu et al., 1973) indicates that the evaporites and interbedded deep-water sediments were formed by often-repeated desiccation and refilling of the basin in a way which is incompatible with a major drop in sea level as a primary cause. Ryan, Hsu and their colleagues (1973) note that dehydration of the Mediter- ranean Basin would lead to a world sea-level rise of ca. 11 m. It seems reasonable that with a Messinian barrier of appropriate sill elevation this effect would create an oscillatory feedback equilibrium which would be reflected

150

in alternating cycles of dehydration and refilling such as those observed. That the Messinian barrier was mainly created by tectonism cannot be dis- puted, and it does not seem necessary to appeal to sea-level lowering to bring the sill elevation within the critical range for the first time. Indeed, for the equilibrium to be maintained it might be necessary that a slight increment in elevation be added to the sill height with each cycle to compensate for erosion during the “waterfall” phase (Hsu, Cita et al., 1973). On the other hand, it can be imagined that recession of the Late Miocene Antarctic ice shelf to present-day limits would be rapid enough, and of sufficient magnitude (cf. newspaper accounts of the D.S.D.P. findings) to sink the Messinian barrier below the critical depth until tectonic activity in the barrier region had ceased. In this way the Antarctic ice recession and the beginning of Pliocene normal marine sedimentation in the Mediterranean, both of which are dated at 5 m.y. or slightly younger, might be linked effects of a single climatological event (see Appendix, note 2).

The appearance of ice-rafted debris in deep ocean sediments, and the variations in the rate of deposition of ice-rafted material is an obvious effect of glaciation, and records the fact that (1) glacial ice sheets were breaking off in bergs at sea level, and (2) that the seasonal limit of floating ice extended beyond the sample locality. Berg-calving and ice-rafted sedimentation is going on today, so that we can assume that the initial appearance of this phenomenon does not imply formation of temperate continental ice sheets at the same time. Nor in the same way do other signs of “initial cooling” necessarily correspond to the initial continental glaciation, but only to critical levels in various paleoclimatically influenced features of the geological record “triggered” by progressive climate deterioration in the Late Neogene.

With this in mind, the first appearance of ice-rafted debris at 3 m.y. in the North Atlantic can be linked to several independent lines of evidence indicating a marked and irreversible cooling trend in this region at about the same age. (See also in this connection McDougall and Stipp, 1968, and Savage and Curtis, 1970.) In terms of biostratigraphy this event occurs approximately at the Reticulofenestra pseudoum bilica/Discoaster surculus boundary in calcareous nannoplankton zonation (Fig. 6) and at a level correlative with the lower part of Zone N 21 (although the system of tropical planktonic foraminiferal zones cannot be systematically applied to biostratigraphy in the North Atlantic). More precisely, the initiation of ice-rafting as seen in D.S.D.P. Site 111 in the Labrador Sea occurs just prior to the local extinction of Globoquadrina altispira, Sphaeroidinellopsis seminulina, S. subdehiscens and Globorotalia multicamerata, and it can be assumed that the disappearance of these forms is caused by the onset of glacial conditions in the region.

A branch of the Gulf Stream flowed along the eastern margin of New- foundland-Labrador during the Cretaceous and Tertiary. This was suddenly blocked by the appearance of icebergs in the Labrador Sea about 3 m.y. ago,

151

and the Gulf Stream was then displaced southwards t o its present position essentially south of latitude 45"N. The Labrador Current probably formed at this time as a response t o the drastic cooling in this region and the Polar Faunal Realm was established with the development of a cold-water fauna (characterized by the appearance of Globigerina pachyderrna and Globorota- lia inflata) in the North Atlantic. The development of a Polar Faunal Realm in the Late Pliocene is reflected in the scarcity of discoasters in North Atlan- tic cores as well.

McDougall and Wensink (1966) dated a normally magnetized lava flow in the Graue Stufe, Jokuldalar region, in northeastern Iceland at 3.10 f 0.10 m.y., and placed it within the Gauss Normal Epoch. They have noted that the oldest glacial intercalation in the Graue Stufe in this area occurs just below the reversed polarity lavas of the Mammoth Event and immediately above the dated basalt. The fact that older glaciations did not erode deep valleys in the basalts in contrast to the more recent glaciations suggest to the authors that the older tillites are ground moraines from extensive ice sheets rather than moraines from modern-type valley glaciers.

This agrees precisely with our evidence in North Atlantic cores where the first ice-rafted detritus appears a t 3 m.y., but i t is probably not valid to suggest that this was the time of the first major expansion of lowland ice sheets over Europe and North America, as well as the maximal mountain glaciations of the Sierra Nevadas, the Rockies, and the Alps, as has been suggested by Curry (1966, p. 771).

The record of ice-rafting in the North Atlantic, most recently summarized by Briskin and Berggren (1974), can be closely compared with the studies of Conolly and Ewing (1970) and Kent e t al. (1971) in the North Pacific. According t o the revised ice-rafting stratigraphy of the latter paper, trans- ported glacial debris first appears at about 2.5 m.y. in North Pacific cores, but this should be viewed in the light of the present-day ice limit in this region which suggests that expansion of the berg range into the North Pacific would possibly have been slower than in the North Atlantic.

Referring t o Fig. 13, comparison of the stratigraphy outlined by Kent e t al. (1971) and that described in North Atlantic core V16-205 by Briskin and Berggren (1974) shows a progressive increase in the average rate of ice-rafted debris in both ocean areas and indicates that new base levels of climatic deterioration are reached in the Northern Hemisphere a t approximately 1.3-1.2 m.y. ago, and again at approximately 0.9 m.y. (North Pacific) or 0.8 m.y. (North Atlantic). Minor, low-amplitude peaks in the ice-rafted sedimentation at 1.2, 1.1 and 0.95 m.y. in the North Pacific (Kent e t al., 1970, fig. 3) are clearly paralleled by temperature minima in North Atlantic core V16-205, and a major ice-rafting peak between 0.82 and 0.75 in the North Pacific may correspond t o a cooling trend beginning a t ca. 0.8 m.y. in

Corroboration of the ice-rafting chronology of glaciation in the Northern V16-205.

Hemisphere comes from studies on the stratigraphy of Tjornes, northern

152

Iceland (Einarsson et al., 1967). A dramatic influx of Pacific boreal molluscs occurred in Iceland and the North Sea Basin either slightly more than 3.0 m.y. ago or more than 3.35 m.y. ago (depending upon the interpretation of paleomagnetic measurements made on the volcanic rocks of Tjornes). Einarsson et al. (1967, p. 322) point out that this invasion occurred:

“at least half a million and perhaps more than a million years before the first Pleistocene glaciation was recorded there. The migration must have taken place at a time when the Arctic Ocean was warmer than at present, for some of the migrating taxa no longer range as far north as the Arctic Ocean.”

Ten glacial cycles alternating with nine interglacial intervals are recorded in the Tjornes sequences and indicate that (Einarsson et al., 1967, p. 324):

“Pleistocene climatic history is much more complex than is suggested by the classical concepts evolved for the Alps and for the mid-continent of North America, which involve only four or five glaciations, and it casts serious doubt on intercontinental correlations of the older Quaternary glacial deposits.”

found a distinct shift towards cooler ocean waters in the Central Atlantic about 1.3 m.y. ago and the beginning of the first sustained cold-water event at ca. 0.9 m.y. Likewise, Donahue (1970) found an increase in the proportions of cold-water diatoms across the interval marked by the Jaramillo 0.9-m.y. Event. Additional evidence for a cold phase of this age comes from D.S.D.P. site 36 in the eastern Pacific, although in the absence of direct calibration of the paleomagnetic sequence the cooling trend may actually be associated with the Brunhes/Matuyama reversal rather than the Jaramillo. The ice-rafting peak of approximately 0.75 m.y. ago seen in North Pacific and North Atlantic cores is not clearly evident in the paleontological record; instead, Ruddiman (1971, figs. 12 and 14) shows a general, intensive cold- water interval from 0.9 to 0.775 m.y. A slight warming trend is seen in the ice-rafting stratigraphy at the Brunhes/Matuyama boundary, from ca. 0.7 to 0.65 m.y. in the North Atlantic (Briskin and Berggren, 1974) and from 0.71 to 0.69 in the North Pacific (Kent et al., 1971). Ruddiman (1971, fig. 14) shows a similar warming trend at the same level.

In Europe the Brunhes/Matuyama boundary has been determined in the “Cromerian” interglacial deposits of The Netherlands (Van Montfrans and Hospers, 1969; Van Montfrans, 1971a,b) and in the Cerveny Kopic loess sequence of Czechoslovakia (Bucha et al., 1969). Nine complete loess cycles, equated with glacial cycles, and lettered B through J from younger to older, have been recognized by Kukla (1969a). The Brunhes/Matuyama boundary occurs in cycle I; thus at least one glacial cycle (cycle J) lies in the latest Matuyama and 7 or 8 within the Brunhes (last 0.7 m.y.). Loess cycle J was correlated by Kent et al. (1971, p. 2751) with the cold (glacial) event in the North Pacific dated at about 0.75 m.y. ago and with the Sherwin Till of the

In studies of climatically influenced micropaleontology, Ruddiman (1971)

153

Sierra Nevada. The pre-“Cromerian” Menapian cold-climate deposits of The Netherlands were correlated with cycle 3 of Czechoslovakia as well.

In Japan, the Gokenya cold phase was determined paleomagnetically to lie between the Jaramillo Normal Event (ca. 0.9 m.y.) and the Brunhes/Matuya- ma boundary (ca. 0.7 m.y.) by Ikebe et al. (1972), and Van Montfrans (1971a) indicates that these paleomagnetic events occur at the base and just above the Menapian cold phase, respectively. The end of the Kansan glaciation is also dated close to 0.7 m.y. (Izett et al., 1972). This agrees with Ruddiman’s 0.9-0.775-m.y. cold-water interval, but on the other hand evidence from the marine paleontological record suggests that, in general, climate in the interval between ca. 0.9 and 0.7 m.y. was relatively warmer than that of the following period (Gamper-Bravo, 1971; J. Imbrie, personal communication, 1973; Briskin and Berggren, 1974), and this data would tend to agree with K-Ar and paleomagnetic evidence which locates the Biharian land mammal fauna in the “interglacial” interval between a terminal-Villafranchian cold-climate period ca. 0.9 m.y. ago and the beginning of the Mindel glaciation, ca. 0.6 m.y. ago. As we have emphasized in Chapter 6 on mammalian biochronology , the Biharian/Villafranchian transition is characterized by an extremely marked change in the European fauna; a much less pronounced change is associated with a medial Biharian cold phase. Thus, if all the calibrations and climate interpretations are correct, we are faced with various lines of evidence which indicate: (1) a sustained cold- climate interval from 0.9 to ca. 0.75 m.y.; (2) a warm-climate interval in the same period; (3) a major cold-climate peak at ca. 0.75 m.y. but not at 0.9 m.y.; and (4) a major cold-climate peak at 0.9 m.y. but not at 0.75 m.y.

The solution to this dilemma, if it is even partly real, is probably to be sought in the third “rule of reason” above, the quantitative inconsistency of different secondary effects. It is suggested here that the Northern Hemi- sphere climate in the latest part of the Matuyama Reversed Epoch ( 0 . 9 4 . 7 m.y.) consisted of two temperature minima separated by a period of moderate climate which was cooler than the preceding “interglacial” but possibly not as cool as the following “interglacial” of the early Brunhes epoch. In the high resolution offered by ice-rafting stratigraphy it can be seen that the 0.9-m.y. minimum observed in North Pacific and North Atlantic cores was a relatively minor fluctuation close to the beginning of a relatively rapid decline in the temperature base curve. Because of its location on the base curve it could have had a disproportionate effect on marine and land mammal faunas, such as that observed a t the coeval Biharian/Villafranchian boundary. The succeeding “interglacial” was characterized by a higher level of ice-rafting than hitherto (Kent et al., 1971) and (by inference) relatively cooler temperatures, such that cold-indicative fossil assemblages might have been sustained (cf. Ruddiman, 1971). The relatively more intense 0.8-0.75-m.y. thermal minimum at the end of this period was less influential on the previously modified marine and mammalian faunas than on non-adaptable re- sponses such as the rate of ice formation and berg-calving.

154

In this r%spect, it should be noted that Zeuner (1959) determined that the Gune glacial episode in the Alps was a doublet, and that Kretzoi (1965) observed a distinct cold-climate interval in the Biharian faunal succession a little less than midway above the 0.9-m.y. old base; on an uncalibrated interpolation this medial cold phase would be ca. 0.75 m.y. The earliest observed till in East Anglia (Cromer Till) appears to have formed not a t the beginning of the Biharian Age, but rather a t approximately the same faunal level where Kretzoi places the medial cold phase on the continent. This indicates that in Europe the most notable continental glaciofluvial event of the pre-Brunhes time did not correspond t o the marked paleontological change at ca. 0.9 m.y. but rather to the ice-rafting peak at ca. 0.75 m.y. On this basis the most intense cold-climate phase of the Menapian, just across the channel to the southwest, should be its later part just prior to the Brun- hes/Matuyama boundary, and is to be correlated with the Cromer Till. The post-Menapian “Cromerian” of The Netherlands, which has never been satisfactorily correlated with the type Cromer Beds below the Cromer Till (West, 1968; Zagwijn et al., 1971) is equivalent to the Late Biharian Bacton Forest Beds only. The type Cromer, on the other hand, is equivalent to a (still unrecognized) Menapian interstadial, age from ca. 0.8 t o 0.9 m.y., since these beds appear to have a normal paleomagnetic sign according with an age within the Jaramillo Normal, and the earliest Menapian cold phase lies just above or within the Jaramillo (Van Montfrans, 1971a). Mammalian evidence, which locates the Biharian/Villafranchian age boundary near the base of the type Cromerian and also near the Waalian/Menapian stage boundary, and which furthermore places the Brunhes/Matuyama paleomagnetic boundary in a later Biharian faunal sequence at Tiraspol (see Mammalian Biochronolo- gy, Chapter 6) is in full agreement. I t is also logical, if not proven, t o assume that the later of two Gunz advances would correspond t o the Cromer Till and the Upper Menapian, although vertebrate paleontologists naturally looked t o the changes wrought by the relatively minor first advance t o signify the maximum Gunz glaciation.

ly known, but evidence has been produced for two glacial maxima in the Kansan episode (G. Izett, personal communication, 1973) and geophysical dating of ash beds indicates that the Kansan, which is known t o predate 0.6 m.y., probably ended just prior t o 0.7 m.y. ago (Cooke, 1972; Izett e t al., 1972); presumably a Kansan interstadial with Jaramillo/late Matuyama paleomagnetic signature and Early Biharian-like fauna can eventually be recognized. Kukla and Opdyke (1972) report a (+/-) paleomagnetic reversal in middle Kansan deposits which they suggest is equivalent t o the Brunhes/ Matuyama boundary; if correct this would put the Kansan late stages, if not late maximum glaciation, younger than 0.7 m.y. a t a time when most other calibrated indices show a warming trend or interglacial conditions. The ash stratigraphy, which puts the 0.7-m.y. Bishop Tuff above Sherwinn Till in the Sierra Nevada (Birkeland et al., 1971) and below “S”-type Pearlette ash but

The stratigraphy of this interval in North America is much less satisfactori-

155

above pluvio-glacial levels in Lake Bonneville cores (Izett et al., 1970; Shuey et al., 1970), rather strongly suggests, if it does not prove, that this reversal is the Jaramillo base instead.

Six climatic warmings within the last 0.7 m.y. in the Subantarctic were observed by Kennett (1969). A significant warm interval was noted within his Globorotalia inflata Zone and was dated at approximately 0.5-0.4 m.y. ago. Kent et al. (1971, p. 2749) have pointed to the coincidence of this and the period of low ice-rafting and warmer surface water (based on diatom evidence) in the North Pacific between 0.53 and 0.46 m.y. ago.

Pacific (Hays et al., 1969) to the seven ice-rafting cycles observed in the North Pacific was suggested by Kent et al. (1971, fig. 7 on p. 2750; see Fig. 13 this study). This supports the idea that high carbonate levels in the equatorial Pacific are a reflection of glaciation at high latitudes, at least within the Brunhes, and probably within the Matuyama as well.

Although correlation of the Wurm-Weichsel = Wisconsin is generally agreed upon, the Riss and Mindel include moraines and terraces of multiple glacial cycles and this correlation with their North American counterparts are still hypothetical and speculative. Again, reference to the more completely preserved cycles in the deep-sea sediments allow a first-order approxima- tion of gross spatial and temporal relationships between marine and continental glacial events.

The correlation of the U Zone of Ericson et al. (1963,196410) with the Mindel is generally agreed upon (Broecker and Van Donk, 1970; Kukla, 1970; Berggren, 197213; Cooke, 1972; J. Imbrie, personal communication, 1973). However, the age determinations on its limits have varied considerably. We have seen above that it has been correlated with an interval of intense ice-rafting between 0.7 and 0.5 m.y., whereas the Mindel/Riss interglacial is correlated with the interval between 0.53 and 0.46 m.y. (Kent et al., 1971). The data of Briskin and Berggren (1974) suggest that a long cold cycle occurred between 0.65 and 0.43 m.y. with a peak at about 0.6 m.y., within the upper part of the T Zone.

The trend in V16-205 is towards a gradual warming over the interval of 0 -45-0.275 m.y . , after which a gradual cooling is once more resumed which reaches a maximum in the W Zone (- 0.18-0.13 m.y.). (As will be seen below, this is just the time interval estimated by Richmond, 1970, p. 11, for the Sangamon interglacial.) In correlating the Mindel with the period between - 0.7 and 0.5 and the Mindel/Riss with the succeeding ice-rafted low between 0.53 and 0.46 and the Alt Riss with the period between - 0.27 and 0.29 m.y., Kent et al. (1971, p. 2752, fig. 8) are left with a large question mark for the interval between 0.45 and 0.3 m.y. This, as we have shown, corresponds to a time of gradually increasing temperatures in the North Atlantic within the V Zone (see also Ruddiman, 1971, p. 287, fig. 4). This interval is reinterpreted here as roughly equivalent to the Mindel/Riss (Hol- steinian) interglacial in part (see discussion below).

A correlation of the eight Brunhes age carbonate cycles in the equatorial

156

The Mindel as envisaged here (- 0.65-0.43) corresponds to about a third of the Brunhes Normal Epoch and encompasses several of the carbonate maxima observed in equatorial Pacific cores by Hays et al. (1969). The U Zone of Ericson and Wollin (Ericson et al., 1963; Ericson and Wollin, 1968) contains a long significant cold faunal episode at about > 0.4-0.5 m.y. (Ruddiman, 1971; Briskin and Berggren, 1974). J. Imbrie (personal communication, 1973) has found that the U Zone has the coldest climatic extremes of the Late Pleistocene and he is therefore of the opinion that they correlate with the Mindel glaciation at least in part.

On the basis of the marine data, the Mindel/Riss interglacial is estimated here to span the interval between approximately 0.4 and 0.3 m.y. The Hol- steinian (= Mindel/Riss) interglacial is shown in Fig. 14 to correspond to the lower, warmer part of Kukla’s (1969a) loess cycle E and the upper part of loess cycle F.

glacial specialists (see summary in Richmond, 1970). The Riss is correlated by Richmond (1970) with the Rocky Mountain Bull Lake glaciation, which is in turn correlated with the lower part of the Altonian substage of the Wisconsinian. Richmond (1970, p. 15) suggests an age span of ca. 0.12 m.y. to some time older than 0.05 m.y. for the Riss, but this is because he correlates only the type Riss to later Bull Lake advance.

On the other hand, Kukla (1970, p. 162, fig. 10) indicates that the Saalian (which is generally correlated with the Riss) is correlative with his loess cycle D and/or E. Cycle D and E span the time interval of approximately 0.34 to 0.22 m.y. (J. Imbrie, personal communication, 1973), which is considerably older than the time interval estimated by Richmond for the Riss interglacial (- 0.12-0.075 m.y.). Richmond (1970, p. 12) discusses evidence for a separate glaciation between the type Mindel and type Riss. These deposits are generally referred to as late Mindel or Alt Riss by other specialists. If we include them in the older part of the Riss, then they can probably be correlated with the Early Saalian at ca. 0.35 m.y.

The marine record suggests that a positive or near-positive sea-level stand occurred about 200,000 years ago (Broecker and Van Donk, 1970). Follow- ing this there is a cold interval corresponding to the upper part of Kukla’s (1969a) loess cycle C. This may correspond to the upper, or type Riss. Broecker and Van Donk (1970, p. 185) point out that the ice sheets that had built up by 130,000 years ago were largely destroyed during Termination I1 (- 127,000 years). Although glaciers grew anew after maximum deglacia- tion, - 124,000 years ago, their expansion was interrupted by two periods of reef coral growth (Barbados Terrace 11: - 103,000 years ago; and Barbados Terrace I: 82,000 years ago) (Mesolella et al., 1969). This interglacial interval, preceding the most recent expansion of glaciers, the Wisconsin-Wurm, is correlated with the high sea-level stand : Barbados Terrace 111, at about 125,000 years ago and may represent the Sangamon = Eemian (Broecker et al., 1968; Mesolella et al., 1969). In Fig. 14 the Sangamonian is correlated

The limits of the Riss have been the subject of considerable debate among

157

with the Eemian and is estimated to have spanned the time interval of - 125,000 to 75,000 years ago. Richmond (1970, p. 15) suggests that the Bull Lake glaciation began

about 120,000--130,000 years ago (and was correlative with the Riss, exclusive of the Alt Riss). As we have discussed above the division between Alt Riss and type Riss is lowered to about 180,000 years (Fig. 14) so that the type Riss is equivalent to the cold period corresponding to the upper part of Kukla’s (1969a, fig. 10) loess cycle C . Accordingly, the base of the Bull Lake has been drawn at 180,000 years to correspond with the base of the type Riss in Fig. 14.

The Wurm is generally divided into two parts: an older, Alt Wurm, and a younger, or Main Wurm. Richmond (1970) places the boundary at about 28,000 years; he correlates a Rocky Mountain Bull Lake-Pinedale glaciation with the Alt Wurm, and the Pinedale glaciation itself with the Main Wurm. In northern Europe the Weichsel is correlative with the Wurm. Deep-sea sediments record two maximum cold periods in the Wisconsin-Wurm: at about 70,000 years and about 17,000 years.

The find recession of the Main W i i m was completed about 11,000 years ago and this event is dramatically recorded in the deep sea and on land by sea-level change as well as in floral and faunal assemblages. This level forms the boundary between the Pleistocene/Holocene, although for all practical purpose we are living within an interglacial interval within the Pleistocene.

CHAPTER 9

PLIOCENE-PLEISTOCENE INTERCONTINENTAL CORRELATIONS

The age of the boundary between the Irvingtonian (Pleistocene) and Blancan (Plio-Pleistocene) land mammal ages has been a vexing question inasmuch as it involves the early part of the glacially influenced climates of North America and the philosophical and stratigraphical arguments over their correlation to world climate cycles and European glacial stratigraphy. Cooke (1972) reviewed the evidence available only a short while before this paper was written and came to the conclusion that the problem could not be resolved on the basis of information in hand: local faunas attributed to the Blancan included some which on paleoclimatological grounds seemed to have lived in the post-Nebraskan interglacial (Aftonian), yet K-Ar dates on Blancan levels extend so far back in time that the Nebraskan continental glaciation seemed to have been coeval with minor cold-climate periods in the Early Villafranchian. Moreover, the earliest Irvingtonian faunas were also found in Aftonian interglacial deposits, and showed some considerably more evolved features than the supposedly Aftonian Blancan assemblages. The validity of the K-Ar calibration, including as it did some controversial or uncertain correlations to the later Blancan faunas, was called into question by those who made a point of the evidence for multiple demi-glacial events earlier than the first major glaciation in North American mountainous regions, e.g. the 3-m.y. Deadman Pass Till in the Sierra Nevada (i. al. Birkeland et al., 1971; Cooke, 1972). At the same time the available evidence did not contradict others who compared the beginning of Nebraskan continental glaciation with the earliest ice-rafted debris and middle-latitude cold-climate foraminiferal faunas dated to approximately 3 m.y. in deep-sea cores (i. al. Beard and Lamb, 1968; Beard, 1969; Lamb, 1969; Savage and Curtis, 1970; Berggren, 1972b; Lamb and Beard, 1972).

By returning to the typical Nebraskan sequence of the upper Midwest, Hibbard (1972) appeared to have resolved the question in favor of a relatively young Nebraskan. Following the reappraisal of Pearlette ash-like beds by Izett et al. (1971), which suggested that the Kansan glaciation was bracketed between 0.6 and 1.2 m.y., Hibbard acknowledged the probable existence of pre-Nebraskan cold-climate cycles, as well as an erroneous report of a post- Nebraskan Mimomys in a Blancan assemblage, and concluded that the later Blancan “interglacial” was not the Aftonian but represented a pre-Nebraskan warm-climate period. The Irvingtonian/Blancan boundary is thus placed in the latter part of the Nebraskan Glacial Age.

160

The consensus of K-Ar dates, which when grouped show a relatively unambiguous separation between Irvingtonian and Blancan levels at ca. 1.5 m.y. (Birkeland et al., 1971) thus falls into place with the bracketing age limits of the Kansan estimated by Izett et al. (1971). Comparing the Kansan continental glaciation with the sequence of dated glaciations in the Sierra Nevada and Rockies, Cooke (1972) proposed a correlation between an 0.8-0.9 m.y. Kansan and the Menapian (?Gunz) cold-climate period in Europe. This would make the Biharian land mammal age of Europe entirely post-Kansan, although paleontological correlations in Europe and deep-sea data indicate a slightly different correlation (see above).

Recent geophysical studies are reasonably consistent with a 1.5-m.y. or slightly younger Irvingtonian/Blancan boundary. In an abstract, Boellstorff (1973) reports that fission-track ages indicate that the entire Nebraskan glacial sequence of eastern Nebraska is older than 1.2 m.y. This is not irreconcilable with the K-Ar consensus which indicates that the youngest Nebraskan should be ca. 1.4-1.3 m.y., especially in view of the experimental and stratigraphic uncertainties involved. Izett et al. (1972) report a minimum age for the top of the Blanco Formation (West Texas) at the locality from which the type Blancan assemblage comes. This is based on the presence, in red sandstones conformably (?) overlying the Blanco Formation, of an ash layer which the authors correlate to the reversely magnetized Guaje Pumice of New Mexico, dated at 1.4 m.y. At Cita Canyon, some 80 miles (130 km) north of Blanco Mountain, Johnston and Savage (1955) recorded a limited Equus-mummu thus Irvingtonian-like collection from just beneath “Pearlet- te” ash, which C.A. Repenning (written communication, 1973) is of the opinion may also be Guaje Pumice ash. This would agree very closely with an Irvingtonian/Blancan transition at ca. 1.5 m.y., and because this transition is apparently associated with the end of Nebraskan glaciation, and because the type Blancan fauna is a warm-climate association (Hibbard, 1972) then the upper part of the Blanco Formation and the overlying red beds should represent part or all of the Nebraskan climate period. *

The time of the Nebraskan glaciation is closely limited by these rather preliminary observations to the period 1.6 to 1.4-1.3 m.y. ago. As we have pointed out elsewhere there is nothing to distinguish one cold phase from another on paleoclimatological grounds alone and arguments as to the correlation of the Nebraskan glaciation in other parts of the world must eventually refer to the paleontology and dating of the Midwestern Pleistocene, including the evidence summarized above.

One of the characteristic differences between Blancan and Irvingtonian mammal faunas is the appearance of Mummuthus (Hibbard et al., 1965). Its absence from Blancan faunas up to as late as 1.5 m.y. ago is not in good agreement with the European calibration which indicates that Mammu thus was ubiquitous in Eurasian mammal faunas by middle Villafranchian time,

* See Appendix, note 7. __-

161

ca. 2.5 m.y. ago. According to Hibbard et al. (1965) and Repenning (1967) other taxa represented in Eurasian faunas as early as mid-Villafranchian (Equus s.s., Smilodon, Euceratherium, possibly Lepus) also make their first appearance in North America during the Aftonian interglacial in early Irving- tonian time, but this group is said to be accompanied by immigrants of Biharian or younger age (Microtus, Pitymys, Gulo, Bison). The apparently delayed prochoresis of the “Villafranchian” group could be explained in a number of ways (cf. Repenning, 1967) and might logically have t o do with the temporary opportunity offered by eustatic sea-level lowering during the first major continental glaciation of the Pleistocene, but the presence of the immigrant “Biharian” group during a 1.5-0.9-m.y. Aftonian interglacial is obviously inconsistent with the calibration which restricis ihe evolution of such an assemblage, in Europe at least, t o an age younger than the 0.9-m.y. Biharian/Villafranchian boundary. We are unwilling t o plunge into the depths of Asian and North American Pleistocene mammal paleontology, but it seems probable that much of the apparent biochronological inconsistency is linked to the former confusion over the Irvingtonian/Blancan transition. For instance the Cudahy fossil site with abundant Microtus and Pitymys (Hibbard, 1949) is now known to be Late Kansan-Early Yarmouth age (mid-Irvingtonian) overlain by “0-type” (0.6 m.y.) Pearlette-like ash, whereas the warm-climate Borchers fauna including only the North American endemic microtine Synaptomys (Hibbard, 1949) represents a Blancan level just above “B-type” (2.0 m.y.) Pearlette-like ash (Naeser e t al., 1973; not Hibbard, 1972); before the ash beds were distinguished from one another these faunal levels were thought t o be in the inverted order (Hibbard et al., 1965), and the Cudahy to be of early Irvingtonian age.

manuscript reporting on the paleomagnetic stratigraphy of the continental San Pedro Valley sequence, Arizona. Differing versions have been presented in meetings as this work progressed (e.g., N.M. Johnson et al., 1972) but the authors’ present conclusion is that reversely magnetized sediments of the Lower Gauss interval contain Late Blancan mammals, whereas Olduvai-C;ils& normally magnetized beds contain Early Irvingtonian mammals. In this interpretation the Irvingtonian Age begins about 1.9 1n.y. ago, which conflicts seriously with other evidence that this period, representing post-Nebraskan evolution, began not earlier than 1.5 m.y. During conversations with E.H. Lindsay, it became clear that in the San Pedro Valley, where most of the fossil fauna is small vertebrates, the beginning of Irvingtonian time was correlated to the first appearance of Lepus. Although this is common practice in small-mammal sequences, there is some evidence that this “datum” does not coincide with the first appearance of Mammuthus, the criterion which is used by large-mammal workers for the base Irvingtonian. Lepus is also known from the Borchers local fauna, dated ca. 2.0 m.y. ago (Izett et al., 1971) but Birkeland et al. (1971) record no occurrence of Mammuthus earlier than 1.5 m.y. despite the relatively abundant number of determinations on Late Blancan levels. Johnson and his co-workers also

N.M. Johnson, N.D. Opdyke, and E.H. Lindsay have sent us a copy of their

162

originally reported that the unnamed ash in the upper type Blancan had positive paleomagnetic sign (contra Izett et al., 1972) but E.H. Lindsay confirmed (personal communication, 1974) that subsequent work found that the type Blancan, including the unnamed ash, yielded only reversely magnetized samples, and was thus entirely older than the Olduvai-Gilsa Event. It now seems probable that the Olduvai-GilsA Event coincides roughly with a “transitional” Blancan/Irvingtonian period of evolution with Lepus but without Mammuthus.

The correlations of earlier Plio-Pleistocene mammal faunas between Eura- sia and North America appear to be more easily dealt with. The absence of Mirnomys from the Blancan faunas (Hibbard, 1972) is more a matter of provincial taxonomic custom than a real difference from the Eurasian rodent faunas of similar age since the Blancan microtids would probably be assigned to Mimomys if found in European sites (Michaux, 1971a, p. 152). In fact the ancestral Prornimomys is recognized both from the Hemphillian of North America and the Early-Middle Ruscinian of Europe, ca. 4.5 m.y. ago, and the Early Blancan microtines (Cosornys, Ophiomys, etc.) of ca. 3.5 m.y. ago have the Type I dental characters (Fejfar, 1964) on the coeval Late Ruscin- ian Mimornys species (Michaux, 1971a). On the other hand, hypselodonty or open-rootedness indicative of ever-growing cheek teeth such as characterizes most modern genera of microtines first develops in European Mirnornys (M. savini group of Fejfar, 1964, with Group I11 dentition) about 1.0 m.y. ago, just earlier than the beginning of the Biharian Age (cf. Kretzoi, 1965) whereas Hibbard (1972) notes that this condition is already present in cheek teeth of endemic North American microtines (Pliolemrnus, Synaptomys) in levels attributed to the Late Blancan (Sand Draw, ?Dixon local faunas) or even earlier (Bender l.f.), with an age at least as great as 2.0 m.y. when the Borch- ers 1.f. is included (Naeser et al., 1971).

Referring to the inter-calibrated continental glaciation sequence and mammalian biochronology , the argument for correlating European and North American continental glacial ages as shown in Fig. 14 is summarized as follows:

(1) The Nebraskan glaciation is earlier than any comparable lowland ice sheet in temperate Europe (Repenning, 1967; Cooke, 1972) but is not the time equivalent of the earliest evidence for glaciation or cold-climate maxima in montane and polar regions (Birkeland et al., 1971; Dalrymple, 1972; Poag, 1972). New and still largely untested calibration appears to limit the Nebras- kan glaciation event to an interval between approximately 1.6 m.y. (estimated upper age limit of warm-climate Blancan fauna and 1.4--1.3-m.y. (earliest unequivocal dates on Aftonian interglacial beds with Irvingtonian mammals). I t is certainly older than the 1.2-m.y. “S” type Pearlette ash (Izett et al., 1972). Aftonian-Early Irvingtonian local faunas containMarnrnuthus and other immigrants related to Villafranchian and younger faunas of Eurasia, but possibly not the exclusively post-Villafranchian (post-0.9 m.y .) elements as reported earlier.

(2) The Aftonian interglacial (early Irvingtonian age) is equivalent to the

163

Waalian climate period (late Villafranchian age) on the basis of geophysical dating of late Villafranchian faunas and North American glacial cycles which indicate that: (a) the later Villafranchian does not extend younger than 0.9 m.y. and ranges back to ca. 1.5 m.y.; fossils of this age are found in Waalian beds of The Netherlands and in East Anglian equivalents; (b) the Aftonian interglacial is limited by dates of approximately 1.4-1.3 m.y. (slightly younger than Irvingtonian/Blancan boundary) and between 1.2 and 0.6 m.y. (maximum age limits of Kansan glaciation); and (c) the Jaramillo Normal Event, midpoint 0.91 m.y., is probably represented both in the uppermost Waalian beds of The Netherlands (Van Montfrans, 1971a) and in Aftonian beds in North America (J. Kukla, quoted by E.V. Lindsay, personal communication, 1973), which would suggest a very close coincidence between the end of the Aftonian interglacial and the end of the Waalian warm- climate phase especially in view of the probable age of the succeeding Kansan and Menapian cold-climate phases (below).

(3) The Kansan glaciation is equivalent to the Menapian cold-climate phase, and probably the Gunz alpine glaciation, on the basis of Pearlette- like/and Bischop ash dates limiting the typical Kansan deposits to the 1.2-0.6-m.y. interval, and evidence for a major pre-Brunhes Normal Epoch glaciation (Sherwin Till) in the North American mountains during approximately the same time (Birkeland et al., 1971; Dalrymple, 1972; Richmond and Obradovich, 1972). The age of the Sherwin montane glaciation could correlate validly to both the Kansan continental glaciation within the given limits and the Gunz-Menapian cold phase in Europe, at or near the 0.9-m.y. Biharian/Villafranchian boundary. The presence of the Jaramillo Normal Event in pre-Kansan deposits, if confirmed, would greatly strengthen this correlation. As Ruddiman (1971) and Briskin and Berggren (1974) noted, the Jaramillo Event in deep-sea cores is linked with the beginning of the first prolonged cold-climate phase in what Cooke (1972) called “a very suitable position” to correlate with the Kansan and the Giinz. Deep-sea evidence (see Chapter 8 on chronology of climate events) indicates however that the major part of the Kansan-Gunz glaciation dates from ca. 0.75 m.y., in mid-Bihar- ian time, and Kukla and Opdyke (1972) place the 0.7-m.y. Brunhes/Matuya- ma boundary within the Kansan glacial deposits.

(4) The Yarrnouthian interglacial (Mid or ?Late Irvingtonian mammal age) covers the same 0.7-0.5-m.y. time span as the “Cromerian” or mid-Taxan- drian interglacial (Late Biharian mammal age) in Europe. This is based on the recently improved geophysical calibration and deep-sea evidence which centers the Kansan glaciation at 0.75 m.y. (lower limiting age 0.9 m.y.) and the Illinoian glaciation at 0.5 m.y. (Cooke, 1972). The Brunhes/Matuyama paleomagnetic boundary ca. 0.7 m.y., is fairly well identified with the medial Biharian Age in Europe, and this and other dating indicates that the Mindel- Elsterian continental glaciation, which terminates the “Cromerian” interglacial, is also centered within the limits 0.6-0.4 m.y. A direct comparison of microtine rodent taxa and other mammals of Europe and North America,

164

of Biharian and late Irvingtonian age respectively, is supported by the revised 0.6-m.y. age of the Cudahy local fauna. Evidence that the cold climate of approximately 0.5 m.y. ago was one of the most intense world temperature declines of the Pleistocene is noted from the Mindel-Elsterian of Europe (Van Montfrans, 1971a; Ambrosetti et al., 1972), the “Riss” of Siberia (Vangengeim and Sher, 1970) post-dating a Biharian fauna, and in deep-sea cores (Cooke, 1972; Briskin and Berggren, 1974) just above the Brunhes/ Matuyama boundary. Cooke (1972) also notes that the first glaciation on Mt. Kilimanjaro is closely dated to this same interval, and the wide extent of Illinoian glacial sediments and soils in North America is well documented.

CHAPTER 10

SUMMARY

We have treated a wide-ranging variety of topics in this review. Perhaps the best way to summarize the basic information and conclusions we have reached here is to present an itemized account which follows the chronology of the study itself.

(1) The integration of modern paleomagnetic, radiometric, and biostratigraphic studies has provided an accurate geochronological framework for the past ten million years -the Late Neogene.

(2) Marine zones based on calcareous and siliceous planktonic organisms are recognized from the Subarctic to Subantarctic regions (60"N to 60"s) and their correlation to the paleomagnetic time-scale is feasible in some detail over the past 5 m.y. Zones in sediments older than 5 m.y. are primarily developed in tropical-temperate regions, and correlation with high-latitude regions awaits further research.

(3) Late Neogene chronostratigraphic boundaries can be recognized by reinforcing multiple paleontologic criteria and the following ages are assigned to them:

Pliocene/Pleistocene (base Calabrian): ca. 1.6-1.8 m.y. Zanclian/Piacenzian (Early/Late Pliocene): ca. 3.3 m.y. Miocene/Pliocene (Messinian/Zanclian): ca. 5.0 m.y. Tortonian/Messinian: ca. 6.6 m.y. Serravallian/Tortonian (Middle/Late Miocene): ca. 10.5-10.7 m.y.

(4) Discrepancies in the literature regarding the age and position of the Miocene/Pliocene and Pliocene/Pleistocene boundaries, as well as other boundaries lower in the chronostratigraphic hierarchy, are discussed and shown to be due primarily to different interpretations of paleontologic criteria, incorrect biostratigraphic correlation and calibration to inaccurate radiometric dates.

to marine stages is possible with a relatively high degree of accuracy. A sequence of major events in the marine plankton and continental mammals is shown for the past 15 m.y.

(6) Late Neogene earth history must be viewed within the framework of a cooling climatic trend which accelerated during the Pleistocene. Although climatic changes characterize the Pleistocene history of the globe, they cannot be used in delineating the age of its lower boundary. Polar ice built

(5) Correlation and calibration of Late Neogene continental mammal ages

166

up great thicknesses in Antarctica much earlier than in the Arctic region. Initial ice-cap formation in Antarctica may be older than 40 m.y., and the presence of thick ice sheets is indicated throughout the Neogene. A great. advance of grounded shelf ice is linked to Late Miocene Antarctic cooling and the recession of this ice may be correlated to initial Pliocene inundation of the desiccated Mediterranean Basin. Floating ice, calved from sea-level glaciers first appeared in the North Atlantic and the North Pacific about 3.0 m.y. ago. The “long school” of glacial chronology which puts the continental glacial history of Europe and North America in a chronological framework in excess of 2 m.y. is based on false assumptions linking evidence for initial cooling and montane glacial advances with the initiation of major temperate-region continental ice sheets. The evidence of opinion is that Neogene cooling was progressive, with relatively rapid decrements of lower temperature approximately 1.5-1.2 m.y. ago and again from 0.9 to 0.5 m.y. There is no well-known calibrated evidence to link the base of the Calabrian (beginning of the Pleistocene) about 1.8 m.y. ago with a marked cooling event, and all indications point to the first temperate glaciation at a significantly younger age with the next general lowering of the average temperature.

(7) A general correlation between North American and European paleoclimatological sequences, based on radiometric and paleomagnetic calibration of mammalian biostratigraphy, appears to be corroborated by the deep- sea paleoclimatological record. A salient feature of this correlation is support for earlier research which suggested, contrary to conventional views, that the first two North American glaciations (Nebraskan and Kansan) are represented in Central European climatological history by cold intervals without major lowland glaciation (Donau, Gunz). Details of this correlation bring out the fact that far too many former correlations were naive in supposing that glaciation was independent of regional variations in topography, meteorolo- gy or latitude, or that the influence of changing climate on one set of evidence might show different apparent intensities than on another coeval set. In consequence of this it is emphasized that the first continental lowland glaciation in Central Europe, the Mindel, is equivalent to the most intense glacial conditions of the Pleistocene and to the Illinoian glaciation of North America. This interval of maximum cold climate extended from ca. 0.6 to 0.45 m.y. in deep-sea cores and is in agreement with dates on the Mindel and Illinoian continental deposits.

(8) The principle that Northern Hemisphere climate cycles were, in general, synchronous on both continental masses, is upheld, but the assumption that correlated climate cycles had comparable absolute temperature levels is shown to be insufficient.

tal rock record and its calibration to an ordinal time-scale provides the appropriate background for interpretive studies in the historical geological evolution of our earth - in this case over the past 10-15 million years.

(9) The integration of biostratigraphic data from the marine and continen-

APPENDIX

INTRODUCTION

We have covered a broad spectrum of data in this study. New information is now appearing at such a rapid rate that it is virtually impossible to keep pace. We have chosen to utilize this Appendix as a format for presenting a summary of new data that has appeared since the completion of this manuscript.

1. PALEOMAGNETIC CORRELATIONS IN DEEP-SEA CORES BELOW EPOCH 5

Recent information warrants some modifications to the Middle Miocene part

One of the basic calibration points in the marine time-scale in these figures of the time-scale shown in Fig. 1 and 11.

is the correlation of sea-floor Magnetic Anomaly 5 to Geomagnetic Polarity Epoch 9 at ca. 9.5 m.y. as suggested in a presentation by M. Dreyfus and W.B.F. Ryan at the 24th International Geological Congress in Montreal, September 1973. In the preceding text, Berggren attributed to Dreyfus and Ryan a further calibration of the Epoch 10/11 and Epoch 11/12 at 11.6 and 12.4 m.y. respectively, resulting in a relatively long Epoch 10 and a relatively short Epoch 11 within the time span of 10.0-12.4 m.y., but this has been found to be a mistaken interpretation of their presented data. No age values were suggested for paleomagnetic epoch boundaries below the Epoch 9/10 boundary at 10 m.y., and in fact the paleomagnetic record shows that Epoch 10 is the shorter and Epoch 11 the longer of the two.

Two separate investigations of the Neogene paleomagnetic time-scale have since come to our attention which deal with this erroneously interpreted period. Theyer and Hammond (1973, 1974) * report on the correlation of radiolarian zones and paleomagnetic stratigraphy in the Central Pacific, and Opdyke et al. (1974) deal with the relationship of planktonic zones and paleomagnetic stratigraphy in the same region.

Theyer and Hammond (1973) accept the correlation of Anomaly 5 and Epoch 9 at 9.5 m.y., and an age of 15 m.y. for the Orbulina Datum. Using

* For additional references see pp. 215-216.

168

local sedimentation rates in the Central Pacific (0.3-0.5 cm/103 year), the ages of Epoch boundaries in cores are interpolated from these calibration points and the standard Equatorial Pacific radiolarian zonation of Riedel and Sanfilippo (1970, 1971) is then correlated to the calibrated paleomagnetic stratigraphy.

Opdyke et al. (1974) also accept the Dreyfus and Ryan correlation of Magnetic Anomaly 5 and Paleomagnetic Epoch 9. They have estimated the age of the paleomagnetic boundaries by measuring the cumulative length of their cores with the paleomagnetic stratigraphy plotted against depth. They then enter the age of observed planktonic foraminifera1 datum events from Berggren (1972a) as well as their own and previously publikhed dates on paleomagnetic boundaries (e.g., the Gilbert/Epoch 5 boundary at 5.1 m.y., the Epoch 8/9 boundary at 9 m.y., and the Epoch 9/10 boundary at 10 m.y.) and then connect the points. Their data form a straight line for the interval T = 5-15 m.y. which would seem to indicate a relatively constant rate of sediment accumulation in the area of study. The boundary ages of Paleomagnetic Epochs 9-15 are virtually the same as that of Theyer and Hammond (1974).

Drawing on these new conclusions the following data are important for the discussion of a revised Miocene calibration presented below :

(a) The Serravallian/Tortonian (= Lower/Middle Miocene) boundary occurs within Zones N15 and "9. The Cannartus peterssoni/Ommatartus antepenultimus radiolarian zonal boundary occurs within the interval of overlap of these two calcareous planktonic zones and probably lies at or very near a level equivalent to the base of the Tortonian. This boundary is near the Epoch 10/11 boundary (Theyer and Hammond, 1974).

(b) The Cannartus petterssoni Zone is confined to the limits of Pdeoriagnetic Polarity Epoch 11 (Theyer and Hammond, 1974).

(c) The planktonic foraminifera1 zone N14/15 boundary occurs within the interval of the Cannartus petterssoni Zone (see ahove). The N12/13 boundary occurs below the base of the Cannartus pet te ssoni Zone (i.e., within the upper part of the Dorcadospyris alata Zone as currently used; this was formerly called the Cannartus laticonus Zone; see Riedel and Sanfilippo, 1971, p. 1574-1579; Hays et al., 1972a, p. 88, Martini, 1971, p. 750). The relative position of the N13/14 (G. nepenthes Datum) is somewhat ambiguous. Bronnimann et al. (1971, p. 1742) indicate that it also falls within the interval of the Cannartus petterssoni Zone. Hays et al. (1972a, p. 66, 74) indicate, however, that the initial appearance of G. nepenthes occurs below the base of the Cannartus petterssoni Zone, and within the uppm part of the Cannartus laticonus Zone (i.e., within the upper part of the Dorcadospyris alata Zone as presently used; we have also observed this relationship at D.S.D.P. Site 214 in the Indian Ocean). In fact G. nepenthes is shown to overlap stratigraphically with Globorotalia fohsi lobata for several meters at the top of the C. laticonus Zone (Hays et al., 1972a, p. 66). We have observed a similar overlap in the course of our work (Berggren and

169

Amdurer, 1974) at D.S.D.P. Site 11 in the North Atlantic. The weight of evidence would appear to indicate that the Globigerina nepenthes Datum is older than the Dorcadospyris alata/Cannartus petterssoni zonal boundary.

(d) An important calibration point between the biostratigraphy and the paleomagnetic stratigraphy is afforded by Burckle in Opdyke et al. (1974, and personal communication to W.A.B., 1974). This is the correlation of the sediment adjacent to the ashes dated at 11.4 t 0.6 m.y. and 12.3 t 0.4 m.y. (Dymond, 1966) in the Experimental Mohole to Paleomag- netic Epoch 12 and/or 13 based on diatom biostratigraphy. This level has been dated by calcareous nannofossils as being within the Discoaster kugleri ("7) Zone and correlated with the Globorotalia fohsi robusta Zone (= N12--13) of Blow (1969). Thus calcareous nannoplankton Zone "7 and the planktonic foraminifera1 Zone N12/N13 boundary should occur at or near the base of Paleomagnetic Epoch 12. This agrees well with data summarized in points b and c above that the N12/13 boundary antedates both the C. petterssoni Zone and Paleomagnetic Epoch 11.

(e) The Far East letter stage Tfl-2 /Tf, boundary (which is equivalent to the N12/13 boundary) was dated at about 12-12.5 m.y. by Page and McDougall (1970). This should provide a further point of calibration for dating the paleomagnetic epoch sequence at or near the base of Epoch 1 2 (see point d above).

A consideration of the above leads us to the following conclusions : (1) The paleomagnetic time-scales of Theyer and Hammond (1974) and

Opdyke et al. (1974) are essentially the same. We have adopted their scale here as a standard of reference for Miocene biochronologic calibration. Within the Middle Miocene particularly it can be demonstrated that the calibration to it of several biostratigraphic events agree well with known geologic data.

(2) The Serrauallian/Tortonian boundary lies within Zones N15 and NN9 and approximately at the Cannartus petterssoni/Ommatartus antepenultimus boundary and coincides with the Paleomagnetic Epoch 10111 boundary. An age of 10.5-10.7 m.y. is estimated for this level here.

(3) The GZobigerina nepenthes Datum (N13/14 boundary) is approximately correlative with the NN7/8 boundary, lies within the upper part of the Dorcadospyris alata Zone and within Paleomagnetic Epoch 12. An estimated age of 11.7 m.y. for the G. Nepenthes Datum is made here.

In a recent study on the biostratigraphic basis of the Neogene time-scale (Van Couvering and Berggren, 1974) we have taken into consideration the new data summarized above in preparing an updated time-scale for the Neo- gene. In that paper, Late Oligocene and Miocene interhemispherical calcareous and siliceous planktonic zonal correlations and calibrations to the paleomagnetic time-scale were shown and this is reproduced here as Fig. 15. The calibration of the radiolarian and diatom zones to the paleomagnetic time- scale follows the investigations of Theyer and Hammond (1973, 1974) and Burckle (1972). The inter-zone correlations are based upon an evaluation

170

Fig. 15. Updated time-scale for Late Oligocene and Miocene.

171

of the data and/or correlations presented in recent investigations on siliceous plankton by Koizumi (1968, 1973), Moore (1971, 1972), Sanfillippo (1971), Martini (1972), Sanfillippo et al. (1973) Schrader (1973a, b) and on calcareous plankton by Berggren (1972b), Martini (1971), Jenkins (1971), Jenkins and Orr (1972), Kennett (1973),Parker (1973a,b), and Berggren and Amdurer (1974).

It should be pointed out, perhaps, that the Middle Miocene part of the time-scale in Fig. 15 is similar to that in Fig. 11 with the following modifications: the Orbulina Datum is lowered from 14 m.y. to 15 m.y. and the numerical ages of the zonal boundaries are slightly adjusted according to the more accurate correlation between the radiolarian zonal boundaries and the polarity epochs made by Theyer and Hammond (1974) and the calibration of the N12/13 boundary to the paleomagnetic 12/13 boundary at about 12.2 m.y. as warranted by the data of Burckle in Opdyke et al. (1974). Finally, the erroneous time-span for Paleomagnetic Epochs 10 and 11 has been corrected.

2. THE ANDALUSIAN STAGE

One of us (W.A.B.) is currently investigating the foraminifera1 biostratigraphy and paleoecology of the Andalusian stratotype section. It is important to bear in mind that the concept of the Andalusian Stage was modified by Perconig (1971) and the base was extended approximately 250 m lower in the stratotype section. The presence of Globorotalia margaritae “throughout the Andalusian” referred to in the text refers to its supposed occurrence in the beds immediately subjacent (down to about 50 m) to the Caliza Tosca. Our investigations suggest that the form referred to as Globorotalia margaritae by Perconig (1968), as well as by Verdenius (1970) in the Ecija Formation, Carmona Section, western Guadalquivir Basin, may be Globorotalia cibaoensis Bermudez, which we believe is the ancestral form of G. margaritae. One of the more important results which has been obtained from this study of the Andalusian stratotype section is that the benthonic foraminifera indicate that a single eustatic fall of sea level on the order of 50-70 m occurred at the boundary between the sandy marls (below) and the Caliza Tosca (above). The Caliza Tosca is a shallow water (- 20-30 m water depth) calcarenite, the base of which (middle zone N17) can be correlated with the base of the upper Messinian Stage in the Mediterranean. The facies developed in these two stages are probably a reflection of this sudden eustatic sea-level fall, which is, in turn, apparently linked with a rapid expansion of the Antarctic ice-sheet to its maximum recorded extent.

3. LEG 13 PALEOMAGNETIC STUDIES

Kennett and Watkins (1974) have recently criticized and rejected the validity of the paleomagnetic interpretations on D.S.D.P. Leg 13.

172

4. THE ISCHIA DATE

The Ischia date is now considered unreliable (personal communication from M.B. Cita to J.A. Van Couvering, April, 1974).

5. CALABRIAN STRATOTYPE

The field-section thickness reported by Bayliss (1969) were incorrectly measured (personal communication from D.D. Bayliss to W.A. Berggren, September 1972). The correct values are presented here in parenthesis and will also appear in a future paper by Dr. Bayliss.

6. THE HIPPARION DATUM

Current investigations are tending to change the definition of the Hipparion Datum slightly. The Ngorora Hipparion, ca. 11 m.y. old, has been identified as H. primigenium by D.A. Hooijer (personal communication, 1974), and work by M.T. Alberdi in Spain (written communication, 1973; see also Col- Pa, 21:7-8, 1972) confirms the conclusion of Forstkn (1968, 1972) that the initial population of Hipparion in Eurasia and North Africa consisted of a single, if variable, species, H. primigenium v. Meyer. Local fossil populations appear t o have been unmercifully split because of essentially meaning- less statistical variations (P.Y. Sondaar, personal communication, 1974). So far, H. nagriensis (Hussain, 1971) of the Siwaliks had not been included within H. primigenium, but with this exception we can conclude tentatively that the species spread initially from Siberia to East Africa and Spain without significant adaptive changes. This implies that there was little to hinder its expansion except for reproductive rate and territorial inhibitions, and that the rate of expansion could have been comparable t o that of rabbits in Australia or English sparrows in North America, i.e., 10 km/year a t a minimum.

N12 should be at approximately 12.5 m.y. (see also this Appendix), but that somewhat uncertain correlations suggest N12 is also earlier in time than the Hipparion Datum. K-Ar dates to be published by the German team working in Anatolia (cf. Tobien, 1970b) indicate that the Hipparion Datum there is younger than 11.9 m.y., and the 12.5-m.y. Howenegg date is being re-investigated because of this discrepancy. S.T. Hussain and E. Delson have suggested t o J.A.V.C. (written communication, 1972) that a 12.5-m.y. first appearance of Hipparion in Eurasia is inconsistent with Clarendonian dates in North America which suggest a younger evolution of the genus. Evident- ly there is growing reason to doubt the 12.5-m.y. age given t o the Hipparion Datum, and it may move up t o 1 m.y. younger t o conform with the Anatolian and East African evidence, and t o marine .and intercontinental correlations as well.

We have noted that the calibration of marine sequences indicates that Zone

173

7. APPEARANCE OF ELEPHAS IN EURASIA

After our text was prepared, V.J. Maglio published a monograph on the origin and evolution of the Elephantidae (Am. Philos. SOC., Trans., 6 3 (3): 149 pp, 1973) in which he concluded that the first representative of the genus Elephas in Eurasia (exclusive of India and the Far East), E. namadicus (= Mammuthus antiquus in most modern works) was an immigrant taxon and did not descend from Mammuthus meridionalis. The meridionalis lineage continues through M. armenacius (= M. trogontherii) and M . primigenius in high latitude faunas of Eurasia with almost imperceptible gradations; the first appearance of Elephas and the development of M. armenacius are nearly simultaneous and help to characterize the radical re-organization of Eurasian faunas at the beginning of the Biharian mammal age.

REFERENCES

Aguirre, E., 1972. Utilization of proboscideans in Pleistocene stratigraphy. In: M.N. Alek- seev, E.A. Vangengeim, K.V. Nikiforova and I.M. Khoreva (Editors), Int. Colloq. Prob- lem “The Boundary Between Neogene and Quaternary”, 111. Moscow, pp.3-13.

Akers, W.H., 1965. Pliocene-Pleistocene boundary, northern Gulf of Mexico. Science, 149: 741-742.

Alekseev, M.N., 1961. Stratigraphy of continental Neogene and Quaternary deposits of the Vilyuy Depression and the lower Lena Valley. Trans. Geol. Inst. Acad. Sci. U.S.S.R., 51 : 117 pp.

the southeastern United States. Geol. SOC. Am., Spec. Pap., 87 : 3.

Palaeoclimatol., Palaeoecol., 5 : 87-94.

406-411.

Alt, D., 1965. Miocene age and paleoclimatic significance of the Citronelle Formation of

Alt, D., 1968. Pattern of post-Miocene eustatic fluctuation of sea-level. Palaeogeogr.,

Alt, D. and Brooks, H.K., 1965. Age of the Florida marine terraces. J. Geol., 73:

Ambrosetti, P., 1967. Cromerian fauna of the Rome area. Quaternaria, 9: 267-284. Ambrosetti, P., Azzaroli, A., Bonadonna, F.P. and Follieri, M., 1972. A scheme of Pleisto-

cene chronology for the Tyrrhenian side of central Italy. Boll. SOC. Geol. Ital., 91: 169-184.

American Commission on Stratigraphic Nomenclature, 1961. Code of stratigraphic nomen

Andrusov, N., 1899. Environs de Kertch. Guide Excurs. VIIe Congr. GBol. Int., St.

Anglada, R., 1971a. Sur la position du datum

clature. Bull. Am. Assoc. Pet. Geol., 45(5): 645-665.

Petersburg, 1897, 30: 16 pp.

N4 (Blow, 1967) et la limite oligo-miocsne en Mediterranee. C.R. Acad. Sci. Paris, 272D: 1067-1070.

Anglada, R., 1971b. Sur la limite Aquitanien-Burdigalien, sa place dans 1’Qchelle des For- aminifdres planctoniques et sa signification dans le Sud-Est de la France. C.R. Acad. Sci. Paris, 272D: 1948-1951.

Aoki, N., 1963. Pliocene and Pleistocene foraminifera from along the Yoro River, Boso Peninsula. Sci. Rep. Tokyo Kyoiku Daigaku, Sect. C., 8(78): 203-227.

Arambourg, C., 1970. Les corr6lations palContologiques et chronoligiques entre le Pleisto- chne inf6rieur de 1’Europe et celui de I’Afrique. Soc. G601. Fr., Bull., Se‘r. 7, l l (1969) :

Globigerinoides (Foraminiferida), la zone

106-115.

(1947-1948), 5(1): 1-89.

Japan. K. Ned. Akad. Wet., Proc., Ser. B, 65(1): 1-16.

ifera from Japan. Sci. Rep. Tohoku Univ., Ser. 2 (Geol.), 37(1): 1-14.

sequence in northeastern Japan. In: P. Bronnimann and H.H. Renz (Editors), Proc. Int. Conf. Planktonic Microfossils, l s t , Geneva, 1967, I. E.J. Brill, Leiden, pp. 14-25.

Asano, K., Takayanagi, Y. and Takayama, T., 1974. Late Neogene epoch boundaries in

Arrhenius, G., 1952. Sediment cores from the East Pacific. Rep. Swed. Deep Sea Exp.

Asano, K., 1962. Faunal change of planktonic foraminifera through the Neogene of

Asano, K. and Takayanagi, Y., 1965. Stratigraphic significance of the planktonic foramin-

Asano, K., Ingle Jr., J.C. and Takayanagi, Y., 1969. Neogene planktonic foraminifera1

176

Japan. In: T. Saito and L. Burckle (Editors), Late Neogene Epoch Boundaries. Micro- palaeontology Press, Am. Mus. Nat. History, New York, N.Y. (in press).

Azzaroli, A., 1963. Rinoceronti pliocenici del Valdarno Inferiore. Palaeontogr. Ital., 57 : 11-20.

Azzaroli, A,, 1965. The two Villafranchian horses of the Upper Valdarno. Palaeontogr., Ital., 59: 1-12.

Azzaroli, A., 1970. Villafranchian correlations based on large mammals. G. Geol., 35(2):

Azzaroli, A. and Ambrosetti, P., 1970. Late Villafranchian and Early Mid-Pleistocene

Azzaroli, A. and Berzi, A., 1970. On an Upper Villafranchian fauna at Imola, northern

11 1-1 31.

faunas in Italy. Palaeogeogr., Palaeoclimatol., Palaeoecol., 8: 107-111.

Italy, and its correlation with the marine sequence of the Po Plain. Palaeontogr. Ital., 66: 1-12.

Bagdasarjan, G.P., Slavik, J. and Vass, D., 1971. Chronostratigraficky a biostratigraficky vek niektorych vyznamnych neovulkanitov Vychodneho Slovenska. Geol. Prace, Zpravy, 55: 87-96.

Bagdasarjan, G.P., Bajanik, S. and Vass, D., 1973. Age radiometrique du volcanisme nCo- g&ne dans le Nord Tunisien. Travaux de ge‘ologie tunisienne No. 8. Notes Serv. GBol. Tunis, 40: 79-93.

Balasundaram, M.A. and Sastry, M.V.A., 1972. Plio-Pleistocene boundary in sediments of the Indian sub-continent. In: M.N. Alekseev, E.A. Vangengeim, K.V. Nikiforova and I.M. Khoreva (Editors), Int. Colloq. Problem “The Boundary Between Neogene and Quaternary”, 111. Moscow, pp. 14-45.

Doc. Lab. G6ol. Univ. Lyon, (hors series) 1971: 201-239.

bigerina pachyderma (Ehrenberg). Geol. Soc. Am. Bull., 70(12-2): 1708.

bigerina pachyderma (Ehrenberg). J. Paleontol., 34(4): 671-681.

Rep. Tohoku Univ., Ser. 2 (Geol.), Spec. Vol., 4: 1-8.

140(3574): 1402-1403.

Tertiary stratigraphy of deep-sea sediments. Science, 142( 3597): 1290-1292.

in Philippines. Bull. Am. Assoc. Pet. Geol., 47(9): 1733-1745.

Ballesio, R., 1971. Le Pliockne rhodanien. In: Ve Congress du Neogkne Mediteraneen, I.

Bandy, O.L., 1959. The geologic significance of coiling ratios in the foraminifer Glo-

Bandy, O.L., 1960a. The geologic significance of coiling ratios in the foraminifer Glo-

Bandy, O.L., 1960b. Planktonic foraminiferal criteria for paleoclimatic zonation. Sci.

Bandy, O.L., 1963a, Aquitanian planktonic Foraminifera from Erben Guyot. Science,

Bandy, O.L., 196313. Miocene-Pliocene boundary in the Philippines as related to Late

Bandy, O.L., 1963c. Cenozoic planktonic foraminiferal zonation and basinal development

Bandy, O.L., 1964. Cenozoic planktonic foraminiferal zonation. Micropaleontology,

Bandy, O.L., 1966. Faunal evidence of Miocene-to-Recent paleoclimatology in Antarctic.

Bandy, O.L., 1967a. Foraminifera1 definition of the boundaries of the Pleistocene in

l O ( 1 ) : 1-17.

Bull. Am. Assoc. Pet. Geol., 50: 643-644 (abstract).

southern California, U.S.A. In: M. Sears (Editor), Progress in Oceanography, 4. Per- gamon, Oxford, pp. 27-49.

Bandy, O.L., 1967b. Problems of Tertiary foraminiferal and radiolarian zonation, Cir- cum-Pacific area. In: K. Hatai (Editor), Proc. Pac. Sci. Congr., l l t h , Tokyo. Sasaki, Sendai, pp. 95-102.

Bandy, O.L., 1968a. Paleoclimatology and Neogene planktonic foraminiferal zonation. G. Geol., Ser. 2, 35(2): 277-290.

Bandy, O.L., 1968b. Cycles in Neogene paleoceanography and eustatic changes. Palaeo- geogr., Palaeoclimatol., Palaeoecol., 5: 63-75.

Bandy, O.L., 1969. Relationships of Neogene planktonic foraminifera to paleoceanography and correlation. In: P. Bronnimann and H.H. Renz (Editors), Proc. Int. Conf. Planktonic Microfossils, l s t , Geneva, 1967, I. E.J. Brill, Leiden, pp. 46-57.

177

Bandy, O.L., 1970a. Introduction to radiometric dating and paleontologic zonation. Geol. SOC. Am., Spec. Pap., 124: 1-6.

Bandy, O.L., 1970b. Upper Cretaceous Cenozoic paleobathymetric cycles, eastern Pana- ma and northern Colombia. Trans. Gulf Coast Assoc. Geol. SOC., 19 : 181-193.

Bandy, O.L., 1971. Recognition of Upper Miocene Neogene Zone N18, Experimental Mohole, Guadelupe Site. Nature, 233: 476-478.

Bandy, O.L., 1972a. The Plio-Pleistocene boundary, Europe and California and the paleomagnetic scale. Int. Colloq. Problem “The Boundary Between Neogene and Quater- nary”, I. Moscow, 1 : 15-62.

Bandy, O.L., 197213. Neogene planktonic foraminifera1 zones, California, and some geologic implications. Palaeogeogr., Palaeoclimatol., Palaeoecol., 12: 131-150.

Bandy, O.L., 1972c. A review of the calibration of deep-sea cores based upon species variation, productivity, and O/’ 8 O ratios of planktonic foraminifera - including sedimentation rates and climatic inferences. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Acad. Press, Edinburgh, pp. 37-62.

Bandy, O.L., 1972d. Time-transgressive aspects of the Late Neogene Wheelerian-Hallian boundary, Southern California. Rev. Micropaleontol., 14( 5): 73-77.

Bandy, O.L. and Casey, R.E., 1969a. Major Late Cenozoic planktonic datum planes, Ant- arctica to the tropics. Antarct. J. U.S., 4: 170-171.

Bandy, O.L. and Casey, K.E., 1969b. Epoch boundaries and paleotemperature cycles of the Late Cenozoic. Geol. SOC. Am. Abstr. Progr., 1969, pt. 7, pp. 252-254.

Bandy, O.L. and Ingle Jr., J.C., 1970. Neogene planktonic events and radiometric scale, California. In: O.L. Bandy (Editor), Paleontologic Zonation and Radiometric Dating. Geol. SOC. Am., Spec. Pap., 124: 133-174.

Bandy, O.L. and Wade, M.E., 1967. Miocene-Pliocene-Pleistocene boundaries in deepwater environments. In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 51-66.

Bandy, O.L. and Wilcoxon, J.A., 1970. The Pliocene-Pleistocene boundary, Italy and California. Geol. SOC. Am. Bull., 81: 2939-2948.

Bandy, O.L., Ingle Jr., J.C. and Frerichs, W.E., 1967. Isomorphism in Sphaeroidinella and Sphaeroidinellopsis. Micropaleontology, 13(4): 483-488.

Bandy, O.L., Butler, R.C. and Wright, R.C., 1969a. Alaskan Upper Miocene marine glacial deposits and the Turborotalia pachyderma datum plane. Science, 166: 607-609.

Bandy, O.L., Morin, R.W. and Wright, R.C., 196913. Definition of the Catapsydrax stain- forthi Zone in the Saucesian Stage, California. Nature, 222( 5192): 468-469.

Bandy, O.L., Vincent, E. and Wright, R.C., 1969c. Chronologic relationships of orbulines to the Globorotalia fohsi lineage. Rev. Esp. Micropaleontol., 1( 2): 131-145.

Bandy, O.L., Casey, R.E. and Wright, R.C., 1971a. Late Neogene planktonic zonation, magnetic reversals, and radiometric dates, Antarctic to the Tropics. Am. Geophys. Union, Antarct. Res. Ser. (Antarctic Oceanology, l), 15: 1-26.

Bandy, O.L., Ingle Jr., J.C. and Wright, R.C., 1971b. Globorotaloides suteri Bolli subsp. relizensis n. subsp. J. Foramin. Res., l (1) : 15-16.

Bandy, O.L., Casey, R.E. and Theyer, F., 1972. Planktonic events, upper Gilbert-Gauss magnetic epochs. Geol. SOC. Am., Abstr. Progr., 4(7): 442.

Banner, F.T. and Eames, F.E., 1966. Recent progress in world-wide Tertiary stratigraphical correlation. Earth-Sci. Rev., 2: 157-179.

Banner, F.T. and Blow, W.H., 1959. The classification and stratigraphical distribution of the Globigerinaceae. Paleontology, 2: 1-27.

Banner, F.T. and Blow, W.H., 1960. Some primary types of species belonging to the SU-

per-family Globigerinaceae. Cushman Found. Foramin. Res. Contrib., 11 : 1-41. Banner, F.T. and Blow, W.H., 1965a. Two new taxa of the Globorotaliinae (Globigeri-

naceae, Foraminifera) assisting determination of the late Miocene/middle Miocene boundary. Nature, 207: 1351-1354.

178

Banner, F.T. and Blow, W.H., 1965b. Progress in the planktonic foraminiferal biostratig-

Banner, F.T. and Blow, W.H., 1967. The origin, evolution and taxonomy of the

Barbieri, F., 1952. I1 Pliocene di Val Recchio (Parma). Boll. SOC. Geol. Ital., 74(1):

Barbieri, F., 1967. The Foraminifera in the Pliocene section VernascaXastell’Arquato including the “Piacenzian stratotype” (Piacenza Province). Mem. SOC. Ital. Sci. Nat.,

Barbieri, F. and Medioli, G., 1964a. Nota preliminare sullo studio micropaleontologico della serie pliocenica Vernasca4astell’Arquato (Piacenza). Boll. SOC. Geol. Ital.,

Barbieri, F. and Medioli, F., 1964b. Significato paleontologico di alcuni generi di For-

raphy of the Neogene. Nature, 208: 1164-1166.

foraminiferal genus Pulleniatina Cushman, 1927. Micropaleontology, 13 : 133-162.

17 1-1 79.

15 : 147-163.

83( 1): 207-212.

aminiferi nella serie pliocenica Vernasca-Castell’Arquato. Ateneo Parmense, 3 5( Suppl. ) : 8-3 4.

Barbieri, F. and Petrucci, F., 1967. La serie stratigraphique du Messinien au Calabrien dans la valle‘e du T. Crostolo (Reggio Emilia-Italie Sept). Mem. SOC. Ital. Sci. Nat.,

Barbot de Marny, N., 1866a. Untersuchungen iiber den Charakter der osterreichischen

Barbot de Marny, N., 1866b. Uber die jungeren Ablagerungen des siidliches Russlands.

Baumann, P. and Roth, P., 1969. Zonierung des ObereozPns und OligozEns des Monte

15(3) : 181-188.

Tertiarablagerungen. Sitzungsber. K. Akad. Wiss. Wien, 54( 1): 87-152.

Sitzungsber. K. Akad. Wiss. Wien, 53: 339-342.

Cagnero (Zentralapennin) mit planktonischen Foraminiferen und Nannoplankton. Eclog. Geol. Helv., 62(1): 303-323.

Bayliss, D., 1969. The distribution of Hyalinea balthica and Globorotalia truncatulinoides in the type Calabrian. Lethaia, 2: 133-143.

Be, A.W.H. and Tolderlund, D.S., 1971. Distribution and ecology of living planktonic foraminifera in the surface waters of the Atlantic and Indian oceans. In: B.M. Funnel1 and W.R. Riedel (Editors), The Micropaleontology of Oceans. Cambridge University Press, Cambridge, pp. 105-149.

Beard, J.H., 1969. Pleistocene paleotemperature record based on planktonic foraminifers, Gulf of Mexico. Trans. Gulf Coast Assoc. Geol. SOC., 19 : 535-553.

Beard, J.H. and Lamb, J.L., 1968. The lower limit of the Pliocene and Pleistocene in the Caribbean and Gulf of Mexico. Trans. Gulf Coast Assoc. Geol. SOC., 18: 174-186.

Beck, R.B., Funnel, B.M. and Lord, A.R., 1972. Correlation of Lower Pleistocene Crag at depth in Suffolk. Geol. Mag., 109: 137-139.

Behrensmeyer, A.K., 1970. Preliminary geological interpretation of a new hominid site in the Lake Rudolf basin. Nature, 226: 225-226.

Benda, L. and Meulenkamp, J.E., 1972. Discussion of biostratigraphic correlations in the Eastern Mediterranean Neogene. Z. Dtsch. Geol. Ges., 123: 559-564.

Benson, R.H., 1972a. Psychrospheric and continental Ostracoda from ancient sediments in the floor of the Mediterranean. In: W.B.F. Ryan, K.J. Hsu et al., Initial Reports of the Deep Sea Drilling Project, XV, no. 2. U.S. Government Printing Office, Washington, D.C., pp. 1002-1008.

during the Pliocene. In: D.J. Stanley (Editor), The Mediterranean Sea. Dowden, Hut- chinson and Ross, Stroudsburg, Pa., pp. 63-73.

(Editor), Messinian Events in the Mediterranean, Utrecht, 1973. North-Holland, Amsterdam, pp. 235-242.

Benson, R.H. and Ruggieri, C., 1974. The end of the Miocene, the time of crisis in the Tethys-Mediterranean history. In: Proc. 75th Anniv. Egypt. Geol. Surv., Cairo, 1971, Symp. Afr. Paleontol. Spec. Bull. (in press).

Benson, R.H., 1972b. Ostracodes as indicators of threshold depth in the Mediterranean

Benson, R.H., 1974. An ostracodal view of the Messinian salinity crisis. In : C.W. Drooger

179

Benson, R.H. and Sylvester-Bradley, P.C., 1971. Deep-sea ostracodes and the transformation of ocean to sea in the Tethys. In: H.J. Oertli (Editor), Paleoecologie d’ostracodes, Pau, 1970. Bull. Centre Rech. Pau-SNPA, 5 Suppl., pp. 63-91.

deep-sea core from the south-central North Atlantic. G. Geol., Ser. 2, 35(2): 291-312.

al., Initial Reports of the Deep Sea Drilling Project, I. U.S. Government Printing Office, Washington, D.C., pp. 584-607.

and the radiometric time-scale. Nature, 224(5224): 1072-1075.

cropaleontology, 15( 3): 351-365.

and synthesis of results. In: E.T. Degens and D.A. Ross (Editors), Hot Brines and Re- cent Heavy Metal Deposits in the Red Sea. Springer-Verlag, New York, N.Y., pp.329-335.

Riedel (Editors), The Micropaleontology of Oceans. Cambridge University Press, Cam- bridge, pp. 693-809.

and the radiometric time-scale. Hung. Geol. SOC. Bull., 101: 162-169.

tonic foraminifera. In: A. Farinacci (Editor), Proc. Planktonic Conf., 2nd, Rome, 1970. Edizioni Tecno-Scienza, Rome, pp.41-56.

Berggren, W.A., 1972a. A Cenozoic time-scale - some implications for regional geology and paleobiogeography. Lethaia, 5 : 195-215.

Berggren, W.A., 1972b. Late Pliocene-Pleistocene glaciation. In: A.S. Laughton, W.A. Berggren et al., Preliminary Reports of the Deep Sea Drilling Project, XII. U.S. Govern- ment Printing Office, Washington, D.C., pp. 953-963.

and calcareous nannoplankton zones. Nature, 243( 5407): 391-397.

and Messinian). In : C.W. Drooger (Editor), Symposium on Messinian Events in the Mediter- ranean, Utrecht, 1973. North-Holland, Amsterdam, pp. 10-20.

Berggren, W.A. and Boersma, A., 1969. Late Pleistocene and Holocene planktonic foraminifera from the Red Sea. In: E.T. Degens and D.A. Ross (Editors), Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer-Verlag, New York, N.Y., pp. 282-298.

Berggren, W.A. and Phillips, J.D., 1971. Influence of continental drift on the distribution of the Tertiary benthonic foraminifera in the Caribbean and Mediterranean regions. Symposium on the Geology of Libya, Faculty of Science, University of Libya. Catholic Press, Beirut, pp. 263-299.

Berggren, W.A., Phillips, J.D., Bertels, A. and Wall, D., 1967. Late Pliocene-Pleistocene stratigraphy in deep-sea cores from the south central North Atlantic. Nature, 216:

Bertolino, V., Borsetti, A.M., Cati, F., Cinelli, D., Colalongo, M.L., Crescenti, U., Dallan, L., De Francesco, A., Dondi, L., D’Onofrio, S., Giannelli, L. Papetti, I., Pomesano Cherchi, A., Salvatorini, G., Sampo, M., Sartoni, S. and Tedeschi, D., 1968. Proposal for a biostratigraphy of the Neogene in Italy based on planktonic foraminifera. C. M. N. S. Proc. IV Sess., Bologna, 1967. G. Geol., Ser. 2, 35(2): 23-30.

Berzi, A., Michaux, J., Hutchinson, J.H. and Lindsay, E., 1970. The Arondelli local fauna, an assemblage of small vertebrates from the Villafranchian stage near Villafranca d’Asti, Italy. C. M. N. S. Proc. IV Sess., Bologna, 1967. G. Geol., 35(1): 133-136.

Berggren, W.A., 1968. Micropaleontology and the Pliocene/Pleistocene boundary in a

Berggren, W.A., 1969a. Biostratigraphy : Cenozoic foraminiferal faunas. In: M. Ewing et

Berggren, W.A., 196913. Cenozoic chronostratigraphy, planktonic foraminiferal zonation

Berggren, W.A., 1969~ . Rates of evolution in some Cenozoic planktonic foraminifera. Mi-

Berggren, W.A., 1969d. Micropaleontologic investigations of Red Sea cores - summation

Berggren, W.A., 1971a. Tertiary boundaries and correlations. In: B.M. Funnel1 and W.R.

Berggren, W.A., 197 l b , Neogene chronostratigraphy, planktonic foraminiferal zonation

Berggren, W.A., 1 9 7 1 ~ . Multiple phylogenetic zonations of the Cenozoic based on plank-

Berggren, W.A., 1973. The Pliocene time-scale: calibration of planktonic foraminiferal

Berggren, W.A., 1974. Biostratigraphy and biochronology of the late Miocene (Tortonian

253-254.

180

Beu, A.G. and Maxwell, P.A., 1968. Molluscan evidence for Tertiary sea temperatures in New Zealand: a reconsideration. Tuatara, 16: 68-74.

Biberson, P., 1970. The problem of correlation between south Europe and North Africa during the Pleistocene. Palaeogeogr., Palaeoclimatol., Palaeoecol., 8 : 113-1 27.

Birkeland, P.W., Crandell, D.R. and Richmond, G.M., 1971. Status of correlation of Quaternary stratigraphical units in the western conterminous United States. Quaterna- ry Res., 1 : 208-227.

Bishop, W. W., 1967. Annotated lexicon of Quaternary stratigraphical nomenclature in East Africa. In: W.W. Bishop and J.D. Clark (Editors), Background to Evolution in Africa. Chicago University Press, Chicago, Ill., pp. 375-395.

Bishop, W.W., 1971. The Late Cenozoic history of East Africa in relation to hominoid evolution. In: K.K. Turekian (Editor), Late Cenozoic Glacial Ages. Yale University Press, New Haven, Conn., and London, pp. 493-527.

northern Kenya rift valley. Nature, 226: 914-918.

tions relevant t o the Miocene fossil mammal sequence in East Africa. Am. J. Sci., 267 : 669-699.

et des Iles ioniennes (Grkce occidentale). Editions Technip., Paris, 142 pp.

the Southern Ocean. Thesis, Florida State Univ., Tallahassee, Fla., 141 pp. (unpublished).

Blanc, A.C., 1957. On the Pleistocene sequence of Rome: paleoecologic and archeologic correlations. Quaternaria, 4: 95-109.

Blow, W.H., 1959. Age, correlation and biostratigraphy of the Upper Tocuyo (San Lo- renzo) and Pozon Formations, Eastern Falcon, Venezuela. Bull. Am. Paleontol.,

Blow, W.H., 1969, Late Middle Eocene to Recent planktonic foraminifera1 biostratigraphy. In: P. Bronnimann and H.H. Renz (Editors), Proc. Int. Conf. Planktonic Micro- fossils, lst. , Geneva, 1967, 1. E.J. Brill, Leiden, pp. 199-421.

Bodelle, J., Lay, C. and Odin, G.S., 1969. Determination d’8ge par la methode geochro- nologie “potassium-argon’’ de glauconies du Bassin de Paris. C.R. Acad. Sci. Paris,

Bishop, W.W. and Chapman, G.R., 1970. Early Pliocene sediments and fossils from the

Bishop, W.W., Miller, J.A. and Fitch, F.J., 1969. New potassium-argon age determina-

Bizon, G.: 1967. Contribution a la connaissance des foraminif6res planctoniques d’Epire

Blair, D.G., 1965. The Distribution of Planktonic Foraminifera in Deep Sea Cores from

39( 178): 67-235 (19 pls.).

268: 1474-1477. Boelstorff, J., 1973. Correlating and dating some older “Pleistocene” tills in the mid-

Bolli, H.M., 1957. Planktonic Foraminifera from the Oligocene-Miocene Cipero and Len-

Bolli, H.M., 1959. Planktonic Foraminifera as index fossils in Trinidad, West Indies, and

Continent. Geol. SOC. Am., Abstr. Progr., 5(4): 301.

gua Formations of Trinidad, B.W.I. U.S. Natl. Mus. Bull., 215: 97-123.

their value for worldwide stratigraphic correlation. Eclogae Geol. Helv., 52( 2): 627-637.

Bolli, H.M., 1964. Observations on the stratigraphic distribution of some warm water planktonic Foraminifera in the young Miocene to Recent. Eclogae Geol. Helv., 57(2): 541-552.

Bolli, H.M., 1966. Zonation of Cretaceous to Pliocene marine sediments based on planktonic foraminifera. Bol. Inf. Asoc. Venez. Geol. Min. Pet., 8 : 119-149.

Bolli, H.M., 1970. The foraminifera of sites 23-31, Leg 4. In: R.G. Bader et al., 1970, Initial Reports of the Deep Sea Drilling Project IV. U.S. Government Printing Office, Washington, D.C., pp. 577-643 (9 pls.).

middle Miocene to Pliocene warm-water sediments. Bol. Inf., Asoc. Venez. Geol. Min. Pet., 8: 119-149.

Bolli, H.M. and Premoli-Silva, I., 1973. Oligocene to Recent planktonic foraminifera and stratigraphy of the Leg 15 sites in the Caribbean Sea. In: N.T. Edgar, J. Saunders et al.,

Bolli, H.M. and Bermudez, P.J., 1965. Zonation based on planktonic Foraminifera of

181

Initial Reports of the Deep Sea Drilling Project, XV. U.S. Government Printing Office, Washington, D.C. (in press).

Bonadonna, F.P., 1968. Studi sul Pleistocene del Lazio, 5. La biostratigrafia di Monte Ma- rio e la “Fauna Malacologica Mariana” di Cerulli Irelli. Mem. SOC. Geol. Ital., 7(2): 261-3 22.

Borisiak, A.A., 1962. A survey of fossil sites of Tertiary land mammals in the U.S.S.R.

Bout, P., 1969. Datations absolues de quelques formations volcaniques d’Auvergne et du Int. Geol. Rev., 4: 845-867.

Velay et chronologie du Quaternaire europCen. Rev. Sci. Nat. Auvergne, 83(4):

Bout, P., 1970. Absolute ages of some volcanic formations in the Auvergne and Velay 45-58.

areas and chronology of the European Pleistocene. Palaeogeogr., Palaeoclimatol., Pala- eoecol., 8 : 95-106.

Bout, P., Frechen, J. and Lippolt, H.-J., 1966. Datations stratigraphiques et radiochrono- logiques de quelques coulees basaltiques de Limagne. Rev. Sci. Nat. Auvergne, BO(4): 2 07 -2 2 3.

Bramlette, M.N., 1961. Pelagic sediments. In: M. Sears (Editor), Progress in Oceanography. Am. Assoc. Adv. Sci., Publ. 67 : 345-366.

Bramlette, M.N. and Wilcoxon, J.A., 1967. Middle Tertiary calcareous nannoplankton of the Cipero Section, Trinidad, W.I. Tulane Stud. Geol., 5(3): 93-130.

Briskin, M. and Berggren, W.A., 1974. Pleistocene stratigraphy and quantitative paleo- oceanography of tropical North Atlantic core V16-205. In: T. Saito and L. Burckle (Editors), Late Neogene Epoch Boundaries. Micropaleontology Press, Am. Mus. Nat. Hist., New York, N.Y. (in press).

Skr. 3., 22: 17-29.

592-600.

c&ne du Dahra. Bull. SOC. GCol. Fr., S6r. 3, 24: 283.

no. 21, Lyon, Alger, 136 pp.

Brives, A., 1894. Terrains mioches de la re‘gion de Carnot (Algbrie). Bull. SOC. Geol. Fr.,

Brives, A,, 1895. Note sur les terrains pliocknes du Dahra. Bull. SOC. Gkol. Fr., Ser. 3, 23:

Brives, A., 1896. Reponse aux observations de M. Welsch au sujet de ma note sur le Plio-

Brives, A., 1897. Les terrains tertiaires du bassin du Chklif et du Dahra. These Sciences

Brocchi, G., 1814. Conchiologia fossile subappenina. Milano, 111 pp. Broecker, W.S., 1965. Isotope geochemistry and the Pleistocene climatic record. In: H.E.

Wright and D.G. Frey (Editors), The Quaternary of the United States. Proc. INQUA Congr., 7th, Princeton, N.J., 1965, pp. 737-753.

Broecker, W.S. and Ku, T.L., 1969. Caribbean cores P6304-8 and P6304-9: new analysis of absolute chronology. Science, 166: 404.

Broecker, W.S. and Thurber, D.L., 1965. Uranium-series of corals and oolites from Ba- haman and Florida (Key) limestones. Science, 149( 3679): 58-60.

Broecker, W.S. and Van Donk, J., 1970. Insolation changes, ice volumes, and the 0 l 8 record in deep-sea cores. Rev. Geophys. Space Phys., 8(1): 169-198.

Broecker, W.S., Turekian, K.K. and Heezen, B.C., 1958. The relation of deep-sea sedimentation rates to variations in climate. Am, J. Sci., 256: 503-517.

Broecker, W.S., Ewing, M. and Heezen, B.C., 1960. Evidence for an abrupt change in climate close to 11,000 years ago. Am. J. Sci., 258: 429-448.

Broecker, W.S., Thurber, D.L., Goddard, J., Ku, T., Matthews, R.K. and Mesolella, K.J., 1968. Milankovich hypothesis supported by precise dating of coral reefs and deep-sea sediments. Science, 159: 297-300.

Biostratigraphic synthesis - Late Oligocene and Neogene of the western tropical Pacific. In: E.L. Winterer et al., Initial Results of the Deep Sea Drilling Project IX, no. 2. U.S. Government Printing Office, Washington, D.C., pp. 1723-1745.

Brtinnimann, P., Martini, E., Resig, J., Riedel, W.R., Sanfilippo, A. and Worsley, T., 1971.

182

Brown, F. and Lajoie, K.R., 197 1. Radiometric age determinations on Pliocene/Pleisto- cene formations in the lower Omo basin, southern Ethiopia. Nature, 229: 483-485.

Bucha, V., HoraEek, J., KoEi, A., Kukla, J., 1969. Palaomagnetische Messungen in Lossen. In: J. Demek and J. Kukla (Editors), Periglazialzone, Loss und Palaolithikum der Tschechoslowakei. INQUA Congress, 8th. Czechoslovak Academy of Sciences, Brno,

Bukry, D., 1970. Coccolith age determinations, Leg 4, Deep Sea Drilling Project. In: R.G. pp. 123-131.

Bader et al., Initial Reports of the Deep Sea Drilling Project, IV. U.S. Government Printing Office, Washington, D.C., pp. 375-381.

Soc. Nat. Hist., Trans., 16(14): 303--328.

Tracey Jr. et al., Initial Reports of the Deep Sea Drilling Project, 111. U.S. Government Printing Office, Washington, D.C., pp. 791-807.

Hays et al., Initial Reports of the Deep Sea Drilling Project, IX. U.S. Government Print- ing Office, Washington, D.C., pp. 817-832.

Bukry, D., 197213. Coccolith stratigraphy - Leg 11, Deep Sea Drilling Project. In: C.D. Hollister, J.I. Ewing et al., Initial Reports of the Deep Sea Drilling Project, XI. U.S. Government Printing Office, Washinton, D.C., pp. 475-482.

Bukry, D., 1972c. Further comments on coccolith stratigraphy - Leg 12, Deep Sea Dril- ling Project. In: A.S. Laughton, W.A. Berggren et al., Initial Reports of the Deep Sea Drilling Project, XII. U S . Government Printing Office, Washington, D.C., pp. 1071-1083.

Bukry, D., 1972d. Coccolith stratigraphy - Leg 14, Deep Sea Drilling Project. In: D. Hays, A.C. Pimm et al., Initial Reports of the Deep Sea Drilling Project, XIV. US . Government Printing Office, Washington, D.C., pp. 487-494.

Bukry, D., 1973. Coccolith stratigraphy - Leg 13, Deep Sea Drilling Project. In: W.B.F. Ryan, K.J. Hsii et al., Initial Reports of the Deep Sea Drilling Project, XIII. U.S. Gov- ernment Printing Office, Washington, D.C., pp. 817-822.

calcareous nannofossils. Tulane Stud. Geol., 6(4): 149-155.

Drilling Project. In: A.E. Maxwell et al., Initial Reports of the Deep Sea Drilling Pro- ject, 111. U.S. Government Printing Office Washington, D.C., pp. 589-611.

Burckle, L.H., 1972. Late Cenozoic planktonic diatom zones from the Eastern Equatorial Pacific. Beih. Nova Hedwigia, 39: 217-246.

Buteux, C., 1843. Esquisse geologique du de'partement de la Somme. Mem. Acad. Sci. Agric., Dept. de la Somme, pp. 187-322.

Butzer, K.W. and Thurber, D.L., 1969. Some late Cenozoic sedimentary formations of the lower Omo basin. Nature, 222: 1132-1143.

Bukry, D., 1971a. Cenozoic calcareous nannofossils from the Pacific Ocean. San Diego

Bukry, D., 1971b. Coccolith stratigraphy - Leg 8, Deep Sea Drilling Project. In: J.I.

Bukry, D., 1972a. Coccolith stratigraphy - Leg 9, Deep Sea Drilling Project. In: J.D.

Bukry, D. and Bramlette, M.N., 1968. Stratigraphic significance of two genera of Tertiary

Bukry, D. and Bramlette, M.N., 1970. Coccolith age determinations, Leg 3, Deep Sea

Campbell, A S . and Clark, R.L., 1944. Miocene radiolarian faunas from southern Califor-

Casey, R.E., 1971. Neogene radiolarian biostratigraphy and paleotemperatures: Califor- nia. Geol. SOC. Am., Spec. Pap., 51: 76 pp.

nia, the experimental Mohole, Antarctic. Geol. SOC. Am., Abstr. Progr. Ann. Meet., 67th Cord. Sect., 3(2): 93.

Casey, R.E., 1972. Neogene radiolarian biostratigraphy and paleotemperatures: Southern California, the Experimental Mohole, Antarctic Core E14-8. Palaeogeogr., Palaeocli- matol., Palaeocol., 12 : 115-130.

Catalano, R. and Sprovieri, R., 1969. Stratigrafia e micropaleontologia dell' intervallo tri- polaceo di torrente Rossi (Enna). Atti Accad. Gioenia, Sci. Nat., Catania, 7(1): 513-517.

183

Catalano, R. and Sprovieri, R., 1971. Biostratigrafia di alcune serie Saheliane (Messiniano inferiore) in Sicilia. Proc. Planctonic Conf., 2nd, Roma, 1970. Edizioni Tecno-Scienza, Rome, 1 : 211-249.

Cati, F., Colalongo, M.L., Crescenti, U., D’Onofrio, S., Pomesano Cherchi, A., Salvatorini, G., Sartoni, S., Premoli-Silva, I., Wezel, C.F., Bertolino, V., Bizon, G., Bolli, H.M., Bor- setti Cati, A.M., Dondi, L., Feinberg, H., Jenkins, D.G., Perconig, E., Sampo, M., and Sprovieri, R., 1968. Biostratigrafia del Neogene mediterraneo basata sui foraminiferi planctonici. Bol. SOC. Geol. Ital., 87 : 491-503.

Chaline, J. and Michaux, J., 1972. An account of Plio-Pleistocene rodent fauna of central and western Europe and the question of the Plio-Pleistocene boundary. In: M.N. Alek- seev, E.A. Vangengeim, K.V. Nikiforova and I.M. Khoreva (Editors), Int. Colloq. Prob- lem “The Boundary between the Neogene and Quaternary”, 111. Moscow, pp. 46-57.

union. Nature, 210: 1212-1214.

foraminifera and its bearing on the Tertiary geohistory of Taiwan. Proc. Geol. SOC. China, 3: 7-30.

Chang, L.S., 1962. Tertiary planktonic foraminifera1 zones of Taiwan and overseas correlation. Mem. Geol. SOC. China, 1: 107-112.

Chang, L.S., 1967. Tertiary biostratigraphy of Taiwan and its correlation. In: K. Hatai (Editor), Tertiary Correlations and Climatic Changes in the Pacific. Sasaki, Sendai.

Chang, L.S., 1974. Miocene/Pliocene boundary in Taiwan. In: T. Saito and L. Burckle (Editors), Late Neogene Epoch Boundaries. Micropaleontology Press, Am. Mus. Nat. Hist., New York, N.Y., (in press).

1967. Note au sujet de 1’Pge isotopique de la limite Mioc&ne-PliocSne au Maroc. C.R. Acad. Sci. Paris, Ser. D, 249: 222-224.

Chepalyga, A.L., 1967. Antropogenoviye Presnovodniye Mollyuski Yuga Russkoy Ravnini i ikh Stratigraficheskoye Znacheniye. Nauka, Moscow, 222 pp.

Chepalyga, A.L., 1972. Neogene-Quaternary boundary on base of fresh-water mollusc fauna. In: M.N. Alekseev, E.A. Vangengeim, K.V. Nikiforova and I.M. Khoreva (Edi- tors), Int. Colloq. Problem “The Boundary Between the Neogene and Quaternary”, 111.

Choubert, G., Hottinger, L., Marcais, J. and Suter, G., 1964. Stratigraphie et micropaleon- tologie du NQogGne au Maroc septentrional. Cursillos Conf. Inst. “Lucas Mallada”, 9 : 229-259.

Choubert, G., Charlot, R., Faure-Muret, A., Hottinger, L., Marcais, J., Tisserant, D. and Vidal, P., 1968. Note pr6liminaire sur le volcanisme messinien-“pontien” au Maroc. C.R. Acad. Sci. Paris, 266: 187-199.

Churcher, C.S., 1970. Two new Upper Miocene giraffids from Fort Ternan, Kenya, East Africa: Palaeotragus primaevus n. sp. and Samotherium africanum n. sp. In: L.S.B. Leakey and R.J.G. Savage (Editors), Fossil Vertebrates oE Africa, 2. Academic Press, London, pp. 1-105.

Cicha, I., 1970. Stratigraphical problems of the Miocene in Europe. Ustred. Ustav Geol., 134 pp.

Cicha, J. and Seneg, J., 1971. F’robleme der Beziehung zwischen bio- und chronostratigra- phie des jiingeren Tertiars. Geol. Zb., Geol. Carpathica (Bratislava), 12( 2): 209-228.

Cicha, J. and Sene& J., 1973. The present state of the biozonal division of the Central Paratethys. Congr. Carpatho-Balkan Geol. Assoc., l o th , Bratislava, Sect. 1 , Stratigra- phy, 1 2 pp. (preprint).

Karpaten: ein Vergleich mit Hilfe planktonischer Organismen. Mitt. Bayer. Staatssamml. Palaontol. Hist. Geol., 11: 279-293.

Chamalaun, F.H. and McDougall, I., 1966. Dating geomagnetic polarity epochs in Re-

Chang, L.S., 1960. Tertiary biostratigraphy of Taiwan with special reference to smaller

Charlot, R., Chaubert, G., Faure-Muret, A., Hottinger, L. Marcais, J. and Tisserant, D.,

MOSCOW, pp.159-164.

Cicha, I., Hagn, H. and Martini, E., 1971. Das Oligozan und Miozan der Alpen und der

184

Cicha, I., Fahlbusch, V. and Fejfar, O., 1972. Die biostratigraphische Korrelation einiger jungtertiarer Wirbeltierfaunen Mitteleuropas. Neues Jahrb. Geol. Palaontol., Abh., 140(2): 129-145.

cene of Cape Verde (North Atlantic). Rev. Micropaleontol., 14( 5): 17-42.

Hsii et al., Initial Reports of the Deep Sea Drilling Project, XIII. U S . Government Printing Office, Washington, D.C., pp.1343-1379.

Cita, M.B., 1974. The Miocene/Pliocene boundary: history and definition. In: T. Saito and L.H. Burckle (Editors), Symposium 109 : Late Neogene Epoch Boundaries. Micro- paleontology Press, Am. Mus. Nat. Hist., New York, N.Y. (in press).

Cita, M.B. and Blow, W.H., 1969. The biostratigraphy of the Langhian, Serravallian and Tortonian stages in the type sections in Italy. Riv. Ital. Paleontol. Stratigr., 75(3): 549-603.

Cita, M.B., Premoli-Silva, I. and Rossi, R., 1965. Foraminiferi planctonici del Tortonia- no-type. Riv. Ital. Paleontol. Stratigr., 71 : 217-308.

Civetta, L., Gasparini, P. and Adams, J.A.S., 1970. Geochronology and geochemical trends of volcanic rocks from Campania, S. Italy. Eclog. Geol. Helv., 65: 57-68.

Clark, D.L., 1969. Paleoecology and sedimentation in part of the Arctic Basin. Arctic 22: 233-245.

Clark, D.L., 1970. Magnetic reversals and sedimentation rates in the Arctic Ocean. Geol. SOC, Am. Bull., 81: 3129-3134.

Clark, D.L., 1971. Arctic Ocean ice cover and its late Cenozoic history. Geol. SOC. Am. Bull., 82: 3313-3324.

Cocconi, G., 1873. Enumerazione sistematica dei molluschi miocenici e pliocenici della provincie di Parma e Piacenza. Mem. Accad. Sci. Bologna, 3(3).

Colalongo, M.L., 1968a. Cenozone a foraminiferi ed ostracodi nel Pliocene e basso Pleistocene della serie del Santerno e dell’Appennino Romagnolo. G. Geol., Ser. 2, 35(3): 29-61.

67-72.

515-542.

villa, F., Manzoni, M., Poluzzi, A. and Russo, A,, 1972. Biostratigrafia e cronostratigra- fia del Pliocene. Boll. SOC. Geol. Ital., 91: 489-509.

Trans. Am. Philos. SOC., N.S., 26: 1-381.

the Talbot Formation in South Carolina. Geol. SOC. Am., Spec. Pap., 87: 243.

northwest of the Surry Scarp in central South Carolina. Southeastern Geol., 5: 119-142.

Conato, V., 1952. Una microfauna pliocenia del subappennino romagnolo. Boll. SOC. Ge- 01. Ital., 74: 155-167.

Conolly, J.R. and Ewing, M., 1965a. Pleistocene glacial-marine zones in North Atlantic deep-sea sediments. Nature, 204: 135-138.

Conolly, J.R. and Ewing, M., 1965b. Ice-rafted detritus as a climate indicator in Antarctic deep-sea cores. Science, 150: 1822-1824.

Conolly, J.R. and Ewing, M., 1970. Ice-rafted detritus in northwest Pacific deep-sea sediments. In: J.D. Hays (Editor), Geological Investigations of the North Pacific. Geol. SOC. Am. Mem., 126: 219-231.

Cita, M.B., 1971. Biostratigraphy, chronostratigraphy and paleo-environment of the Plio-

Cita, M.B., 1973. Pliocene biostratigraphy and chronostratigraphy. In: W.F.B. Ryan, K.H.

Colalongo, M.L., 196813. Ostracodi del neostratotipo del Messiniano. G. Geol., Ser. 2, 35:

Colalongo, M.L., 1970. Appunti biostratigrafici sul Messiniano. G. Geol., Ser. 2, 36:

Colalongo, M.L., Padovani, A., Sartoni, S., Tampieri, R., D’Onofrio, S., Elmi, C., Franca-

Colbert, E.H., 1935. Siwalik mammals in the American Museum of Natural History.

Colquhoun, D.J., 1965. Paleoecological and rock-stratigraphic proof for the existence of

Colquhoun, D.J. and Duncan, J.R., 1964. Rock-stratigraphic distribution of sediments lying

Cooke, C.W., 1945. Geology of Florida. Fla. Geol. Surv., Geol. Bull., 29: 339 pp.

185

Cooke, H.B.S., 1972. Pleistocene chronology: long or short? Marit. Sed., 8: 1-12. Cooke, H.B.S. and Coryndon, S.C., 1970. Pleistocene mammals from the Kaiso Forma-

tion and other related deposits in Uganda. In: L.S.B. Leakey and R.J.G. Savage (Edi- tors), Fossil Vertebrates of Africa, 2. Academic Press, London, pp.107-224.

Cooke, H.B.S. and Maglio, V.J., 1972. Plio-Pleistocene stratigraphy in relation to proboscidean and serid evolution. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Academic Press, Edinburgh, pp. 303-329.

Coppens, Y., 1965. Les Proboscidiens du Tchad. Leur contribution a la chronologie du Quaternaire Africain. Actas V Congr. Panafr. Prehist. Estud. Cuaternario. 1963, 1 : 331-387,

Coppens, Y., 1971. Les vertebres villafranchiens de Tunisie: gisements nouveaux, signification. C.R. Acad. Sci. Paris, 273: 51-54.

Coppens, Y., 1972. La paleontologie de I’homme et la palhontologie des vertebres du Ter- tiaire et du Quaternaire d’Afrique dans les recents Congrhs. Assoc. senegal. Et. Quatern. Ouest Afrique. Bull. Liais. Labs. Cons. Int. Explor. Sci. Mer. Mediterr., 33-34; 57-66.

3247-3260.

tion. Meteorol. Monogr., 8(30): 112-125.

Cox, A., 1968a. Lengths of geomagnetic polarity intervals. J. Geophys. Res., 73( 10):

Cox, A., 1968b. Polar wandering, continental drift, and the onset of Quaternary glacia-

Cox, A., 1968c. Radiometric time-scale for geomagnetic reversals. Q.J. Geol. SOC. Lond.,

Cox, A., 1969. Geomagnetic reversals. Science, 163(3864): 237-245. Cox, A., 1972. Geomagnetic reversals - their frequency, their origin and some problems

of correlation. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Academic Press, Edinburgh, pp.93-106.

Cox, A. and Dalrymple, G.B., 1967a. Geomagnetic polarity epochs: Nunivak Islands, Alaska. Earth Planet. Sci. Lett., 3 : 173-177.

Cox, A. and Dalrymple, G.B., 1967b. Statistical analysis of geomagnetic reversal data and the precision of potassium-argon dating. J. Geophys. Res., 72: 2603-2614.

Cox, A. and Doell, R.R., 1968. Paleomagnetism and Quaternary correlation. Jn: R.B. Morrison and H.E. Wright Jr. (Editors), Means of correlation of Quaternary Succes- sions. Proc. INQUA Congr., 7th, Princeton, N.J., 1965, 8: 253.

Cox, A., Doell, R.R. and Dalrymple, G.B., 1963a. Geomagnetic polarity epochs and Pleis- tocene geochronometry. Nature, 198: 1049-1051.

Cox, A., Doell, R.R. and Dalrymple, G.B., 196313. Geomagnetic polarity epochs: Sierra Nevada 11. Science, 142: 382-385.

Cox, A., Doe\\, R.R. and. ’Dairymple, G.B., 1964a. Geomagnetic polarity epochs. Science, 143: 351-352.

Cox, A., Doell, R.R. and Dalrymple, G.B., 1964b. Reversals of the earth’s magnetic field. Science, 144: 1537-1543.

Cox, A., Doell, R.R. and Dalrymple, G.B., 1965. Quaternary paleomagnetic stratigraphy. In: H.E. Wright Jr. and D.G. Frey (Editors), Quaternary of the United States. Prin- ceton University Press, Princeton, N.J. pp. 817-830.

Cox, A,, Hopkins, D.M. and Dalrymple, G.C., 1966. Geomagnetic polarity epochs: Pri- bilof Islands, Alaska. Geol. Soc. Am. Bull., 77: 883-910.

Cox, A., Doell, R.R. and Dalrymple, 1968. Radiometric time-scale for geomagnetic reversals. Q. J. Geol. SOC. Lond., 124: 53-66.

Craddock, C., Bastien, T.W. and Rutford, R.H., 1964. Geology of the Jones Mountains area. In: R.J. Adie (Editor), Proc. Int. Symp. Antarctic Geology, l s t , Cape Town, 1963. North-Holland, Amsterdam, pp. 171-187.

Crescenti, U., Giannelli, L., Martinez Diaz, C. and Salvatorini, G., 1973. Tentativo di correlazione tra i piani Andalusiano e Messiniano. Atti SOC. Tosc. Sci. Nat., Mem., Ser. A, 80(1973): 17-39.

124: 53-66.

186

Creuze de Lesser, H., 1824. Statistique du departement de 1’HLrault. Boehm, Montpellier,

Crowell, J.C. and Frakes, L.A., 1970. Phanerozoic glaciation and the causes of ice ages.

Crusafont-Pairo, M., 1950. La cuestion del llamado Meotico espafiol. Arrahona (Saba-

Crusafont-Pairo, M., 1964. La biota de Can Llobateres (Sabadell) y su significacibn bio-

Crusafont-Pairo, M., 1965. Observations ‘a un travail de M. Freudenthal et P.Y. Sondaar

606 pp.

Am. J. Sci., 268: 193.

dell), 1.

Idgica. Cursillos Conf. Inst. “Lucas Malladas”, Estud. Geol., 9: 177-179.

sur des nouveaux gisements‘a Hipparion d’Espagne. K. Ned. Akad. Wet., Proc., Ser. B, 68: 121-126.

Crusafont-Pairo, M., 1972. Les Ischyrictis de la transition Vindobonien-Vallesien. Palaeo- vertebr., 5 : 253-260.

Crusafont-Pairo, M. and Aguirre, E., 1971. A new species of Percrocuta from the Middle Miocene of Kenya. Abh. Hess. Landesamt. Bodenforsch., 60: 51-58.

Curry, R.P., 1966. Glaciation about 3,000,000 years ago in the Sierra Nevada. Science, 154: 770-771.

Curtis, G.H., 1965. Potassium-argon date for Early Villafranchian of France. Trans. Am. Geophys. Union, 46: 178.

Curtis, G.H. and Hay, R.L., 1972. Further geological studies and potassium-argon dating at Olduvai Gorge and Ngorongoro Crater. In: W.W. Bishop and J.A. Miller (Editors), Cali- bration of Hominoid Evolution. Scottish Academic Press, Edinburgh, pp. 289-302.

Curtis, G.H., Savage, D.E. and Evernden, J.F., 1961. Critical points in the Cenozoic. Ann. N.Y. Acad. Sci., 91: 342-346.

Dalrymple, G.B., 1963. Potassium-argon dates of some Cenozoic volcanic rocks of the

Dalrymple, G.B., 1964a. Cenozoic chronology of the Sierra Nevada, California. Univ.

Dalrymple, G.B., 196413. Potassium-argon dates of three Pleistocene interglacial basalt

Dalrymple, G.B., 1972. Potassium-argon dating of geomagnetic reversals and North Amer-

Sierra Nevada, California. Geol. SOC. Am. Bull., 74: 379-390.

Calif. Berkeley Publ., Geol. Sci., 47 : 41 pp.

flows from the Sierra Nevada, California. Geol. SOC. Am. Bull., 75: 753-757.

ican glaciations. In: W.W. Bishop and J. Miller (Editors), Calibration of Hominoid Evolu- tion. Scottish Academic Press, Edinburgh, pp. 107-134.

gy of some volcanic rocks from COCOS Island. Nature, 217: 323-326.

ques, and Applications to Geochronology. Freeman and Co., San Francisco, Calif., 258 pp.

Dalrymple, G.B., Cox, A. and Doell, R.R., 1965. Potassium-argon and palaeomagnetism of the Bishop Tuff, California. Geol. SOC. Am. Bull., 76: 665-673.

Dalrymple, G.B., Cox, A., Doell, R.R. and Gromme, C.S., 1967. Pliocene geomagnetic polarity epochs. Earth Planet. Sci. Lett., 2: 163-173.

Dansgaard, W. and Tauber, H., 1969. Glacial oxygen-18 content and Pleistocene ocean temperatures. Science, 166( 3904): 499-502.

Dansgaard, W., Johnson, S.S., Moller, J. and Langway Jr., C.C., 1969. One thousand centuries of climatic record from Camp Century on the Greenland Ice Sheet. Science, 166(3903): 377-381.

tional Arctic Symposium (abstract), Bull. Am. Assoc. Pet. Geol., 54: 2476.

Fr., Paris, 2: 441-1100.

Calatayud-Teruel basin. Part V. K. Ned. Akad. Wet., Proc., Ser. B, 71 : 74-90.

Dalrymple, G.B. and Cox, A., 1968. Palaeomagnetism, potassium-argon ages and petrolo-

Dalrymple, G.B. and Lanphere, M.A., 1969. Potassium-Argon Dating: Principles, Techni-

Darby, D.A., 1970. Carbonate cycles in Arctic Ocean sediment cores. Second Interna-

D’Archiac, A., 1849. Histoire des progrks de la GQologie de 1834 a 1845. C.R. SOC. Ge‘ol.

De Bruijn, H. and Mein, P., 1968. On the mammalian fauna of the Hipparion-beds in the

187

De Bruijn, H. and Meulenkamp, J.E., 1972. Late Miocene rodents from the Pandanassa Formation (Prov. Reythymnon), Crete, Greece. K. Ned. Akad. Wet., Proc., Ser. B, 75: 54-60.

De Bruijn, H., Dawson, M.R. and Mein, P., 1970. Upper Pliocene Rodentia, Lagomorpha, and Insectivora (Mammalia) from the isle of Rhodes (Greece). I, 11, and 111. K. Ned. Akad. Wet., Proc., Ser. B, 73: 535-584.

from the Neogene of Kastellios Hill (Crete): a correlation of continental and marine biozones. I. K. Ned. Akad. Wet., Proc., Ser. B, 74(5): 22 pp.

De Mercey, A., 1874-1877. Sur la classification de la pkriode quaternaire en Picardie. Mem. Soc. Linn. Nord de la France, 4: 18-29.

Denizot, M.G., 1957. Lexique stratigraphique international, 1. Fasc. VII. Tertiaire: France, Belgique, Pays-Bas, Luxembourg. Centre Nat. Rech. Sci., Paris.

Denizot, M.G., 1958a. Sur la convenance de retablir le Tortonien dans son acception pre- mikre comportant bquivalence de tout le Miocene supe'rieur. C.R. Congr. Soc. Sav., Soussection Gkologie, Colloque sur le Miochne, Aix-Marseille, 1958, pp. 139-141.

Denizot, M.G., 1958b. L'Ctage Aquitanien et la limite Oligoc&ne-Miocene. C.R. Congr. Soc. Sav., Sous-section Ge'ologie, Colloque sur le Miockne, Aix-Marseille, 1958, pp. 143-152.

Denton, G.H. and Armstrong, R.L., 1968. Glacial geology and chronology of the McMurdo Sound region. Antarct. J. U.S., 3: 99.

Denton, G.H. and Armstrong, R.L., 1969. Miocene-Pliocene glaciations in southern Alas- ka. Am. J. Sci., 267: 1121-1142.

Denton, G.H., Armstrong, R.L. and Stuiver, M., 1969. Histoire glaciaire et chronologie de la rkgion due detroit de McMurdo, sud de la Terre Victoria, Antarctide; note prbliminaire. Rev. Geograph. Ph ys. Geol. Dyn., 11 : 265.

Denton, G.H., Armstrong, R.L. and Stuiver, M. 1970. Late Cenozoic glaciation in Antarc- tica: the record in the McMurdo Sound region. Antarct. J. U.S., 5: 15-21.

Denton, G.H., Armstrong, R.L. and Stuiver, M., 1971. The late Cenozoic glacial history of Antarctica. In: K.K. Turekian (Editor), The Late Cenozoic Glacial Ages. Yale Uni- versity Press, New Haven, Conn., 606 pp.

Dep&et, Ch., 1892. Note sur la classification et le parall6lisme du systeme Miochne. C.R. Soc. G601. Fr., 13: 165-168.

Depe'ret, Ch., 1893. Note sur la classification et le paralle'lisme du systirme MiocBne. Bull. Soc. GQol. Fr., Se'r. 3, 21: 170-266.

Deperet, M.C., 1895. Observations 2 propos de la note sur la nomenclature des terrains sidi- mentaires, par MM. Munier-Chalmas et de Lapparent. C.R. Soc. Gkol. Fr., S&. 3, 23: 33-36.

Depgret, M.C., 1918. Essai de coordination chronologique des temps quaternaires. C.R. Acad. Sci., Paris, 166(2): 480-486.

De Rouville, P.G., 1853. Description ge'ologique des environs de Montpellier. Boehm, Montpellier, 185 pp.

De Serres, M., 1830. De la simultanCit6 des terrains de sediment supkrieurs. Extrait de la Geographie Physique, V, de 1'Encyclopedie Me'thodique, 125 pp.

De Serres, M., 1832. Sur les animaux dkcouverts dans diffkrentes couches des dbp6ts quaternaires. Bull. Soc. G601. Fr., Se'r. 1, 2: 450.

De Serres, M., 1855. Des caractbres et de l'importance de la p6riode quaternaire. Bull. Soc. GQol. Fr., SQr. 2, 12 : 1257-1263.

Desnoyers, J., 1829. Observations sur un ensemble de dkp6ts marins plus rkcents que le terrain tertiaire du bassin de la Seine, et pouvant constituer une formation g6ologique distincte; preckdkes d'un aperqu sur la non-simultane'ith des bassins tertiaires. Ann. Sci. Nat., 16 : 171-214; 402-491.

Hot-Brine deep area of the Central Red Sea. In: E.T. Degens and D.A. Ross (Editors),

De Bruijn, H., Sondaar, P.Y. and Zachariasse, W.J., 1971. Mammalia and foraminifera

Deuser, W.G. and Degens, E.T., 1969. 018/016 and C13/C12 ratios of fossils from the

188

Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer-Verlag, New York, pp. 336-347.

Devereux, I., 1967. Oxygen isotope paleotemperature measurements on New Zealand Tertiary fossils. N.Z.J. Sci., 10 : 988-1011.

Devereux, I., Hendy, C.H. and Vella, P., 1970. Pliocene and early Pleistocene sea temperature fluctuations, Mangaopori Stream, New Zealand. Earth Planet. Sci. Lett., 8 :

De Villalta Comella, J.F. and Crusafont-Pairo, M., 1957. Dos nuevas especies de Hippa- 163-1 68.

rion del Pikermiense espaEo1. Cursillos Conf. Inst. “Lucas Malladas”, Estud. Geol., 4: 65-69.

Dieci, G., 1959. I Foraminiferi tortoniani di Montegibbio e Castelvetro (App. Modenese). Paleontogr. Ital., 54: 1-113.

Di Napoli, E., 1952. Foraminifera pelagici e facies in Italia. Atti Conv. Naz. Mek. Pet., 7”, Taormina, 1: 231-254.

Di Stefani, C., 1893. Les terrains tertiaires supkrieurs du bassin de la Mediterranee. Cer- manne, LiPge.

Doderlein, P., 1872. Note illustrative della carta geologica del Modenese e del Reggiano. Memoria terza, Modena, 74 pp.

Doell, R.R. and Dalrymple, G.B., 1966. Geomagnetic polarity epochs: A new polarity event and the age of the Brunhes-Matuyama boundary. Science, 152: 1060-1061.

Doell, R.R. and Dalrymple, G.B., 1974. Potassium-argon ages and paleomagnetism of the Waianae and Koolau Volcanic Series, Oahu, Hawaii. Geol. SOC. Am. Bull. (in press).

Doell, R.R., Dalrymple, G.B. and Cox, A., 1966. Geomagnetic polarity epochs: Sierra Nevada data, 3. J. Geophys. Res., 71: 531-541.

Doell, R.R., Dalrymple, G.B., Smith, R.L. and Bailey, R.A., 1968. Paleomagnetism, potassium-argon ages, and geology of rhyolites and associated rocks of the Valles Calde- ra, New Mexico. Geol Soc. Am. Mem., 116: 211-248.

Pacific sector of the southern ocean, In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 133-140.

Donahue, J.G., 1970. Pleistocene diatoms as climatic indicators in North Pacific sediments. Geol. SOC. Am., Mem., 126: 121-138.

Dondi, L., 1962. Nota paleontologico-stratigrafica sul pede-apennino padano. Boll. SOC. Geol. Ital., 81(4): 113-229.

Dondi, L. and Papetti, I., 1966. Studio paleoecologico e stratigrafico sul passagio Plioce- ne-Quaternario nella bassa valle del Santorno (Emilia). Riv. Ital. Paleontol. Stratigr.,

Dondi, L. and Papetti, I., 1968. Biostratigraphical zones of Po Valley Pliocene. C.M.N.S.

Donn, W.L. and Ewing, M., 1966. A theory of Ice Ages 111. Science, 152: 1706. Donn, W.L. and Shaw, D.M., 1966. The heat budget of an ice-free and an ice-covered

Arctic Ocean. J. Geophys. Res., 71: 1087-1093. Donn, W.L. and Shaw, D.M., 1967. The maintenance of an ice-free Arctic Ocean. In: M.

Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 105-113. D’Onofrio, S., 1964. I Foraminiferi del neostratotipo del Messiniano. G. Geol., Ser. 2,

D’Orbigny, A., 1852. Cours klbmentaire de PalContologie et de Gkologie stratigraphique,

Dorman, F.H., 1966. Australian Tertiary paleotemperatures. J. Geol. 74 : 49. Doumani, G.A., 1964. Volcanoes of the Executive Committee Range, Byrd Land. In: R.J.

Donahue, J.G., 1967. Diatoms as indicators of Pleistocene climatic fluctuations in the

72: 231-244.

Proc. IV Sess., Bologna, 1967, G. Geol., Ser. 2, 35(3): 63-99.

32(2): 409-461.

2.

Adie (Editor), Proc. Int. Symp. Antarctic Geology, lst, Cape Town, 1963. North-Hol- land, Amsterdam, pp. 666-675.

189

Durham, J.W., Jahns, R.H. and Savage, D.E., 1954. Marine-nonmarine relationships in the Cenozoic section of California. In: R.H. Jahns (Editor), Geology of Southern Cali- fornia. Calif. Dep. Nat. Resour., Div. Mines, Bull., 170: 59-71.

Dymond, J.T., 1966. Potassium-argon geochronology of deep-sea sediments. Science,

Dymond, J.T., 1969. Age determinations of deep-sea sediments: a comparison of three 152: 1239-1241.

methods. Earth Planet. Sci. Lett., 6 : 9-14.

Eardley, A.J. and Gvodetsky, V., 1960. Analysis of Pleistocene core from Great Salt La-

Einarsson, T., Hopkins, D.M. and Doell, R.R., 1967. The stratigraphy of Tjornes, North- ke, Utah. Geol. SOC. Am. Bull., 71: 1323-1344.

ern Iceland and the history of the Bering Land Bridge. In: D.M. Hopkins (Editor), The Bering Land Bridge. Stanford Univ. Press, Stanford, Calif., p. 312.

Emiliani, C., 1955. Pleistocene temperatures. J. Geol., 63: 538-578. Emiliani, C., 1956. Oligocene and Miocene temperatures of the equatorial and subtropical

Atlantic Ocean. J. Geol., 64: 281-288. Emiliani, C., 1961. Cenozoic climatic changes as indicated by the stratigraphy and chro-

nology of deep-sea cores of Globigerina ooze facies. Ann. N.Y. Acad. Sci., 95: 521. Emiliani, C., 1964. Paleotemperature analysis of the Caribbean cores A254-BR-C and

CP-28. Geol. SOC. Am. Bull., 75 : 129. Emiliani, C., 1966a. Paleotemperature analysis of the Caribbean cores P6304-8 and

P6304-9 and a generalized temperature curve for the last 425,000 years. J. Geol., 74: 109-126.

Emiliani, C., 1966b. Isotopic paleotemperatures. Science, 154: 851-856. Emiliani, C., 1967a. The Pleistocene record of the Atlantic and Pacific oceanic sediments;

correlations with the Alaskan stages by absolute dating; and the age of the last reversal of the geomagnetic field. In: M. Sears (Editor), Progress in Oceanography, 4. Perga- mon, Oxford, pp. 219-224.

Emiliani, C., 1967b. Stratigraphic correlations and the Pleistocene Epoch. In: Time and Stratigraphy in the Evolution of Man. Nat. Acad. Sci., Nat. Res. Counc. Publ., 1469: 56-62.

Emiliani, C., 1969. Amplitude of the Pleistocene climatic cycles at low latitudes and the oxygen isotopic composition of the ice caps. Geol. SOC. Am., Abstr. Progr. 7: 56-57.

Emiliani, C., 197 1. Paleotemperature variations across the Plio-Pleistocene boundary. Science, 171: 60-62.

Emiliani, C., 1972. Quaternary paleotemperatures and the duration of the high-temperature intervals. Science, 178: 398-401.

Emiliani, C. and Geiss, J., 1957. On glaciations and their causes. Geol. Rundsch., 46: 576-601.

Emiliani, C. and Milliman, J.D., 1966. Deep-sea sediments and their geologic record. Earth-Sci. Rev., 1: 105-132.

Emiliani, C., Mayeda, T. and Selli, R., 1961. Paleotemperature analysis of the Plio-Pleisto- cene section at Le Castella, Calabria, southern Italy. Geol. SOC. Am. Bull., 72: 67 9-68 8.

Emiliani, C., Gartner, S. and Lidz, B., 1972. Neogene sedimentation on the Blake Plateau and the emergence of the Central American Isthmus. Palaeogeogr., Palaeoclimatol., Palaeoecol., 11 : 1-10,

Epstein, S., Sharp, R.P. and Gow, A.J., 1970. Antarctic ice sheet: stable isotope analysis of Byrd Station cores and interhemispheric climatic implications. Science, 168( 3939): 1570-1 57 2.

Ericson, D.B., 1959. Coiling direction of Globigerina pachyderma as a climatic index.

Ericson, D.B. and Wollin, G., 1956a. Correlation of six cores from the equatorial Atlantic Science, 130: 219-220.

and the Caribbean. Deep-sea Res., 3: 104-125.

190

Ericson, D.B. and Wollin, G., 1956b. Micropaleontological and isotopic determinations of Pleistocene climates. Micropaleontology, 2( 3): 257-270.

Ericson, D.B. and Wollin, G., 1959. Micropaleontology and lithology of Arctic sediment cores. In: Scientific Studies at Fletcher’s Ice Island, T-3 (1952-1955). U.S. Air Force Cambridge Res. Lab., Geophysics Research Directorate, Geophys. Res. Pap., 63 : 219 pp.

Ericson, D.B. and Wollin, G., 1968. Pleistocene climates and chronology in deep-sea sediments. Science, 162: 1227-1234.

Ericson, D.B. and Wollin, G., 1970. Pleistocene climates in the Atlantic and Pacific oceans: a comparison based on deep-sea sediments. Science, 167: 1483-1485.

Ericson, D.B., Wollin, G. and Wollin, J., 1954. Coiling direction of Globorotalia truncatulinoides in deep-sea cores. Deep-sea Res., 2(2): 152-158.

Ericson, D.B., Broecker, W.S., Kulp, J.L. and Wollin, G., 1956. Late Pleistocene climates and deep-sea sediments. Science, 124( 3218): 385-389.

Ericson, D.B., Ewing, M., Wollin, G. and Heezen, B.C., 1961. Atlantic deep-sea sediment cores. Bull. Geol. SOC. Am., 72: 193-286.

Ericson, D.B., Ewing, M. and Wollin, G., 1963. Pliocene-Pleistocene boundary in deep-sea sediments. Science, 139( 3556): 727-737.

Ericson, D.B., Ewing, M. and Wollin, G., 1964a. Sediment cores from the Arctic and subarctic seas. Science, 144: 1183-1192.

Ericson, D.B., Ewing, M. and Wollin, G., 1964b. The Pleistocene Epoch in deep-sea sediments. Science, 146: 723-732.

Evernden, J.F. and Curtis, G.H., 1965. The potassium-argon dating of late Cenozoic rocks in East Africa and Italy. Curr. Anthropol., 6 : 343-364 (with comments pp. 364-385).

Evernden, J.F. and Evernden, R.K.S., 1970. The Cenozoic time scale. In: O.L. Bandy (Editor), Radiometric Dating and Paleontologic Zonation. Geol. SOC. Am., Spec. Pap., 124: 247 pp.

Evernden, J.F., Savage, E.D., Curtis, G.H. and James, G.T., 1964. Potassium-argon dates and the Cenozoic mammalian chronology of North America. Am. J. Sci., 262: 145-1 98.

Ewing, M. and Donn, W.L., 1956. A theory of ice ages. Science, 123: 1061-1066. Ewing, M. and Donn, W.L., 1958. A theory of ice ages, 2. Science, 127: 1159-1162.

Fahlbusch, V. and Gall, H., 1970. Die obermiozhe Fossil-Lagerstatte Sandelzhausen. 1. Entdeckung, Geologie, Fauneniibersicht und Grabungsbericht fur 1969. Mitt. Bayer. Staatssamml. Palaontol. Hist. Geol., 10 : 365-396.

Fedorov, P.V., 1957. Stratigrafia Chetvertichnikh Otlozheniy i Istorii Razvitiya Kaspiis- kogo Morya. Akad. Nauk S.S.S.R., Moscow, 298 pp.

Fejfar, O., 1964. The Lower Villafranchian vertebrates from Hajnicka near Filikovo in southern Slovakia. Rozpravy Ustred. Ustav Geol., 30: 115 pp.

Fillon, R.H., 1970. Paleoclimatic fluctuations in glacial-marine sediments of Antarctic deep-sea cores. Geol. SOC. Am., Progr. Ann. Meet., Abstr., 2(7): 350.

Fillon, R.H., 1972. Evidence from the Ross Sea for widespread submarine erosion. Na- ture, Phys. Sci., 238: 40-42.

Fisher, V. and Goodell, H.G., 1967. Southern Ocean sedimentation 7,000,000 years ago. Geol. SOC. Am., Progr. Ann. Meet., Abstr., p. 67.

Fitch, F.J. and Miller, J.A., 1970. Radioisotopic age determinations of Lake Rudolf artifact site. Nature, 226: 226-228.

Fleck,R.J., Mercer, J.H., Nairn, A.E.M. and Peterson, D.N., 1972. Chronology of Late Pliocene and Early Pleistocene glacial and magnetic events in southern Argentina. Earth Planet. Sci. Lett., 16: 15-22.

Fleming, C.A., 1950. The upper and lower limits of the Pliocene in New Zealand. N.Z. Geol. Surv., 6th Ann. Staff Conf. (mimeographed, unpublished).

191

Flerow, C.C., 1971. The evolution of certain mammals during the Late Cenozoic. In: K.K. Turekian (Editor), Late Cenozoic Glacial Ages. Yale University Press, New Ha- ven, Conn., and London, pp. 479-491.

Flint, R.F., 1947. Glacial Geology and the Pleistocene Epoch. John Wiley, New York, N.Y., 589 pp.

Flint, R.F., 1965. The Pliocene-Pleistocene boundary. In: International Studies on the Quaternary. Geol. SOC. Am., Spec. Pap., 84: 497-533.

Flint, R.F., 1971. Glacial and Quaternary Geology. John Wiley, New York, N.Y., 892 pp.

Forbes, E., 1846. On the connexion between the distribution of the existing fauna and flora of the British Isles and the geographical changes which have affected their area, especially during the epoch of the Northern Drift. G. B. Geol. Surv., Mem., 1: 336-4 3 2.

Forsten, A.-M., 1968. Revision of the Paleoarctic Hipparion. Acta 2001. Fenn., 119: 1-134.

Forsten, A.-M., 1972. Hipparion primigenium from Southern Tunisia. Travaux de Ge‘ologie tunisienne, No. 5 (1971). Notes Serv. Geol. Tunis., 35: 7-28.

Foster, J.H. and Opdyke, N.D., 1968. Evidence for the existence of the Kaena and Mam- moth events in deep-sea cores. Am. Geophys. Union, 49(1): 156.

Foster, J.H. and Opdyke, N.D., 1970, Upper Miocene to Recent magnetic stratigraphy in deep-sea sediments. J. Geophys. Res., 75( 23): 4465-4473.

Frechen, J. and Lippolt, H.J., 1965. Kalium-Argon-Daten zum Alter des Laacher Vulka- nismus der Rheinterrassen und der Eiszeiten. Eiszeitalter Gegenw., 16: 5-30.

Freudenthal, M., 1968. On the mammalian fauna of the Hipparion-beds in the Calata- yud-Teruel basin (Prov. Zaragoza, Spain), Part IV. The Genus Megacricefodon (Roden- tia). K. Ned. Akad. Wet., Proc. Ser. B., 71: 57-72.

(Pikermian) of Samos, Greece. Am. Mus. Novit., 2402: 10 pp.

Geol., 3: 1-10.

Pacific sediments. Micropaleontology, 13( 2): 217-232.

tors), The Phanerozoic Time-Scale. Q.J. Geol. Soc. Lond., 120 (S supplement): 17 9-1 91.

Freudenthal M., 1970. A new Ruscinomys (Mammalia, Rodentia) from the late Tertiary

Freudenthal, M., 1971. Neogene vertebrates from the Gargano Peninsula, Italy. Scr.

Friend, J.K. and Riedel, W.R., 1967. Cenozoic orosphaerid radiolarians from tropical

Funnell, B.M., 1964. The Tertiary Period. In: W.B. Harland, A.G. Smith, B. Wilcock (Edi-

Gabunia, L.K., 1972. On the Neogene-Quaternary boundary in Europe (as based on the data of mammalian fauna). Int. Colloq. Problem “The Boundary Between Neogene and Quaternary”, 11. Moscow, pp. 34-46.

Gabunia, L. and Rubenstein, M., 1968. On the correlation of the Cenozoic deposits of Eurasia and North America based on the fossil mammals and absolute age data. Int. Geol. Congr., 23rd, Prague, 1968, Proc. Sect. 10 (TertiarylQuaternary boundary), pp. 9-17.

Gabunia, L.K. and Vekna, A.C., 1968. Kuabebi fauna of the Akchagylian mammals. Int. Geol. Congr., 23rd, Prague, 1968, Proc. Sect. 10 (Tertiary/Quaternary boundary), pp. 49-56.

Gamper-Bravo, M.A., 1971. Correlation and paleoclimatic analysis of three P l iocene Pleistocene deep-sea North Atlantic cores. Thesis, Brown University, Providence, R.I., 60 pp. (unpublished).

Gulf of Mexico. Univ. Kans. Paleontol. Contrib., Pap. 29: 7 pp.

nannofossil zones. Trans. Gulf Coast Assoc. Geol. Soc., 19: 585-599.

Gartner Jr., S., 1967. Calcareous nannofossils from Neogene of Trinidad, Jamaica, and

Gartner Jr., S., 1969. Correlation of Neogene planktonic foraminifera and calcareous

192

Gartner Jr., S., 1973. Absolute chronology of the late Neogene nannofossil succession in

Geitzenauer, K.R., 1969. Coccoliths as late Quaternary paleoclimatic indicators in the

Geitzenauer, K., 1972. The Pleistocene calcareous nannoplankton of the subantarctic

Gentry, A.W., 1970. The Bovidae (Mammalia) of the Fort Ternan fossil fauna. In: R.J.G.

the equatorial Pacific. Geol. SOC. Am. Bull., 84(6): 2021-2034.

subantarctic Pacific Ocean. Nature, 223: 170-172.

Pacific Ocean. Deep-sea Res., 19 : 45-60.

Savage and L.S.B. Leakey (Editors), Fossil Vertebrates of Africa, 11. Academic Press, London, pp. 243-323.

fauna. Br. Mus. (Nat. Hist.), Bull. Geol., 20: 234-296.

SOC. Lond., 1967 (1638) : pp. 75-87.

Society of London, 1969. Recommendations on stratigraphical usage. Proc. Geol. Soc. Lond., 1969 (1656): 139-166.

papers by the Russian geologists. Int. Colloq. Problem “The Boundary Between Neo- gene and Quaternary,” 11. Moscow, pp. 61-74.

Sci., 6(2): 177-208.

animaux vertkbres vivants et fossiles. Paris, 263 pp.

(Tortona-Alessandria). Riv. Ital Paleontol. Stratigr., Mem. 6: 167-302.

C.R. Acad. Sci. Paris, 150: 841-844.

de la Sicile. Ann. Univ. Lyon, N.S., 36: 693 pp.

I’Apennin, entre le Monte Gargano et Plaisance. Bull. SOC. GCol. Fr., SPr. 4, 4(6): 324- 348.

Gentry, A.W., 1971. The earliest goats and other antelopes from the Samos Hipparion

George, T.N., et al., 1967. Report of the stratigraphical code sub-committee. Proc. Geol

George, T.N. (Chairman) and Members of the Stratigraphy Committee of the Geological

Gerbova, V.G. and Bromov, V.I., 1972. The lower boundary of the Anthropogene in

Gervais, P., 1865. De 1’anciennetP de 1’Homme. Acad. Sci. Lettr. Montpellier, Mem. Sect.

Gervais, P., 1867-69. Zoologie et pal6ontologie gPn6rales. Nouvelles recherches sur les

Gianotti, A., 1953. Microfauna della serie tortoniana del Rio Mazzapiedi-Castellania

Gignoux, M., 1910. Sur la classification du Pliocene et du Quaternaire de 1’Italie du Sud.

Gignoux, M., 1913. Les formations marines plioc‘enes et quaternaires de 1’Italie du Sud et

Gignoux, M., 1914. L’itage Calabrien (Plioc’ene supirieur marin) sur le versant nord-est de

Gignoux, M., 1936. Geologie stratigraphique. Paris, 2e ed., 709 pp. Gignoux, M., 1950. Ge‘ologie stratigraphique. Paris, 4e ed., 735 pp. Gignoux, M., 1954. Pliockne et Quaternaire marines de la Mediterranee occidentale. Int.

Geol. Congr. 19th, Algeria, 1952, Sect. 13. (Questions diverses de geologie gknkrale), 3(15): 249-258.

682 pp.

86-87.

20th, Mexico, 1956, Sect. 7, pp. 389-395.

lia. Tuatara, 16(1): 56-61.

cene-Pliocene volcanism in Fiji. Nature, 241: 176-180.

septentrional. C.R. Acad. Sci. Paris, Ser. D, 274: 1257-1260.

Gignoux, M., 1955. Stratigraphic Geology. Freeman and Co., San Francisco, Calif.,

Gill, E.D., 1957. The Pliocene-Pleistocene boundary in Australia. Aust. J. Sci., 20(3):

Gill, E.D., 1961. The Pliocene-Pleistocene boundary in Australia. Int. Geol. Congr.,

Gill, E.D., 1968. Oxygen isotope paleotemperature determinations from Victoria, Austra-

Gill, J.B. and McDougall, I., 1973. Biostratigraphic and geological significance of Mio-

GlaCon, G. and Rouvier, H., 1972. Age des mouvements tectoniques majeurs en Tunisie

Glaessner, M.F., 1953. Time-stratigraphy and the Miocene Epoch. Geol. SOC. Am. Bull.,

Glass, B., Ericson, D.B., Heezen, B.C., Opdyke, N.D. and Glass, J.A., 1967. Geomagnetic 64: 647-658.

reversals and Pleistocene chronology. Nature, 216: 437-442.

193

Goodell, H.G. and Watkins, N.D., 1966. Paleomagnetic stratigraphy and sedimentary history of the Southern Ocean. Long. 2O0W-l6O0E. Geol. SOC. Am., Progr. Ann. Meet., 1966, p. 78 (abstract).

Goodell, H.G. and Watkins, N.D., 1968. Paleomagnetic stratigraphy of the Southern Ocean; 2O0W-l6O0E. Deep-sea Res., 15 : 98-112.

Goodell, H.G., Watkins, N.D., Mather, T.T. and Koster, S., 1968. The Antarctic glacial history recorded in sediments of the southern ocean. Palaeogeogr. Palaeoclimatol., Palaeoecol., 5: 41-62.

Gramann, F. and Kockel, F., 1969. Das Neogen im Strimonbecken (Griechisch Ostmaze- donien): Teil I, Stratigraphie und Palaogeographie. Geol. Jahrb., 87 : 445-484.

Grichuk, V.P., Hey, R.W. and Venzo, S., 1965. Report of the Subcommission on the Plio-Pleistocene Boundary. Rep. INQUA Congr. 6th, Warsaw, 1961, 1 : 311-329.

Gromme, C.S. and Hay, R.L., 1963. Magnetization of basalt of Bed I, Olduvai Gorge, Tanganyika. Nature, 200: 560-561.

Gromme‘, C.S. and Hay, R.L., 1967. Geomagnetic polarity epochs: new data from Olduvai Gorge, Tanganyika. Earth Planet. Sci. Lett., 2 : 111-115.

Grommi, C.S. and Hay, R.L., 1971. Geomagnetic polarity epochs: age and duration of the Olduvai normal polarity event. Earth Planet. Sci. Lett., 10 : 179-185.

Gromme‘, C.S., Reilly, T.A., Mussett, A.E. and Hay, R.L., 1971. Paleomagnetism and potassium-argon ages of volcanic rocks of Ngorongoro caldera, Tanzania. Geophys. J.R. Astron. Soc., 22 : 101-115.

Gromov, V.I., 1970. The Tiraspol faunal complex. Palaeogeogr., Palaeoclimatol., Palaeo- ecol., 8: 186-195.

Gromov, V.I. and Nikoforova, K.V., 1968. The boundary between the Neogene and Anthropogene (Quaternary Period). Int. Geoi. Congr., 23rd, Prague, 1968, Proc. Sect. 10 (Tertiary/Quaternary boundary), pp. 57-64.

pliocknes du domaine rhodanien. In: Ve Congres du NCogkne Mediterrane‘en, I. Doc. Lab. Gkol. Univ. Lyon (hors skries) 1971: 131-171.

Gugrin, C., Mein, P., Philippe, M. and Truc, G., 1972. Decouverte d’hipparions antetor- toniens dans le bassin de Vaison-la-Romaine (Vaucluse, Sud-Est de la France). C.R. Acad. Sci. Paris, 274: 1276-1279.

GuCrin, C. and Mein, P., 1971. Les principaux gisements de mammif6res miocknes et

Hamilton, W.B. and Armstrong, R.L., 1969. Late Tertiary glaciation of Antarctica. U.S.

Harrison, C.G.A., 1968. Evolutionary processes and reversals of the earth’s magnetic field.

Haug, E., 1900. Article “Quaternaire”. In: La Grande Encyclope‘die, 27 : 1108-1118. Haug, E., 1908-1911. Traite‘ de gkologie: Les pe‘riodes gdologiques. Libraire Armand

Colin, Paris, fasc. 1: 390 pp; fasc. 2: 468 pp.; fasc. 3: 628 pp. Hay, R.L., 1963. Stratigraphy of Beds I through IV, Olduvai Gorge, Tanganyika. Science,

139: 829-833. Hay, W.W., 1970, Calcareous nannofossils from cores recovered on Leg 4. In: R.G. Bader

et al., Initial Reports of the Deep Sea Drilling Project, IV. U S . Government Printing Office, Washington, D.C., pp. 455-501.

Hay, W.W., Mohler, H.P., Roth, P.H., Schmidt, R.R. and Boudreaux, J.E., 1967. Calcar- eous nannoplankton zonation of the Cenozoic of the Gulf Coast and Caribbean- Antillean area and transoceanic correlation. Trans. Gulf Coast Assoc. Geol. Soc., 17: 4 28-480.

Hays, J.D., 1965. Radiolaria and late Tertiary and Quaternary history of Antarctic seas: Biology of the Antarctic Sea 11. Am. Geophys. Union, Antarct. Res., Ser. 5: pp. l‘25-184.

Hays, J.D., 1967. Quaternary sediments of the Antarctic Ocean. In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 117-131.

Geol. Surv., Prof. Pap., 650A: 214.

Nature, 217: 46-47.

194

Hays, J.D., 1970. The stratigraphy and evolutionary trends of Radiolaria in North Pacific deep-sea sediments. Geol. SOC. Am., Mem., 126: 185-218.

Hays, J.D. and Berggren, W.A., 1971. Quaternary boundaries and correlations. In: B.M. Funnel1 and W.R. Riedel (Editors), Micropaleontology of the Oceans. Cambridge University Press, Cambridge, pp. 669-691.

Hays, J.D. and Opdyke, N.D., 1967. Antarctic Radiolaria magnetic reversals and climatic change. Science, 158: 1001-1011.

Hays, J.D., Saito, T., Opdyke, N.D. and Burckle, L.H., 1969. Pliocene-Pleistocene sediments of the equatorial Pacific - their paleomagnetic, biostratigraphic and climatic record. Geol. SOC. Am. Bull., 80 : 1481-1514.

Hays, J.D. et al., 1972a. Initial Reports of the Deep Sea Drilling Project, IX. U.S. Govern- ment Printing Office, Washington, D.C., pp. vii-xxi, 3-1205.

Hays, J.D., Irving, G. and Lozano, J., 1972b. Sub-Antarctic climatic changes during the last one hundred and forty thousand years as deduced from quantitative studies of Radiolaria. Geol. SOC. Am., Abstr. Progr., 7(4): 531.

Heath, G.R., 1969a. Carbonate sedimentation in the abyssal equatorial Pacific during the past 50 million years. Geol. SOC. Am. Bull., 80: 689.

Heath, G.R., 196913. Mineralogy of Cenozoic deep-sea sediments from the equatorial Paci- fic Ocean. Geol. SOC. Am. Bull., 80: 1997.

Hedberg, H.D. (Editor), 1961. Stratigraphic classification and terminology. Int. Geol. Congr., 21st, Copenhagen, 1960, Pt. 24, Int. Subcomm. Stratigr. Classification, pp. 7-38.

192-1 99.

Congr. 24th, Montreal, 1971, Rep., 6: v-viii, 1-39.

classification, terminology and usage. Int. Subcomm. Stratigr. Classification, Rep. 7a. Lethaia, 5 : 283-295.

Hedberg, H.D., 1968. Some views on chronostratigraphic classification. Geol. Mag., 105( 2):

Hedberg, H.D. (Editor), 1971. Preliminary report on chronostratigraphic units. Int. Geol.

Hedberg, H.D. (Editor), 1972. Introduction to an international guide to stratigraphic

Heer, O., 1858. Flora Tertiaria Helvetiae, 2. Winterthur, 110 pp. Heintz, E., 1969. Mise au point sur les Proboscidiens des gisements villafranchiens de

Perrier-Etouaires (Puy-de-D6me) et de Vialette (Haute-Loire). C.R. Soc. Ge'ol. Fr., 1969(2): 56.

Heirtzler, J.R., 1968. Sea-floor spreading. Sci. Am., 219: 60-69. Heirtzler, J.R., Dickson, G.O., Herron, E.M., Pitman, W.C. and Le Pichon, X., 1968.

Marine magnetic anomalies, geomagnetic field reversals, and motions of the ocean floor and continents. J. Geophys. Res., 13(6): 2119-2136.

Ocean. Nature, 201: 386.

climatology. Palaeogeogr., Palaeoclimatol., Palaeoecol., 6: 251-276.

477.

Meade County, Kansas. Kans. Geol. Surv. Bull., 38 (Pt. 7): 197-220.

Am. Bull., 60( 1): 417-428.

Mich. Acad. Sci. Pap., 43(1957): 3-32.

America. Mich. Acad. Sci. Ann. Rep., 62: 5-30.

Mich. Acad. Sci. Pap., 49: 115-127.

Herman, Y., 1964. Temperate water foraminifera in Quaternary sediments of the Arctic

Herman, Y., 1969. Arctic Ocean Quaternary microfauna and its relation to paleo-

Herman, Y., 1970. Arctic paleo-oceanography in late Cenozoic time. Science, 169: 474-

Hibbard, C.W., 1941. The Borchers fauna, a new Pleistocene interglacial fauna from

Hibbard, C.W., 1949. Pleistocene vertebrate paleontology in North America. Geol. SOC.

Hibbard, C.W., 1958. Summary of North American Pleistocene mammalian local faunas.

Hibbard, C.W., 1960. An interpretation of Pliocene and Pleistocene climates in North

Hibbard, C.W., 1963. A late Illinoian fauna from Kansas and its climatic significance.

Hibbard, C.W., 1972. Sand draw local fauna: correlation, age, and paleoecology. In: M.F.

195

Skinner and C.W. Hibbard with others, Early Pleistocene Preglacial and Gla- cial Rocks and Faunas of North-Central Nebraska. Bull. Am. Mus. Nat. Hist., 148(1): 131-1 34.

Hibbard, C.W. and Taylor, D.W., 1960. Two late Pleistocene faunas from southwestern

Hibbard, C.W., Ray, D.E., Savage, D.E., Taylor, D.W. and Guilday, J.E., 1965. Quater- Kansas. Univ. Mich. Mus. Paleontol. Contrib., 16: 1-223.

nary mammals of North America. In: H.E. Wright and D.G. Frey (Editors), The Qua- ternary of the United States. Princeton University Press, Princeton, N.J., pp. 509-526.

Hoernes, M., 1856. Die fossilen Mollusken der Tertiar-Beckens von Wien. Abh. K.K. Geol. Reichsanstalt, 3: 1-733.

Hopkins, D.M., MacNeil, F.S., Merklin, R.L. and Petrov, O.M., 1965. Quaternary correlations across Bering Strait. Science, 147: 1107-1114.

Horn, P., Lippolt, H.J. and Todt, W., 1972. Kalium-Argon Altersbestimmungen an tertiaren Vulkaniten des Oberrheingrabens: I. Gesamtgesteinsalter. Eclogae Geol. Helv.,

Hornibrook, N. de B., 1968. Distribution of some warm water benthic foraminifera in the

Hornibrook, N. de B., 1971. New Zealand Tertiary climate. N. Z. Geol. Surv., Rep., 47:

Howell, F.C., 1969. Remains of Hominidae from Pliocene/Pleistocene formations in the

Howell, F.C., 1972. Pliocene/Pleistocene Hominidae in Eastern Africa: absolute and rela-

65: 131-156.

New Zealand Tertiary. Tuatara, 16 : 11-15.

1 9 PP.

lower Omo basin, Ethiopia. Nature, 223: 1234-1239.

tive ages. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Academic Press, Edinburgh, pp. 331-368.

Hsu, K.J., Cita, M.B. and Ryan, W.B.F., 1973. The origin of the Mediterranean evaporites. In: W.B.F. Ryan, K.J. Hsii et al., Initial Reports of the Deep Sea Drilling Project, XIII. U.S. Government Printing Office, Washington, D.C., pp. 1203-1231.

Huang, T., 1964. Smaller Foraminifera from the Sanhsien-chi, Taitung, eastern Taiwan. Geol. SOC. China, Proc., 7: 63-72.

Huang, T., 1966. Planktonic Foraminifera from the Somachi Formation, Kikai-jima, Kagoshima Prefecture, Japan. Trans. Proc. Paleontol. SOC. Japan, N.S., 62 : 217-233.

Huang, T., 1967. Late Tertiary planktonic Foraminifera from southern Taiwan. Tohoku Univ., Sci. Rep., Ser. 2 (Geol.), 38(2): 165-192.

Huang, T., 1968. Smaller Foraminiferida from the Pliocene Formation of Miyako-jima, Ryukyu. Tohoku Univ., Sci. Rep., Ser. 2 (Geol.) 39: 47-63.

Huang, T., 1971. Planktonic foraminifers from the Miyazaki Group, Kyushu, Japan. Oceanograph. Sin., 10: 7-23.

Huang, T., 1974. Late Neogene foraminifera1 zonation of southwestern Taiwan. In: T. Saito and L. Burckle (Editors), Late Cenozoic Epoch Boundaries. Micropaleontology Press, Am. Mus. Nat. Hist., New York, N.Y. (in press).

Huddlestun, P., 197 1. Pleistocene paleoclimates based on Radiolaria from subantarctic deep-sea cores. Deep-sea Res., 18(11): 1141-1144.

Hudson, J.D., 1967. Speculations on the depth relations of calcium carbonate solution in recent and ancient seas. Mar. Geol., 5: 473-480.

Hunkins, K. and Kutschale, H., 1967. Quaternary sediments in the Arctic Ocean. In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, p. 89.

Hunkins, K. , Ewing, M., Heezen, B. and Menzies, R., 1960. Biological and geoiogical observations on the first photographs of the Arctic Ocean deep-sea floor. Limnol. Oceanogr., 5: 154.

Hunkins, K., BQ, A.W.H., Opdyke, N.D. and Mathieu, G., 1971. The late Cenozoic history of the Arctic Ocean. In: K. Turekian (Editor), Late Cenozoic Glacial Ages. Yale University Press, New Haven, Conn., pp.215-237.

Huot, J.J.N., 1837. Nouveau cours Clkmentaire de Gdologie. I. Paris.

196

Hurzeler, J., 1958. Oreopithecus barnbotii Gervais, a preliminary report. Verh. Natur- forsch. Ges. Basel, 69(1): 1-48.

Hussain, S.T., 1971. Revision of Hipparion (Equidae, Mammalia) from the Siwalik Hills of Pakistan and India. Abh. Bayer. Akad. Wiss., Math.-Naturwiss. Klasse, H, 147: 68 PP.

Iaccarino, S., 1967. Les foraminifhres du stratotype du Tabianien (Pliockne inf6rieur) de Tabiano Bagni (Parme). SOC. Ital. Sci. Nat. Milano, Mem., 15: 165-180.

Iaccarino, S. and Papani, G., 1967. La trasgressione del Pliocene inferiore (“Tabianiano”) sul Tortoniano del Colle di Vigoleno (Piacenza). Riv. Ital. Paleontol., 73(2): 679-700.

Ikebe, N., 1969. A synoptical table on the Quaternary stratigraphy of Japan. J. Geosci., Osaka City Univ., 12(4): 45-51.

Ikebe, N., Chiji, M. and Nakaseko, K., 1969. Problems on the terminal Miocene of Japan. C.M.N.S. Proc. IV Sess., Bologna, 1967. G. Geol., 35(4): 12-25.

Ikebe, N., Takayanagi, Y., Chiji, M. and Chinzei, K., 1972. Neogene biostratigraphy and radiometric time-scale of Japan - an attempt at intercontinental correlation. Pac. Geol., 4 : 39-78.

land) with foraminifera1 paleotemperature curves from North Atlantic deep sea cores. Geol. SOC. Am., Abstr. Progr., 4(7): 550.

Imbrie, J. and Kipp, N.G., 1969. Quantitative interpretations of late Pleistocene climate based on planktonic foraminifera1 assemblages in Atlantic cores. Abstr. Geol. SOC. Am., Ann. Meet., 1969,Part 7: 113.

paleoclimatology : application to a late Pleistocene Caribbean core. In: K. Turekian (Editor), Late Cenozoic Glacial Ages. Yale University Press, New Haven, Conn., pp. 73-181.

cene Caribbean deep-sea core : comparison of isotopic and faunal methods. Quaternary Res., 3(1): 10-38.

Ingle Jr., J.C., 1966. Pliocene-Miocene boundary in temperate eastern Pacific. Bull. Am. Assoc. Pet. Geol., 50(3), pt. 1 : 618-619 (Abstract).

Ingle Jr., J.C., 1967. Foraminifera1 biofacies variation and the Miocene-Pliocene boundary in southern California. Bull. Am. Paleontol., 52( 236): 217-394.

Isaac, G.L., 1972. Chronology and tempo of cultural change during the Pleistocene. In: W.W. Bishop and J. Miller (Editors), Calibration of Hominoid Evolution. Scottish Academic Press, pp. 381-430.

Isaac, G.L., Leakey, R.F. and Behrensmeyer, A.K., 1971. Archeological traces of early hominid activities, east of Lake Rudolf, Kenya. Science, 173: 1129-1134.

Issel, A., 1914. Lembi fossiliferi quaternari e recenti osservati nella Sardegna meridionale dal Prof. D. Lovisato. Atti R. Acc. Lincei, Rendiconti, Roma (5), 23 : 759-77C

Ivanova, I.K. and Popov, G.I., 1961. Novie danniye o vozraste visokikh dnestrovskikh terras v svyazi s nakhodkami fauni molluskov. Dokl. Akad. Nauk S.S.S.R., 136(6): 1425-1 4 27.

Izett, G.A., Wilcox, R.E., Powers, H.A. and Desborough, G.A., 1970. The Bishop ash bed; a Pleistocene marker bed in the western United States. Quaternary Res., 1: 121-132.

Izett, G.A., Wilcox, R.E., Obradovich, J.D. and Reynolds, R.L., 1971. Evidence for two Pearlette-like ash beds in Nebraska and adjoining areas. Geol. SOC. Am., Ann. Meet., Abstr., 1971: 610.

Izett, G.A., Wilcox, R.E., Borchardt, G.A., 1972. Correlation of a volcanic ash bed in Pleistocene deposits near Mount Blanco, Texas, with the Guaje Pumice Bed of the Jemes Mountains, New Mexico. Quaternary Res., 2: 531-553.

Imbrie, J., 1972. Correlation of the climatic record of the Camp Century ice core (Green-

Imbrie, J. and Kipp, N., 1971. A new micropaleontological method for quantitative

Imbrie, J., Van Donk, J. and Kipp, N., 1973. Paleoclimatic investigation of a late Pleisto-

197

Jaeger, J.-J., 197 1. Les micromammifbres du “Villafranchien” inf6rieur du lac Ichkeul (Tunisie): donnies stratigraphiques et biog6ographiques nouvelles. C.R. Acad. Sci. Paris, S&. D., 273: 562-565.

Jaeger, J.J. and Martin, J., 1971. De‘couverte au Maroc des premiers micromammifhres du Pontien d’Afrique. C.R. Acad. Sci. Paris, Se‘r. D, 272: 2155-2158.

Janoschek, R., 1951. Das inneralpine Wiener Becken. In: Geologie von Osterreich. Deutiche, Vienna.

Jenkins, D.G., 1964. Location of the Pliocene-Pleistocene boundary. Cushman Found. Foramin. Res. Contrib., 15(1): 25-27.

Jenkins, D.G., 1967. Recent distribution, origin and coiling ratio changes in Globorotalia pachyderrna (Ehrenberg). Micropaleontology, 13: 195-203.

Jenkins, D.G., 1968. Planktonic foraminifera as indicators of New Zealand Tertiary paleotemperatures. Tuatara, 16: 32-37.

Jenkins, D.G., 1971. Stratigraphic position of the New Zealand Pliocene-Pleistocene boundary: comment. N. Z. J. Geol. Geophys., 14(2): 418-420.

Jenkins, D.G., 1973, The Pliocene-Pleistocene boundary in New Zealand and Australia. In: T. Saito and L.H. Burckle (Editors), Late Neogene Epoch Boundaries. Micro- paleontology Press, Am. Mus. Nat. Hist., New York, N.Y., (in press).

Johnson, H.S. and DuBar, J.R., 1964. Geomorphic elements of the area between the Cape Fear and Pedee Rivers, North and South Carolina. Southeastern Geol., 6: 37-48.

Johnson, N.M., Opdyke, N.D. and Lindsay, E.H., 1972. Paleomagnetic zonation of Blancan and Irvingtonian faunas from the upper San Pedro Valley, Arizona. Geol. Soc. Am., Abstr. Progr., 4: 555.

profiles through the Antarctic and Greenland ice sheets. Nature, 235: 429-434; 236: 249.

faunas of the Panhandle of Texas. Part I. Introduction, description of localities, preliminary faunal lists. Univ. Calif, Publ. Geol. Sci., 31: 27-49.

Julien, A., 1869. Des ph6nomhe.s glaciaires dans le plateau central de la France, en parti- culihre dans le Puy-de-DGme et le Cantal. BailliBre, Paris, 104 pp.

Johnson, S.J., Dansgaard, W., Clausen, H.B. and Langway, C.C., 1972. Oxygen isotope

Johnston, C.S. and Savage, D.E., 1955. A survey of various Late Cenozoic vertebrate

Kahlke, H.-D., 1961. Revision der Saugertierfaunen der klassischen deutschen Pleistozan-

Kahlke, H.-D., 1968. Vertebratenstratigraphie zur Plio/Pleistozan-Grenze. Int. Geol. Fundstellen von Sussenborn, Mosbach, Taubach. Geologie, 10 : 493-532.

Congr., 23rd, Prague, 1968, Proc. Sect. 10 (TertiaryIQuaternary boundary), pp. 27-39.

Keany, J . and Kennett, J.P., 1973. Pliocene--early Pleistocene paleoclimatic history recorded in Antarctic-Subantarctic deep-sea cores. Deep-sea Res., 19 : 529-548.

Kellogg, T.B., 1972. Late Pleistocene climates in the Norwegian and Greenland Seas. Geol. SOC. Am., Abstr. Progr., 4(7): 560.

Kennett, J.P., 1962. The Kapitean Stage (Upper Miocene) a t Cape Foulwind, West Coast. N.Z.J. Geol. Geophys., 5(4): 620-625.

Kennett, J.P., 1965. Faunal succession in two Upper Miocene-Lower Pliocene sections, Marlborough, New Zealand. Trans. R. SOC. N.Z., 3(15): 197-213.

Kennett, J.P., 1966a. Stratigraphy and fauna of the type section and neighbouring sections of the Kapitean Stage, Greymouth, N.Z. Trans. R. SOC. N.Z., 4(1): 1-77.

Kennett, J.P., 1966b. The Globorotalia crassaformis bioseries in north Westland and Marl- borough, New Zealand. Micropaleontology, 12( 2): 235-245.

Kennett, J.P., 1966c. Biostratigraphy and paleoecology in Upper Miocene-Lower Plio- cene Sections in Wairarapa and southern Hawke’s Bay. Trans. R. SOC. N.Z., 4(3): 83-1 02.

Kennett, J.P., 1966d. Four Upper Miocene to Lower Pliocene sections, Hawke’s Bay to East Cape, N.Z. Trans. R. Soc. N.Z., 4(10): 189-209.

198

Kennett, J.P., 1967a. New Foraminifera from the Upper Miocene and Lower Pliocene of

Kennett, J.P., 1967b. New Foraminifera from the Ross Sea, Antarctica. Cushman

Kennett, J.P., 1967c. Recognition and correlation of the Kapitean Stage (Upper Miocene)

Kennett, J.P., 1968a. Ecology and Distribution of Ross Sea Foraminifera, Antarctica.

Kennett, J.P., 1968b. Latitudinal variation of Globigerina pachyderma in Recent sedi-

Kennett, J.P., 1 9 6 8 ~ . Globorotalia truncatulinoides (d’orbigny) as a paleo-oceanographic

Kennett, J.P., 1968d. Paleo-oceanographic aspects of the foraminiferal zonation in the

New Zealand. N.Z. J. Geol. Geophys., lO(4): 989-1008.

Found. Foramin. Res. Contrib., 18(3): 133-135.

New Zealand. N.Z. J. Geol. Geophys., lO(4): 1051-1063.

N.Z. Dept. Sci. Ind. Res., Bull., 186: 48 pp.

ments of the southwest Pacific Ocean. Micropaleontology, 14( 3): 305-318.

index. Science, 159: 1461-1463.

upper Miocene-Lower Pliocene of New Zealand. C.M.N.S. Proo. IV Session, Bologna, 1967. G. Geol., 35: 143-156.

Kennett, J.P., 1969. Foraminifera1 studies of southern ocean deep-sea cores. Antarct. J.

Kennett, J.P., 1970. Pleistocene paleoclimates and foraminiferal biostratigraphy in subantarctic deep-sea cores. Deep-sea Res., 17 : 125-140.

Kennett, J.P., 1972. The climatic and glacial record in Cenozoic sediments of the South- ern Ocean. In: E.M. van Zinderen-Bakker Sr. (Editor), Palaeoecology of Africa, the Surrounding Islands and Antarctica, 6. Balkema, Cape Town, pp. 59-78.

Kennett, J.P. and Brunner, C., 1973. Antarctic Late Cenozoic glaciation: evidence for initiation of ice-rafting and inferred increased bottom water activity. Geol. SOC. Am.

U.S., 4(5): 178-179.

Bull., 84(6): 2043-2052. Kennett, J.P. and Geitzenauer, K.R., 1969. Pliocene-Pleistocene boundary in a South

Pacific deep-sea core. Nature, 224: 899-901. Kennett, J.P. and Huddlestun, P., 1972a. Late Pleistocene paleoclimatology, foraminiferal

biostratigraphy and tephrochronology, western Gulf of Mexico. Quaternary Res., 2( 1 ) : 38-69.

Kennett, J.P. and Huddlestun, P., 1972b. Abrupt climatic change at 90,000 yrs B.P.: faunal evidence from Gulf of Mexico cores. Quaternary Res., 2(3): 384-395.

Kennett, J . and Watkins, N.D., 1970. Plio-Pleistocene volcanism and geomagnetic reversals. Nature, 227 : 930-934.

Kennett, J.P., Watkins, N.D. and Vella, P., 1971. Paleomagnetic chronology of P l i o c e n e Early Pleistocene climates and the Plio-Pleistocene boundary in New Zealand. Science, 171 : 276-279.

Kennett, J.P., Burns, R.E., Andrews, J.E., Churkin Jr., M., Davies, T.A., Dumitrica, P., Edwards, A.R., Galehouse, J.S., Packham, G.H. and Van der Lingen, G.J., 1972. Australian-Antarctic continental drift, palaeocirculation changes and Oligocene deep- sea erosion, Nature, 239(91): 51-55.

Kent, D., Opdyke, N.D. and Ewing, M., 1971. Climate change in the North Pacific using ice-rafted detritus as a climatic indicator. Geol. SOC. Am. Bull., 82: 2741-2754.

Konecn9, V., Bagdasarjan, G.P. and Vass, D., 1969. Evolution of Neogene vulcanism in central Slovakia and its confrontation with absolute ages. Acta Geol. Acad. Sci. Hung., 13: 245-258.

Kowalski, K., 1960. Cricetidae and Microtidae (Rodentia) from the Pliocene of Weze

Kowalski, K., 1963. The Pliocene and Pleistocene Gliridae (Mammalia, Rodentia) from

Kowalski, K., 1967. Rodents from the Miocene of Opole. Acta Zool. Cracov., 12 : 1-18. Kowalski, K., 1971. The biostratigraphy and paleoecology of Late Cenozoic mammals of

(Poland). Acta Zool. Cracov., 5: 447-505.

Poland. Acta Zool. Cracov., 8: 534-567.

Europe and Asia. In: K.K. Turekian (Editor), Late Cenozoic Glacial Ages. Yale Univer- sity Press, New Haven, Conn., pp. 465-477.

199

Kretzoi, M., 1941. Betrachtungen uber das Problem der Eiszeiten. Ann. Mus. Nat. Hung.,

Kretzoi, M., 1962. Fauna und Faunenhorizont von Csarnota. Jahresber. Ungar. Geol. Anstalt, 1959: 344-395.

Kretzoi, M., 1965. Die Nager und Lagomorphen von Voigtstedt in Thiiringen und ihre chronologische Aussage. Palaontol. Abh., Ser. A, 2 : 586-660.

Krommenhock, W., 1969. Mammalian fossils from the Pleistocene of Lake Edward and the Kazinga Channel. Uganda J., 33: 79-84.

Ku, T.L. and Broecker, W.S., 1966. Atlantic deep-sea stratigraphy: extention of absolute chronology t o 320,000 years. Science, 151 : 448-450.

Ku, T.L. and Broecker, W.S., 1967. Rates of sedimentation in the Arctic Ocean. In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 95-104.

Kukla, J., 1961a. Lithologische Leithorizonte der tschechoslowakischen Lossprofile. Vestn. Ustred. Ustav. Geol., 36(5): 369-372.

Kukla, J., 1961b. Quaternary sedimentation cycle: Survey of Czechoslowak Quaternary. Czwartorzed Europy Srodkoweji i Wschodniej, INQUA VIth Congress, Institut Geolo- giczny, Prace, Warszawa, 34: 145-154.

Kukla, J., 1968. Pleistocene Epoch and the evolution of Man; Comment. Curr. Anthro- pol., 9 : 37-39.

Kukla, J . , 1969a. Die zyklische Entwicklung und die absolute Datierung der Loss-Serien. In : J. Demek and J. Kukla (Editors), Periglazialzone, Loss und Palaolithikum der Tschechoslowakei Brno, pp. 75-96.

34: 56-82.

Kukla, J., 1969b. The cause of the Holocene climate change. Geol. Mijnb., 48: 307-334. Kukla, J., 1970. Correlations between loesses and deep-sea sediments. Geol. Foren.

Kukla, G.J., 1972. Insolation and glacials. Boreas, 1: 63-96. Kukla, J. and Loiek, V., 1961. Loess and related deposits: Survey of Czechoslovak Qua-

Stockh. Forh., 92: 148-180.

ternary. Czwartorzed Europy Srodkowej i Wschodniej, INQUA VIth Congress, Institut Geologiczny, Prace, Warszawa, 34: 11-28.

scale. Geol. SOC. Am., Ann. Meet. Abstr., 4(7): 569-570. Kukla, G.J. and Opdyke, N.D., 1972. American glacial stages in paleomagnetic time-

Kulp, J.L., 1961. Geologic time scale. Science, 133: 1105-1114. Kurt&, B., 1952. The Chinese Hipparion fauna. SOC. Sci. Fenn., Comment. Biol., 13(4):

Kurt&, B., 1968. Pleistocene Mammals of Europe. Aldine Publ. Co., Chicago, Ill., 1-82.

317 pp.

Laffitte, R., 1948. Sur 1’6tage SahClian Pomel. Bull. SOC. Hist. Nat. Afr. Nord, 38: 31-56. Lamb, J.L., 1969. Planktonic foraminiferal datums and late Neogene epoch boundaries in

the Mediterranean, Caribbean and Gulf of Mexico. Trans. Gulf Coast Assoc. Geol. SOC., 19: 559-578.

Lamb, J.L., 1971. Planktonic foraminiferal biostratigraphy and paleomagnetics of late Pliocene and early Pleistocene strata a t Le Castella, Sicily. Trans Gulf Coast Assoc. Geol. Soc., 21: 411-418.

Lamb, J.L. and Beard, J.H., 1972. Late Neogene planktonic foraminifers in the Carib- bean, Gulf of Mexico and Italian stratotypes. Univ. Kans. Paleontol. Contrib., Art. 57 (Protozoa 8): 67 pp.

Laughton, A.S., Berggren W.A., et al., 1970. Deep-sea Drilling Project: Leg 12. Geotimes,

Laughton, A.S., Berggren, W.A., et al., 1972. Initial Reports of the Deep Sea Drilling Project, XII. U.S. Government Printing Office, Washington, D.C., pp. iv-xxi +

15(9): 10-14.

3 --1243. Lavocat, R., 1955. Le genre Hipparion, la limite mio-pliochne, les correlations stratigraphi-

ques entre Europe, AmCrique, et Indie. Bull. Soc. Geol. Fr., 6(5): 325-329.

200

Leakey, M.D., 1971a. Olduvai Gorge, 3. Cambridge University Press, Cambridge. Leakey, M.D., 1971b. Discovery of post-cranial remains of Homo erectus and associated

artefacts in Bed IV a€ Olduvai Gorge, Tanzania. Nature, 232: 380-383. Leakey, R.E.F., 1970. Fauna and artefacts from a new Plio-Pleistocene locality near Lake

Rudolf in Kenya. Nature, 226: 223-224. Leakey, R.E.F., 1971. Further evidence of lower Pleistocene hominids from East Rudolf,

north Kenya. Nature, 231: 241-245. LeMasurier, W.E., 1972a. Volcanic record of Antarctic glacial history: implications with

regard to Cenozoic sea levels. Inst. Br. Geogr., Spec. Publ., 4 : 59-74. LeMasurier, W., 197213. Volcanic record of Cenozoic glacial history of Marie Byrd Land.

In: R.J. Adie (Editor), Antarctic Geology and Geophysics, Universitetsforlaget, Oslo, pp.251-2 59.

Voyage dans la Russie meridionale et la Crime, 4. Paris. Le Play, 1842. Exploration des terrains carbonifhres du Donetz. In: Demidoff (Editor),

Lewis, G.E., 1937. A new Siwalik correlation. Am. J. Sci., 33: 191-204. Lidz, L., 1966. Deep-sea Pleistocene biostratigraphy. Science, 154: 1448-1452. Lippolt, H.J., Gentner, W. and Wimmenauer, W., 1963. Altersbestimmungen nach der

Kalium-Argon-Methode an tertiaren Eruptivgesteinen Siidwestdeutschlands. Jahrb. Geol. Landesamtes Baden-Wurttemberg, 6 : 507-538.

Lipps, J. and Kalisky, M., 1972. California Oligo-Miocene calcareous nannoplankton biostratigraphy and paleoecology . In: E.H. Stinemeyer (Editor), Proc. Pac. Coast Miocene Biostratigr. Symp., Pac. Sect. SOC. Econ. Paleontol. Mineral., Bakersfield, Calif., pp. 239-254.

Lona, F., 1962. Prime analisi pollinologiche sui depositi Terziari-Quaternari di Castell’

Lona, F., 1963. Floristic and glaciologic sequence (from Donau to Mindel) in a complete

Lona, F., Bertoldi, E. and Ricciardi, E., 1974. Synchronization of outstanding stages of

Arquato. Boll. SOC. Geol. Ital., 81: 89-91.

diagram of the Leffe deposit. Geobot. Inst. Rube1 (Zurich), Ber., 34: 64-66.

some Italian Upper Pliocene and Lower Pleistocene sequences, especially by means of palynological researches. Ve Congrks du NBogene Mediterraneen, Lyon, 1971. Proc. (in press).

Braunkohlengebietes von Baccinello (Provinz Grosseto - Italien). Riv. Ital. Paleontol. Stratigr., 74: 147-270.

Loiek, V., 1964. Quarttrmollusken der Tschechoslowakei. Rozpr., Ustred. Ustav. Geol., 31: 374 pp.

Loiek, V., 1969. Palaontologische Charakteristik der Loss-Serien. In: J. Demek and J. Kukla (Editors), Periglazialzone, Loss und PalPolithikum der Tschechoslowakei. Brno, pp. 43-60.

Loiek, V., 1970. Le loess et les formations assimilhes: correlation entre L’Europe Centrale et la France par la faune de Mollusques. Mhm. INQUA Congr., Paris (preprint).

Luyendyk, B.P., Forsyth, D. and Phillips, J.D., 1972. An experimental approach to the paleocirculation of the oceanic surface waters. Geol. SOC. Am. Bull., 83(9): 2649-2664.

Lyell, C., 1833. Principles of Geology, 3. John Murray, London, 1st ed., 398 pp. Lyell, C., 1839. Nouveaux 616ments de gkologie. Pitois-Levranet, Paris, 648 pp. Lyell, C., 1852. On the Tertiary strata of Belgium and French Flanders, Part 11: The

Lyell, C., 1857. Supplement to the Fifth Edition. John Murray, London, 40 pp. Lyell, C., 1865. Elements of Geology. John Murray, London, 6th ed., 794 pp.

Maglio, V.J., 1969. The status of the East African elephant Archidishodon exoptatus

Lorenz, H.G., 1968. Stratigraphische und mikropalaontologische Untersuchungen des

Lower Tertiary of Belgium. Q. J. Geol. Soc. Lond., 8: 277-370.

Dietrich 1942. Breviora, 336: 1-25.

201

Maglio, V.J., 1970a. Four new species of Elephantidae from the Plio-Pleistocene of north-

Maglio, V.J., 1970b. Early Elephantidae of Africa and a tentative correlation of African

Maglio, V.J., 1971. Vertebrate fossils from the Kubi Algi, Koobi Fora and Ileret areas,

Maglio, V.J., 1972. Vertebrate faunas and chronology of hominid-bearing sediments east

Maglio, V.J. and Hendey, Q.B., 1971. New evidence relating to the supposed stegolopho-

Malagoli, A., 1892. Foraminifera riscontrati nel Pliocene di Castell’Arquato. Boll. SOC.

Mandra, Y.T. and Mandra, H., 1970. Antarctic Tertiary marine climate based on silicofla-

Mansfield, W.C., 1932. Pliocene fossils from limestones in south Florida. U.S. Geol. Surv.,

Margolis, S.V. and Kennett, J.P., 1970. Antarctic glaciation during the Tertiary recorded

Margolis, S.V. and Kennett, J.P., 1971. Cenozoic paleoglacial history of Antarctica re-

Martini, E., 1971. Standard Tertiary and Quaternary calcareous nannoplankton zonation.

western Kenya. Breviora, 341: 1-43.

Plio-Pleistocene deposits. Nature, 25 : 328-332.

East Rudolf, Kenya. Nature, 231: 243-249.

of Lake Rudolf, Kenya. Nature, 239: 379-385.

dont ancestry of the Elephantidae. S. Afr. Archeol. Bull., 25: 85-87.

Geol. Ital., 11.

gellates. Antarct. J . US., 5: 178-180.

Prof. Pap., 170(D): 43-49.

in sub-Antarctic deep-sea cores. Science, 170: 1085-1087.

corded in sub-Antarctic deep-sea cores. Am. J. Sci., 271: 1-36.

In: A. Farinacci (Editor), Proc. Planktonic Conf., 2nd, Rome, 1970. Edizioni Tecno- Scienza, Rome, pp. 739-785.

Mohole drilling. J. Paleontol., 37 : 845-856.

Nature, 225: 289-290.

piemontese. Riv. Ital. Paleontol. Stratigr., 60(2): 45-114; and 60(3): 125-194.

hominid remains by P.V. Tobias). Nature, 215: 476-479.

Hist., 61 : 437-560.

nary Res., 2(3) : 368-373.

years (preprint, unpublished MS).

New Zealand. Nature, 212: 979-980.

printed).

graphie. Soc. Hist. Nat. Croatica, 4: 14-47 (Zagreb).

Europas. Verh. Schweiz. Nat. Ges., 1857: 32 pp.

Martini, E. and Bramlette, M.N., 1963. Calcareous nannoplankton from the experimental

Martini, E. and Worsley, T., 1970. Standard Neogene calcareous nannoplankton zonation.

Martinis, B., 1954. Richerche stratigrafiche e micropaleontologiche su e Pliocene

Martyn, J., 1967. Pleistocene deposits and new fossil localities in Kenya (with a note on the

Matthew, W.D., 1929. Critical observations upon Siwalik mammals. Bull. Am. Mus. Nat.

Mathews, R.K., 1973. Dynamics of the ocean-cryosphere system: Barbados data. Quater-

Mathews, R.K. and Imbrie, J., 1971. Pleistocene cyclicity: a synthesis of the last 400,000

Mathews, W.H. and Curtis, G.H., 1966. Date of the Pliocene-Pleistocene boundary in

Mayer-Eymar, Ch., 1888. Tableau des systkmes de terrains de s8diments. (privately

Mayer-Eymar, Ch., 1889. Tableau des terrains de s6diment. Extrait du cours do strati-

Mayer-Eymar, K., 1858. Versuch einer synchronistischen Tabelle der Tertiar-Gebilde

Mayer-Eymar, K., 1868. Tableau des terrains tertiaires suphrieurs. Zurich, 4th ed. McCall, G.J.H., Baker, B.H. and Walsh, J., 1967. Late Tertiary and Quaternary sediments

in the Kenya rift valley. In: W.W. Bishop and J.D. Clark (Editors), Background to Evo- lution in Africa. University of Chicago Press, Chicago, Ill., pp. 191-220.

McDougall, I . , 1964. Potassium-argon ages from lavas of the Hawaiian Islands. Geol. SOC. Am. Bull., 75: 107-128.

McDougall, I., 1969. Potassium-argon ages on lavas of Kohala Volcano, Hawaii. Geol. SOC. Am. Bull., 80: 2597-2600.

202

McDougall, I. and Chamalaun, F.H., 1966. Geomagnetic polarity scale of time. Nature,

McDougall, I. and Chamalaun, F.H., 1969. Isotopic dating and geomagnetic polarity studies on volcanic rocks from Mauritius, Indian Ocean. Geol. SOC. Am. Bull., 80: 1419-1442.

McDougall, I. and Page, R.W., 1974. Toward a physical time-scale for the Neogene - data from the Australasian region. In: T. Saito and L.H. Burckle (Editors), Late Neogene Epoch Boundaries. Micropaleontology Press, Am. Mus. Nat. Hist., New York, N.Y., (in press).

McDougall, I. and Rahman, Aziz ur, 1972. Age of the Gauss-Matuyama boundary and of the Kaena and Mammoth Events. Earth Planet. Sci. Lett., 14: 367-380.

McDougall, I. and Stipp, J.J., 1968. Isotopic dating evidence for the age of climatic deterioration and the Pliocene-Pleistocene boundary. Nature, 219 : 51-53.

McDougall, I. and Tarling, D.H., 1963. Dating of polarity zones in the Hawaiian Islands. Nature, 200: 54-56.

McDougall, I. and Wensink, H., 1966. Paleomagnetism and geochronology of the Plio- cene-Pleistocene lavas in Iceland. Earth Planet. Sci. Lett., 1 : 232-236.

McDougall, I., Allsopp, H.L. and Chamalaun, F.H., 1966. Isotopic dating of the Newer Volcanics of Victoria, Australia and geomagnetic polarity epochs. J. Geophys. Res.,

212: 1415-1418.

71: 6107-6118. McIntyre, A,, 1967. Coccoliths as paleoclimatic indicators of Pleistocene glaciation.

Science, 158( 3806) : 1314-1 31 7. McIntyre, A., 1969. The Coccolithophorida in Red Sea sediments. In: E.T. Degens and

D.A. Ross (Editors), Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer-Verlag, New York, N.Y., pp. 299-305.

Placoliths and Cyrtoliths. Deep-sea Res., 14 : 561-597.

deduced from coccolith floras. Geol. SOC. Am. Abstr. Progr., 7: 121-122.

cene boundary. In: M. Sears (Editor), Progress in Oceanography, 4. Pergamon, Oxford, pp. 3-25.

McIntyre, A., B6, A.W.H. and Roche, M.B., 1970. Modern Pacific Coccolithophoridae: a paleontological thermometer. Trans. N.Y. Acad. Sci., 32(6): 720-731.

McIntyre, A,, B6, A.W.H., Biscaye, P., Burckle, L., Gardner, J., Geitzenauer, K., Roche, M., Imbrie, J., Kipp, N., Ruddiman, W., Moore, T., and Heath, R., 1972a. The glacial North Atlantic 17,000 years ago: paleoisotherm and oceanographic maps derived from floral-faunal parameters by CLIMAP. Geol. SOC. Am., Abstr. Progr., 4(7): 590.

McIntyre, A., Ruddiman, W.F. and Jantzen, R., 1972b. Southward penetration of the North Atlantic Polar Front: faunal and floral evidence of large-scale surface water mass movements over the last 225,000 years. Deep-sea Res., 19 : 61-77.

sediments. Quaternary Res., 1 : 72-192.

lia, Rodentia) du Tertiaire de 1’Europe. Scr. Geol., 2: 1-37.

de 1’Europe sub-occidentale. C. R. Acad. Sci. Paris, 270: 2780-2783.

Am. Bull., 79: 471-486.

(Assem.) Berne, Publ., 79: 217-225.

Sci., 266: 91-109.

McIntyre, A. and B6, A.W.H., 1967. Modern Coccolithophoridae of the Atlantic Ocean. I.

McIntyre, A. and Jantzen, R.E., 1969. North Atlantic Pleistocene climatic fluctuations

McIntyre, A., B&, A.W.H. and Priekstas, R., 1967. Coccoliths and the Pliocene-Pleisto-

McManus, D.A., 1970. Criteria of climatic change in the inorganic components of marine

Mein, P. and Freudenthal, M., 1971. Une nouvelle classification des Cricetidae (Mamma-

Mein, P. and Michaux, J., 1970. Un nouveau stade dans 1’Cvolution des Rongeurs plioc‘enes

Mercer, J.H., 1968a. Glacial geology of the Reedy Glacier area, Antarctica. Geol. SOC.

Mercer, J.H., 1968b. Antarctic ice and Sangamon Sea level. Int. Assoc. Sci. Hydrol., Gen.

Mercer, J.H., 1968c. Variations of some Patagonian glaciers since the Late-Glacial. Am. J.

203

Mercer, J.H., 1970a. A former ice sheet in the Arctic Ocean? Palaeogeogr., Palaeoclimatol.,

Mercer, J.H., 1970b. Variations of some Patagonian glaciers since the Late-Glacial: 11.

Mercer, J.H., 1972. Chilean glacial chronology 20,000-11,000 C-14 years ago: some

Mercer, J.H., 1973. Cainozoic temperature trends in the Southern Hemisphere: Antarctic

Palaeoecol., 8: 19-27.

Am. J. Sci., 269: 1-25.

global comparisons. Science, 176: 1118-1120.

and Andean glacial evidence. In: E.M. Van Zinderen Bakker (Editor), Palaeoecology of Africa. A.A. Balkema, Capetown-Rotterdam, 8 : 87-114.

Argentina before 3.6 m.y. ago and origin of the Patagonian gravels (Rodados Patagoni- cos). Geol. SOC. Am. Ann. Meet., Abstr., 4(7): 594.

theory of climatic change: Barbados data. J. Geol., 77(3): 250-274.

Crete, with special reference to the phylogeny of uniserial Uvigerina from the Mediter- ranean region. Utrecht Micropaleontol. Bull., 2: 168 pp. (Schotanus and Jens, Utrecht.)

and paleogeography of the late Cenozoic of the Island of Rhodes. Z. Dtsch. Geol. Ges.,

Mercer, J.H., Fleck, R.J., Mankinen, E.A. and Sander, W., 1972. Glaciation in southern

Mesolella, K.J., Mathews, R.K., Broecker, W.S. and Thurber, D.L., 1969. The astronomical

Meulenkamp, J.E., 1969. Stratigraphy of Neogene deposits in the Rethymnon province,

Meulenkamp, J.E., De Mulder, E.T.J. and Van de Weerd, A., 1972. Sedimentary history

123: 541-553. Meunier, S., 1908. GCologie. Paris, 989 pp. (152 figs). Michaux, J., 1969. Muridae (Rodentia) du Pliocene sup6rieur d’Espagne et du Midi de la

Michaux, J., 1971a. Arvicolinae (Rodentia) du Pliocene terminal et du Quaternaire ancien

Michaux, J., 1971b. Les Rongeurs (ArvicolidGs, Murid&, et Glirid6s) de la localitd Aron-

France. Palaeovertebrata, 3: 1-25.

de France et d’Espagne. Palaeovertebrata, 4 : 137-214.

delli k Villafranca d’Asti (Italie). Essai de corrhlation stratigraphique. Paleontogr. Ital., 66: 67-80.

Migliorini, C., 1950. The Pliocene-Pleistocene boundary in Italy. Proc. Int. Geol. Con@., 18th, London, 9 : 66-72.

Milankovitch, M., 1941. Canon of insolation and the Ice-Age problem. R. Serbian Acad. Spec. Publ., 132, Section of Mathematical and Natural Sciences, 33. (Translated U S . Dept. of Commerce.)

Mili, R., 1962. Le Quaternaire marin du versant Adriatique et Ionien de la peninsule italienne. Quaternaria, 4: 391-413.

Molina, E., P&ez-Gonzhlez, A. and Aguirre, E., 1972. Observaciones geologicas en el Campo de Calatrava. Estud. Geol., 28: 3-11.

Montenat, C. and Crusafont-Pairo, M., 1970. Ddcouverte de Mammifbres dans le N6ogbne et le Pl6istockne du Levant espagnol (Provinces d’Alicante et de Murcia). C.R. Acad. Sci. Paris, SCr. D, 270: 2434-2437.

Montenat, C. et Martinez, C., 1970. Stratigraphie et micropal6ontologie du N6ogbne au Sud de Murcia (Chaines Bktiques, Espagne). C.R. Acad. Sci. Paris, S&. D, 270: 592-595.

Moore, T., 1971. Radiolaria. In: J.I. Tracey Jr., et al., Initial Reports of the Deep Sea Drilling Project, VIII. U.S. Government Printing Office, Washington, D.C., pp. 727-748.

Morlot, A., 1854. Ueber die quaternaren Gebilde des Rhonegebietes. Verh. Schweiz. Naturforsch. Ges. (Versamml. St. Gallen), 39: 161-164.

Morlot, A,, 1858a. Sur le terrain quaternaire du bassin du Lgman. Bull. Soe. Vaudoise Sci. Nat., 6 : 101-108.

Morlot, A., 1858b. Ueber die quaternaren Gebilde des Rhonegebietes. Verh. Schweiz. Naturforsch. Ges. (Versamml. Bern), 43: 144-150.

Morrison, R.B., 1968. Means of time-stratigraphic division and long-distance correlation of Quaternary successions. In: R.B. Morrison and J.E. Wright Jr. (Editors), Means of

204

Correlation of Quaternary Successions. Proc. INQUA Congr. 7th, Princeton, N.J., 1965 ,8 : 1-113.

J. Geol., 57: 380-412.

Lille, 119 pp.

se‘dimentaires. Bull. SOC. Geol. Fr., Sdr. 3, 21: 438-488.

Movius Jr., H.L., 1949. Villafranchian stratigraphy in southern and southwestern Europe.

Munier-Chalmas, M., 1903. Notice sur les travaux scientifiques de M. Munier-Chalmas.

Munier-Chalmas, M. and De Lapparent, A., 1893. Note sur la nomenclature des terrains

Naeser, C.W., Izett, G.A. and Wilcox, R.E., 1971. Zircon fission-track ages of Pearlette- like volcanic ash beds in the Great Plains. Geol. SOC. Am., Abstr. Progr., 3(7): 657.

Naeser, C.W., Izett, G.A. and Wilcox, R.E., 1973. Zircon fission-track ages of Pearlette family ash beds in Meade County, Kansas. Geology, l ( 4 ) : 187-189.

Nagy, E., 1970. Hungary’s Neogene climate on the basis of palynological researches. C.M.N.S. Proc. IV Session, Bologna, 1967, Pt. 1, G. Geol., 35(1): 91-104.

Nakagawa, H., Niitsuma, N. and Hayasaka, I., 1969. Late Cenozoic geomagnetic chronology of the Boso Peninsula. J. Geol. SOC. Jap., 75( 5): 267-280.

Nakagawa, H., Niitsuma, N. and Elmi, C., 1971. Pliocene and Pleistocene magnetic stratigraphy in Le Castella area, southern Italy - a preliminary report. Quaternary Res., l ( 3 ) : 360-368.

Negadaev-Nikonov, K.N. and Yanovskii, P.V., 1969. Chetvertichnye Otlozheniya Mol- davskoi SSR. Akad. Nauk Moldavskoi S.S.R., Otdel. Paleontol. Stratigraf. Kartya Moldovenske, Kishinev, 89 pp.

Nevesskaja, L.A. and Iljina, L.B., 1969. On the scope and position of the Maeotic stage and on the Miocene/Pliocene boundary in the Ponto/Caspian Basin. G. Geol., Ser. 2, 35(4): 27-39.

study of Quaternary sediments in core V20--130. Geol. SOC. Am., Mem., 126: 139-1 83.

In: B.M. Funnel1 and W.R. Riedel (Editors), Micropaleontology of the Oceans. Cam- bridge University Press, Cambridge, pp. 443-461.

Mar. Geol., 6(2): 99-112.

geomagnetic reversal. Tohoku Univ. Sci. Rept., Ser. 2 (Geol.), 43(1): 1-39.

Nigrini, C., 1970. Radiolarian assemblages in the North Pacific and their application to a

Nigrini, C.A., 1971. Radiolarian zones in the Quaternary of the equatorial Pacific Ocean.

Niitsuma, N., 1970. Some geomagnetic stratigraphical problems in Japan and Italy. J.

Niitsuma, N., 1971. Detailed study of the sediments recording the Matuyama-Brunhes

Nikiforova, K.V., 1970. The volume and subdivision of Pliocene. G. Geol., Ser. 2, 35( 1):

Nikiforova, K.V., 1972. State of the problem of the boundary between the Neogene and the Quaternary. In: M.N. Alekseev, E.A. Vangengeim, K.V. Nikiforova, and I.M. Khoreva (Editors), Int. Colloq. Problem “The boundary between the Neogene and Quaternary”, 111. Moscow, pp. 130-147.

Nikiforova, K.V., Ivanova, I.K. and Konstantinova, N.A., 1970. Tiraspol as a type locality for the Pleistocene of Eastern Europe. Palaeogeogr., Palaeoclimatol., Palaeoecol., 8 : 175-185.

Nikiforova, K.V. (Chief Editor), Beliajeva, E.I., Vangengeim, E.A., Konstantinova, N.A., Negadev-Nikonov, K.N., 1971. Pleistocen Triaspolya. Akad. Nauk S.S.S.R., Geol. Inst., Akad. Nauk Moldavskoi S.S.R., Otdel. Paleontol. Stratigraf. Shtiintsa, Kishinev, 187 pp.

17 1-180.

Obradovich, J.D., 1965. Age of the marine Pleistocene of California. Bull. Am. Assoc. Pet. Geol., 49(7): 1087 (abstract).

Obradovich, J.D., 1968. The potential use of glauconite for late-Cenozoic geochronology. In: R.M. Morrison and H.E. Wright Jr. (Editors), Means of Correlation of Quater- nary Successions, 8. University of Utah Press, Salt Lake City, Utah, pp. 267-279.

205

Odin, G.S., 1973. Rhsultats de datations radiometriques dans des series sedimentaires du Tertiaire de 1'Europe occidentale. Rev. Ge'ogr. Phys. G601. Dyn. 2, 15(3): 317-330.

Odin, G.S., Curry, D., Bodelle, J., Lay, C. and Pomerol, C., 1969a. Ge'ochronologie de niveaux glauconieux tertiaires des bassins de Londres et du Hampshire (mbthode potassium-argon). C.R. SOC. Ge'ol. Fr., 8: 309.

Odin, G.S., Gulinck, M., Bodelle, J. and Lay, C., 1969b. Ge'ochronologie de niveaux glauconieux tertiaires du bassin de Belgique (mhthode potassium-argon). C.R. SOC. Ge'ol. Fr., 6: 198.

Odin, G.S., Bodelle, J., Lay, C. and Pomerol, C., 1970. Ggochronologie de niveaux glauconieux pale'ogbnes d'Allemagne du Nord (m6thode potassium-argon). Re'sultats prbliminaires. C.R. SOC. GQol. Fr., 6: 220.

Olausson, E., 1961. Remarks on Tertiary sequences of two cores from the Pacific. Bull. Geol. Inst. Uppsala, 40: 299-303.

Olausson, E., 1965. Evidence of climatic changes in North Atlantic deep-sea cores with remarks on isotopic paleotemperature analysis. In: M. Sears (Editor), Progress in Oceanography, 3. Pergamon, Oxford, pp. 221-252.

carbonate deposition. In: M. Sears (Editor), Progress in Oceanography, 5. Pergamon, Oxford, pp.245-265.

Foren. Stockh. Forh., 93: 459-475.

of the Deep Sea Drilling Project, V. U.S. Government Printing Office, Washington, D.C., pp. 557-567.

Opdyke, N.D., 1969. The Jaramillo event as detected in oceanic cores. In: S.K. Runcorn (Editor), The Application of Modern Physics to the Earth and Planetary Interiors. Wiley-Interscience, London, pp. 549-554.

Opdyke, N.D., 1972. Paleomagnetism of deep-sea cores. Rev. Geophys. Space Phys., 10( 1): 213-249.

Opd-.ke, N.D. and Foster, J.H., 1970 Paleomagnetism of cores from the North Pacific. Gsol. SOC. Am., Mem., 126: 83-1 19.

Opdyke, N.D., Glass, B., Hays, J.D. and Foster, J.H., 1966. Paleomagnetic study of Ant- arctic deep-sea cores. Science, 154. 349-357.

Osm,,nd, J. and Pollard, L., 1967. Sedimentation rate determination in deep-sea cores by g: mma-ray spectrometry. Earth P1 met. Sci. Lett., 3: 476-480.

Ozarsoy, F., 1965. Etude des gisemeits continentaux et des mammifhres du cenozoique de Turquie. Me'm. SOC. GCol. Fr., 44(102): 1-92.

Olausson, E., 1967. Climatological, geoeconomical, and paleooceanographical aspects on

Olausson, E., 1971. Oceanographic aspects of the Pleistocene of Scandinavia. Geol.

Olsson, R.K. and Goll, R., 1970. Biostratigraphy. In: D. McManus et a]., Initial Reports

Page, R.W. and McDougall, I., 1970. Potassium-argon dating of the Tertiary fl-2 stage

Papani, G. and Pelosio, G., 1962. La Perie Plio-Pleistocenica del T. Stirone (Parmense

Papp, A., 1969. Die Koordinierung des Miozans in der Paratethys (mit 2 Tabellen). Verh.

Parandier, 1891. Notice ge'ologique et palContologique sur la nature des terrains travers6

in New Guinea and its bearing on the geological time-scale. Am. J. Sci., 269: 321-342.

occidentale). Boll. SOC. Geol. Ital., 81(4): 293-335.

Geol. Bundesanstalt, 1969 (1): 2-6.

par le chemin de feu entre Dijon et Chllon-sur-Sa8ne. Bull. SOC. G601. Fr., S6r. 3, 19: 794-8 18.

Pare:o, M.F., 1865. Note sur la subdivision que l'on pourrait e'tablir dans les terrains tertiaires de 1'Apennin septentriond. Bull. SOC. Geol. Fr., Ser. 2, 22: 210-277.

Parker, F.L., 1964. Foraminifera from the experimental Mohole drilling near Guadalupe Island, Mexico. J. Paleontol., 38(4): 6 1 7 4 3 6 .

Parker, F.L., 1967. Late Tertiary biostratigraphy (planktonic Foraminifera) of tropical Indo-Pacific deep-sea cores. Am. Paleontol. Bull., 52( 235): 115-203.

206

Patterson, B., 1966. A new locality for early Pleistocene fossils in northwestern Kenya. Nature, 212: 577-578.

Patterson, B., Behrensmeyer, A.K. and Sill, W.D., 1970. Geology and faunal correlations of a new Pliocene locality in northwestern Kenya. Nature, 226: 918-921.

Penck, A. and Briickner, E., 1909. Die Alpen im Eiszeitalter. Tauchnitz, Leipzig, 1199 pp.

Perconig, E., 1952. Stratigrafia del sondaggio profondo n. 29 di Catemaggiore. Atti Conv. Naz. Mek. Pet., 70, Taormina, 1: 135-148.

Perconig, E., 1955. Richerche stratigrafiche e micropaleontologicere nella regione marchi- giora (foglio fermo). Boll. SOC. Geol. Ital., 77(2-3): 199-269.

Perconig, E., 1964. La estratigrafia del Mioceno en Andalucia occidental (Espafia). El limite Oligoceno-Mioceno y la fase terminal marina del Mioceno. Cursillos Conf. Inst. “Lucas Mallada”, Estud. Geol., 9: 219-228.

Perconig, E., 1968. Biostratigrafia della sezione di Carmona (Andalusia, Spagna) in base ai Foraminiferi planctonici. C.M.N.S. Proc. IVth, Bologna, 1967a. G. Geol., Ser. 2, 35 : 191-218.

Perconig, E. and Granados, L.F., 1973. El Estratotipe del Andaluciense. Coloq. Eur. Micropaleontol., 13th, Madrid, 1972, pp. 225-246.

Peron, A., 1891. Notes sur les subdivisions des terrains tertiaires moyen e t sup6rieur en AlgBrie. Bull. SOC. G601. Fr., SQr. 3, 19: 922-928.

Pevzner, M.A., 1970. Paleomagnetic studies of Pliocene-Quaternary deposits of Pridnie- strovie. Palaeogeogr., Palaeoclimatol., Palaeoecol., 8: 215-219.

Pezzani, F., 1963. Studio micropaleontologico di un campione della serie messiniana di Tabiano Bagni (Parma). Riv. Ital. Paleontol. Stratigr., 69(4): 559-662.

Phillips, J.D., Berggren, W.A., Bertels, A. and Wall, D., 1968. Paleomagnetic stratigraphy and micropaleontology of three deep-sea cores from the south-central North Atlantic. Earth Planet. Sci. Lett., 4(2): 118-130.

Piette, E., 1880. Nomenclature des temps anthropiques primitifs. Imp. Le Vasseur, Laon, 78 PP.

Pilbeam, D.R., 1972. Evolutionary changes in hominoid dentition through geological time. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Academic Press, Edinburgh, pp. 369-380.

Pilgrim, G.E., 1938. Are the Equidae reliable for the correlation of the Siwaliks with the Coenozoic Stages of North America? Rec. Geol. Sum. India, 73: 437-482.

Poag, C.W., 1972. Correlation of Early Quaternary events in the U.S. Gulf Coast. Quater- nary Res., 2: 447-469.

Poag, C.W. and Akers, W.H., 1967. Globigerina nepenthes Todd of Pliocene age from the Gulf Coast. Cushman Found. Foramin. Res. Contrib., 18: 168-175.

Pomel, A., 1858. Sur le systbme des montagnes du Mermoucha et sur terrain sah8lien. C.R. Acad. Sci., Paris, 47: 852-855.

Popov, G.I., 1955. Istoryia Manichskogo proliva v svyazi so stratigrafiyei chernomorskikh i kaspiiskikh chetvertichnikh otlozhenii. Byul. Mosk. 0.-Va. Ispyt. Prir., Otd. Geol., 30(2): 31-49.

Popov, G.I. and Konstantinova, N.A., 1966. 0 stratigraphicheskom raschleneni i vozraste kaspiiskikh otlozheniy basseina Nizhnego Dunaya. Byul. Komis. Izuch. Chetvertichn. Perioda, 31: 120-121.

Rahman, Aziz ur and McDougall, I., 1972. Potassium-argon ages of the newer volcanics

Reboul, H., 1833. Geologie de la pdriode Quaternaire. Paris, 222 pp. Renevier, E., 1897. Chronologie g6ologique. Congr. GBol. Int., 4e, Zurich, 1894,

Repenning, C.A., 1967. Palearctic-Nearctic Mammalian dispersal in the late Cenozoic.

of Victoria. R. SOC. Vic., Proc., 85(1): 61-70.

pp. 523-695.

In: D.M. Hopkins (Editor), The Bering Land Bridge. Stanford University Press, Stan- ford, Calif., pp. 288-311.

207

Richmond, G.M., 1957. Three pre-Wisconsin glacial stages in the Rocky Mountain region. Bull. Geol. SOC. Am., 68: 239.

Richmond, G.M., 1965. Glaciation of the Rocky Mountains. In: H.E. Wright and D.G. Frey (Editors), The Quaternary of the United States. Princeton University Press, Princeton, N.J., pp. 217-230.

Richmond, G.M., 1969. Comparison of the Quaternary Stratigraphy of the Alps and Rocky Mountains. INQUA Congress, 8th, Paris, 17 pp. (Presidential address.)

Richmond, G.M., 1970. Comparison of the Quaternary stratigraphy of the Alps and Rocky Mountains. Quaternary Res., 1( 1): 3-28.

Richmond, G.M. and Obradovich, J.D., 1972. Radiometric correlation of some continental Quaternary deposits: a review (Abstract). Am. Quaternary Assoc., Nat. Conf., Abstr., 2: 47-49.

6: 61-96.

Micropaleontology, 5: 285-302.

Pacific Ocean floor. Q.J. Geol. SOC. Lond., 120: 305-368.

R.G. Bader et al., Initial Reports of the Deep Sea Drilling Project, IV U.S. Government Printing Office, Washington, D.C., pp. 503-575.

Pacific, Leg 7. In: E.L. Winterer et al., Initial Reports of the Deep Sea Drilling Project, VII. U.S. Government Printing Office, Washington, D.C., pp. 1529-1672.

Riedel, W.R., Parker, F.L. and Bramlette, M.N., 1963. “Pliocene-Pleistocene” boundary in deep-sea sediments. Science, 140: 1238-1240.

Rittman, A., 1948. Origine e differenciazione del magma Ischitano. Schweiz. Miner. Mitt., 28: 643-698.

Robinson, E., 1967. The Globoquadrina altispira/Globorotalia truncatulinoides zone in Jamaica. Bol. Inf., Asoc. Venez. Geol. Min. Pet., 10: 97-98.

Robinson, E., 1968. Late Tertiary erosion surfaces and Pleistocene sea-levels in Jamaica. Caribb. Geol. Conf., 5th, St. Thomas, Virgin Islands, 1968 (mimeogr. preprint).

Robinson, E., 1969. Coiling directions in planktonic Foraminifera from the Coastal Group of Jamaica. Trans. Gulf Coast Assoc. Geol. SOC., 19: 555-558.

Robinson, E. and Lamb, J.L., 1970. Preliminary paleomagnetic data from the Plio-Pleisto- cene of Jamaica. Nature, 227: 1236-1237.

Robinson, P. and Black, C.C., 1969. Note prbliminaire sur les vertebres fossiles du Vindo- bonien (Formation Beglia) du Bled Douarah, Gouvernerat de Gafsa, Tunisia. Travaux de geologie Tunisienne, 11. Notes Serv. GBol. Tunis., 31: 67-70.

Romeo, M., 1969. Stratigrafia e microfaunes del Miocene de Monte Grammoia presso Gela (Caltanisetta). Atti Accad. Gioenia, Sci. Nat., Catania, Ser. 7, 1: 239-345.

Rona, E. and Emiliani, C., 1969. Absolute dating of Caribbean cores P6304-8 and P6304-9. Science, 163: 66-68.

Rosholt, J.N., Emiliani, C., Geiss, J., Koczy, F.F. and Wangersky, P.J., 1961. Absolute dating of deep sea cores by the PaZ

Ruddiman, W.F., 1971. Pleistocene sedimentation in the equatorial Atlantic: stratigraphy and faunal paleoclimatology. Bull. Geol. SOC. Am., 82: 283-301.

Ruggieri, G., 1940. I1 Calabriano dell’Appennino romagnolo. Rend. R. Accad. Ital., C1. Sci. Fis. Mat. Nat., Ser. 7.

Ruggieri, G., 1944. I1 Calabriano e il Siciliano della valle del Santerno (Imola). G. Geol., Ser. 2, 17: 95-113.

Ruggieri, G., 1949. I1 Pliocene superiore di Capocolle (Forli). G. Geol., Ser. 3, 20: 19-38. Ruggieri, G., 1950. Gli ultimi capitoli della storia geologica della Romagna. Studi romag-

Riedel, W.R., 1957. Radiolaria: A preliminary stratigraphy. Swed. Deep-sea Exped. Rep.,

Riedel, W.R., 1959. Oligocene and lower Miocene Radiolaria in tropical Pacific sediments.

Riedel, W.R. and Funnell, B.M., 1964. Tertiary sediment cores and microfossils from the

Riedel, W.R. and Sanfilippo, A., 1970. Radiolaria, Leg 4, Deep Sea Drilling Project. In:

Riedel, W.R. and Sanfilippo, A., 1971. Cenozoic Radiolaria from the western tropical

/Thz 3 0 method. J. Geol., 69: 162-185.

noli, 1: 303-311.

208

Ruggieri, G., 1954. La limite entre Pliockne et Quaternaire dans la se'rie Plio-P16istockne du Santerno. Congr. Giol. Int., l ge , Alger, 1952, section 1 3 (questions diverses de giologie ge'ne'rale), 3e partie, 15: 235-240.

6 2 7-6 3 7. Ruggieri, G., 1957a. L'interesse paleontologico della Romagna. Studi romagnoli, 8:

Ruggieri, G., 1957b. Nuovi dati sul contatto Pliocene+!alabriano nella sezione del San- terno (Imola). G. Geol., Ser. 2, 26: 8 pp.

Ruggieri, G., 1960. Segnalazione di Globoquadrina altispira nei trubi di Buonfornello (Palermo). Riv. Min. Sicil., 11: 3-7.

Ruggieri, G., 1961. Alcune zone biostratigrafiche del Pliocene e del Pleistocene italiano. Riv. Ital. Paleontol. Stratigr., 67: 405-417.

Ruggieri, G., 1962. La serie Marina Pliocenica e Quaternaria della Romagna. Pubbl. Cam- era Comm. Ind. Agric., Forli, 79 pp.

Ruggieri, G., 1971. Calabriano e Siciliano nei dintorni di Palermo Parte 1. Riv. Min. Sicil., 22( 130-132): 160-1 71.

Ruggieri, G., 1967. The Miocene and later evolution of the Mediterranean Sea. In: As- pects of Tethyan Biogeography. Systematics Assoc. (London), 7 : 283-290.

Ruggieri, G. and Greco, A., 1965. Studi geologici e paleontologici su Capo Milazzo con particolare riguardo al Milazziano. Geol. Rom., 4: 41-88.

Ruggieri, G. and Selli, R., 1950. I1 Pliocene e il Post-Pliocene dell'Emilia. Int. Geol. Congr., 18th, London, 1948, Rep., 9 : 85-93.

Ruggieri, G., Torre, G., Moroni, M.A. and Aruta, L., 1969. Miocene superiore (Saheliano) nei dintorni di Bonfornello (Palermo). Atti Accad. Gioenia, Sci. Nat., Catania, 7(1): 175-193.

Ruscelli, M., 1952. I foraminiferi del deposit0 tortoniano di Marentino (Torino). Riv. Ital. Paleontol. Stratigr., 58: 39-58.

Ruscelli, M., 1953. Microfaune della serie elveziano del Rio Mazzapiedi-Castellania (Tor- tona-Alessandria). Riv. Ital. Paleontol. Stratigr., Mem. 6: 99-166.

Rutford, R.H., 1969. The Glacial Geology and Geomorphology of the Ellsworth Moun- tains, West Antarctica. Dissertation, University of Miniiesota, Minneapolis, Minn., 311 pp.

Rutford, R.H., Craddock, C. and Bastien, T.W., 1965. Possible late Tertiary glaciation, Jones Mountains, Antarctica (abstract). Geol. SOC. Am., Progr. Ann. Meet., 1965, p. 142 (abstract).

Rutford, R.H., Craddock, C. and Bastien, T.W., 1968. La ,e Tertiary glaciation and sea- level changes in Antarctica. Palaeogeogr., Palaeoclimat A., Palaeoecol., 5: 15.

Rutford, R.H., Craddock, C., Armstrong, R.L. and White C.M., 1970. Tertiary glaciation in the Jones Mountains. SCAR/IUGS Symp. Antarctic Geology and Solid Earth Geo- physics, Oslo, 1970. (Summaries of papers.)

Rutten. M.G., 1959. Paleomagnetic reconnaissance of mid-Italian volcanoes. Geol. Mijnb., 10: 373-374.

Rutten, M.G. and Wensink, H., 1960. Paleomagnetic dating, glaciations and the chronology of the Plio-Pleistocene in Iceland. Int. Geol. Congr., 21st, Copenhagen, 1960, 4: 6:;.

Ryan, W.B.F., 1969. The Floor of the Mediterranean, Part I. Structure and Evolution. Part 11. The Stratigraphy of the Eastern Mediterranean. Thesis, Columbia University, New York, N.Y., 281 pp.

Ryan, W.B.F., Hsii, K.J., Nesteroff, W.D., Pautot, G., Wezel, F., Lort, J.M., Cita, M.B., Maync, W., Stradner, H. and Dumitricu, P., 1970. Deep Sea Drilling Project, Leg 13. Geotimes, 15(10): 12-15.

tectonics and geology of the Mediterranean Sea. In: A.E. Maxwell (Editor), The Sea, 4. John Wiley, New York, N.Y., pp. 387-492.

Ryan, W.B.F., Hsu, K.J., Nesteroff, W.D., Pautot, G., Wezel, F., Lort, J.M., Cita, M.B., Maync, W., Stradner, H. and Dumitricu, P., 1973. Initial Reports of the Deep Sea

Ryan, W.B.F., Stanley, D.J., Hersey, J.B., Fahlquist, D.A. and Allan, T.D., 1971. The

209

Drilling Project, Leg 13, I and 111. U.S. Government Printing Office, Washington, D.C., 1447 pp.

Saggerson, E.P., 1952. Geology of the Kisumu area. Geol. Surv. Kenya, Rep., 21: 86. Saito, T., 1969a. The Miocene-Pliocene and Pliocene-Pleistocene boundaries in deep-sea

sediments. Congr. INQUA, 8e, Paris 1969. Rdsum4s des Communication, sect. 2, 72 bis (abstract).

Saito, T., 1969b. Late Cenozoic stage boundaries in deep-sea sediments. Geol. SOC. Am., Ann. Meet., Abstr., 82(pt. 7): 289-290.

Saito, T. and Funnell, B.F., 1971. Pre-Quaternary sediments and microfossils in the oceans. In: A.E. Maxwell (Editor), The Sea, 4. John Wiley, New York, N.Y., pp. 183-204.

Saito, T., Burckle, L. and Hays, J.D., 1974. Micropaleontologic and paleomagnetic data on the Miocene/Pliocene boundary. In: T. Saito and L. Burckle (Editors), Late Neo- gene Epoch Boundaries. Micropaleontology Press, Am. Mus. Nat. Hist., New York, N.Y. (in press).

Salaj, J. and Stranik, Z., 1971. Contribution B la stratigraphie du Miockne du synclinal de Saouaf (Re‘gion du dj. Fkirine, Tunisie orientale). Travaux de ggologie tunisienne, no. 3, 1970. Notes Serv. GCol. Tunis., 32: 79-82.

Savage, D.E. and Curtis, G.H., 1967. The Villafranchian Age and its radiometric dating. Am. Assoc. Pet. Geol., 41st Ann. Meet., April, 1967 (preprint). Bull. Am. Assoc. Pet. Geol., 51: 479 (abstract).

Savage, D.E. and Curtis, G.H., 1970. The Villafranchian Stage-Age and its radiometric dating. In: O.L. Bandy (Editor), Radiometric Dating and Paleontologic Zonation. Geol. Soc. Am., Spec. Pap., 124: 207-231.

Savage, D.E., Downs, T. and Poe, O.J., 1954. Cenozoic land life of southern California. Calif. Div. Mines, Bull., 170(111): 43-58.

Scholl, D.W., Craeger, J.S., Boyce, R.E., Echols, R. J., Fullam, T. J., Grow, J.A., Koizumu, I., Lee, J.H., Ling, H.Y., Supko, P.R., Stewart, R.J. and Worsley, T.R., 1971. Deep Sea Drilling Project. Geotimcs, 16( 11): 12-15.

Seguenza, G., 1862. Prime ricerche intorno ai rizopodi fossili delle argille Pleistoceniche dei dintorni di Catania. Atti Accad. Gioenia Sci. Nat., Catania, Ser. 2, 8: 85-125.

Seguenza, G., 1868. La formation zancle‘anne, ou recherches sur une nouvelle formation tertiaire. Bull. Soc. Ge‘ol. Fr., Se‘r. 2, 25: 465-486.

Seguenza, G., 1873a. Studii stratigrafici sulla Formazione pliocenica dell’Italia meridionale. I1 Pliocene del Messirinse. Boll. R. Com. Geol. Italia, 4: 84-103.

Seguenza, G., 1873b. Brevisimi cenni intorno la serie terziaria della provincia di Messina. Boll. R. Com. Geol. Italia, 4: 231-238; 259-270.

Seguenza, G., 1879. Le formazioni terziarie nella provincia di Reggio Calabria. Mem. R. Accad. Lincei, C1. Sci. Fis. Mat. Nat., 3(6): 1-446.

Seguenza, G., 1902. I Vert2brati fossili della provincia di Messina. Parte seccsnda, Mammi- feri e Geologia del piano Pontico. Boll. Soc. Geol. Ital., 21: 115-175.

Seguenza, G., 1908. I1 Miocene della provincia di Messina. Atti R. Ace. Lincei, 6(17): 3 79-385.

Selli, R., 1954. I1 Bacino del Metauro. G. Geol., Ser. 2, 24: 1-268. Selli, R., 1957. Sulla trasgressione del Miocene nell’Italia meridionale. G. Geol., Ser. 2, 26:

Selli, R., 1960. I1 Messiniano Mayer-Eymar, 1867. Proposta di un neostratotipo. G. Geol.,

Selli, R., 1964. The Mayer-Eymar Messinian, 1867: Proposal for a neostratotype. Int.

Selli, R., 1967. The Pliocene-Pleistocene boundary in Italian marine sections and its

1-54.

Ser. 2, 28: 1-33.

Geol. Congr., 21st, Copenhagen, 1960, Rept. Session, Norden, 28: 311-333.

relationship to continental stratigraphies. In: M. Sears (Editor), Progress in Oceanogra- phy, 4. Pergamon, Oxford, pp. 67-82.

210

Selli, R., 1970. Report on the absolute age. C.M.N.S. Proc., IV Sess., Bologna, 1967.

Selli, R., 1971. Calabrian. G. Geol., Ser. 2, 37(2): 55-64. Shackleton, N., 1967. Oxygen isotope analyses and Pleistocene temperatures re-assessed.

Nature, 215(5096): 15-17. Shackleton, N.J. and Opdyke, N.D., 1973. Oxygen isotope and palaeomagnetic stratigra-

phy of Equatorial Pacific Core V28-238: oxygen isotope temperature and ice volumes on a l o5 year and lo6 year scale. Quaternary Res., 3( 1): 39-55.

Shackleton, N.J. and Turner, C., 1967. Correlation between marine and terrestrial Pleisto- cene successions. Nature, 216(5120): 1079-1082.

Sharp, R.P., 1968. Sherwin Till-Bishop Tuff geological relationships, Sierra Nevada, California. Geol. SOC. Am. Bull., 79: 351-364.

Sharp, R.P., 1969. Semiquantitative differentiation of glacial moraines near Convict Lake, Sierra Nevada, California. J. Geol., 77 : 68-91.

Sharp, R.P., and Birman, J.H., 1963. Additions to classical sequence of Pleistocene glaciations, Sierra Nevada, California. Geol. SOC. Am. Bull., 74: 1079-1086.

Shaw, D.M. and Donn, W.L., 1968. Milankovitch radiation variations: a quantitative evaluation. Science, 162(3859): 1270-1272.

Shinbo, K. and Maiya, S., 1971. Neogene Tertiary planktonic foraminifera1 zonation in the oil-producing provinces of Japan. In: Stratigraphic Correlation Between Sedimen- tary Basins of the ECAFE Region, 2. United Nations, Min. Resour. Dev. Ser., 36 (pt. B): 135-142 (1970).

Shuey, R.T., Johnson, R.B. and Eardley, A.J., 1970. Paleomagnetism of Lake Bonneville sediments. EOS, 51(4): 277 (abstract).

Sickenberg, 0. and Tobien, H., 1971. New Neogene and Lower Quaternary vertebrate faunas in Turkey. Newslett. Stratigr., 1: 51-61.

Simionescu, J., 1902. Hipparion gracile en Roumanie. Ann. Sci. Univ. Jassy, 11: 1-3 (Extract, Verh. Geol. Reichanstalt Wien, 1901: 15).

Simionescu, J., 1903. Contributiuni la geologia moldovei dintre Siret si Prut. Publ. Fond. Vasilie Adamachi Acad. RomPne, l l ( 9 ) : 73-116.

Simionescu, J., 1904. Ueber einige tertiare Saugertierreste aus der Moldau (Rumanien). Verh. Geol. Reichanstalt Wien, 1904: 70-73.

Simons, E.L. and Pilbeam, D.R., 1965. Preliminary revision of the Dryopithecinae. Folia Primatol., 3: 81-152.

Sissingh, W., 1972. Ostracodes from the Sahelian near Carnot, N. Algeria. K. Ned. Akad. Wet., Proc., Ser. B, 75(1): 84-95.

Smith, J.D. and Foster, J.H., 1969. A short geomagnetic reversal in the Brunhes Normal Polarity Epoch. Science, 163: 565-567.

Smith, L.A., 1965. Paleoenvironmental variation curves and paleoeustatics. Trans. Gulf Coast Assoc. Geol. SOC., 15: 47-60 .

Smith, L.A., 1969. Pleistocene discoasters from the stratotype of the Calabrian Stage (Santa Maria di Catanzaro) and the section at Le Castella, Italy. Trans. Gulf Coast Assoc. Geol. SOC., 19: 579-583.

G. Geol., Ser. 2, 35( 1): 51-59.

Sondaar, P.Y., 1971. The Samos Hipparion. K. Ned. Akad. Wet., Proc., Ser. B., 74:

Sprovieri, R., 1968. La serie Plio-Pleistocenica di Agrigento. G. Geol., 35: 295-301. Sprovieri, R., 1969. Distribuzione dei Foraminiferi nella sezione saheliana di Bonfornello

Steininger, F., 1967. Ein weiterer Zahn von Dryopithecus (Dry.) fontani darwini Abel,

4 1 7-44 1.

(Palermo). Atti Accad. Gioenia Sci. Nat., Catania, Ser. 7, 1: 195-210.

1902 (Mammalia, Pongidae) aus dem Miozan des Wiener Beckens. Folia Primatol., 7: 243-275.

Recent Arctic Ocean sediment. J. Paleontol., 46(4): 573-580. Steuerwald, B.A. and Clark, D.L., 1972. Globigerina pachyderrna in Pleistocene and

21 1

Steuerwald, B.A., Clark, D.L. and Andrew, J.A., 1968. Magnetic stratigraphy and faunal patterns in Arctic Ocean sediments. Earth Planet. Sci. Lett., 5: 79-85.

Stevanovic, P.M., 1964. Contribution ‘a la connaissance de 1’6tage Pontien de Grkce avec une remarque speciale sur la Pontien de la Mer Noire. Cursillos Conf. Inst. “Lucas Mallada”,

Pleistocene boundary in New Zealand. Am. J. Sci., 265: 462-474.

rungen. Sitzungsber. K. Akad. Wiss., Wien, 54(1): 66 + 40 pp.

Stipp, J.J., Chappell, J.M.A. and McDougall, I., 1967. K/Ar age estimate of the Pliocene-

Suess, E., 1866. Untersuchungen uber den Charakter der osterreichischen Tertiarablage-

Takayama, T., 1969. Discoasters from the Lamont Core V21-98 (Preliminary reports of

Takayama, T., 1970. The Pliocene-Pleistocene boundary in the Lamont core V21-98 and

Takayanagi, Y. and Saito, T., 1962. Planktonic Foraminifera from the Nobori Formation,

Talwani, M., Windisch, C.C. and Langseth Jr., M.G., 1971. Reykjanes Ridge crest: a

Tanner, W.F., 1965. Cause and development of an ice age. J. Geol., 73: 413. Tanner, W.F., 1968. Tertiary sea-level symposium - introduction. Palaeogeogr., Palaeo-

climatol., Palaeoecol., 5: 7-14. Tedford, R.H., 1970. Principles and practices of mammalian geochronology in North

America. Proc. North Am. Paleontol. Conv., September 1969, Part F, pp. 666-703. Teilhard de Chardin, P. and Stirton, R.A., 1934. A correlation of some Miocene and

Pliocene mammalian assemblages in North America and Asia with a discussion of the Mio-Pliocene boundary. Univ. Calif. Publ., Dept. Geol. Sci., 23( 8): 277-300.

Thaler, L., 1966. Les Rongeurs fossiles du Bas-Languedoc dans leurs rapports avec l’histoire des faunes et la stratigraphie du Tertiaire d’Europe. M6m. Mus. Natl. Hist. Nat. (Paris), S6r. C, 17: 295 pp.

Thaler, L., Crusafont-Pairo, M. and Adrover, R., 1965. Les premiers micromammifkres du Pliockne d’Espagne: pr6cisions chronologiques et biogbographiques sur la faune d’Al- coy. C.R. Acad. Sci. Paris, 260: 4024-4037.

Thenius, E., 1952. Die Saugetierfauna aus dem Torton von Neudorf an der March (CSR). Neues Jahrb. Geol. Palaontol. Abh., 96: 27-136.

Thenius, E., 1959. In : F. Lotze (Editor), Handbuch der stratigraphischen Geologie. Band 111, Tertiar. Zweiter teil, Wirbeltierfaunen. Enke Verlag, Stuttgart, 328 pp.

Tjalsma, R.C., 1971. Stratigraphy and foraminifera of the Neogene of the eastern Guadal- quivir Basin (southern Spain). Utrecht Micropaleontol. Bull., 4: 161 pp.

Tjalsma, R.C. and Wonders, A.A.H., 1972. Stratigraphic position and planktonic Fora- minifera of Brives’ classic reference sections of the Sahelian a t Carnot, N. Algeria. Proc. K. Ned. Akad. Wet., Ser. B, 75(1): 69-83.

Tobien, H., 1952. Die oberpliozane Saugerfauna von Wolfersheim-Wetterau. Z. Dtsch. Geol. Ges., 104: 191.

Tobien, H., 1956. Zur Okologie der jungtertiaren Saugetiere vom Howenegg/Hegau und zur Biostratigraphie der europaischen Hipparion Fauna. Schr. Ver. Gesch. Naturgesch. Baar Angrenzenden Landesteile Donaueschingen, 24 : 208-223.

les faciks marins en Europe et Afrique du Nord. C.R. Congr. SOC. Sav., Paris, Colloque sur le MiocBne, pp. 299-303.

the Philippine Sea cores, Part 2). Bull. Nat. Sci. Mus., Tokyo, 12(2): 431-450.

at Le Castella, Italy. J. Mar. Geol., 7(2): 70-77.

Shikoku, Japan. Tohoku Univ., Sci. Rep., Ser. 2 (Geol.), 5: 67-106.

detailed geophysical study. J. Geophys. Res., 76(2): 473-517.

Tobien, H., 1958. Relations stratigraphiques entre la faune mammalogique pontienne et

Tobien, H., 1970a. Subdivision of Pontian mammalian faunas. G. Geol., Ser. 2, 35(1):

Tobien, H., 1970b. Biostratigraphy in the mammalian faunas at the Pliocene-Pleistocene 191-195.

boundary in Middle and Western Europe. Palaeogeogr., Palaeoclimatol., Palaeoecol., 8: 77-93.

212

Tobien, H., 1971. Oeningian. In: G.C. Carloni, P. Marks, R.F. Rutsch and R. Selli (Edi- tors), Stratotypes of Mediterranean Neogene Stages. G. Geol., 37( 2): 135-143.

Tongiorgi, E. and Tongiorgi, M., 1964. Age of the Miocene-Pliocene limit in Italy. Nature, 201: 365.

Trevisan, L., 1958. La Terra, 1. Turin. Truyols, J. and Crusafont-Pairo, M., 1951. Caracterizaci6n de un sistema de cufias marinas

en el Mioceno del Pened6s. Inst. “Lucas Mallada” Invest. Geol., Estud. Geol., 14: 443-4 54.

Scale and San Andreas Fault Displacement. Dissertation, University of California, Berkeley, Calif., 99 pp.

Turner, D.L., 1970. Potassium-Argon dating of Pacific Coast Miocene foraminifera1 stages. In: O.L. Bandy (Editor), Paleontologic Zonation and Radiometric Dating. Geol. SOC. Am., Spec. Pap., 124: 91-129.

Turner, D.L., 1968. Potassium-Argon Dates Concerning the Tertiary Foraminifera1 Time

U.S. Geological Survey, 1970. Geological Survey Research 1969, Chapter A. U.S. Geol. Surv., Prof. Pap., 650-A: 214.

Valentine, J.W. and Lipps, J.H., 1970. Marine fossils at Rancho La Brea. Science, 169: 277.

Van Couvering, J.A., 1972. Radiometric calibration of the European Neogene. In: W.W. Bishop and J.A. Miller (Editors), Calibration of Hominoid Evolution. Scottish Aca- demic Press, Edinburgh, pp. 247-271.

Van Couvering, J.A. and Miller, J.A., 1971. Late Miocene marine and non-marine time scale in Europe. Nature, 230: 559-563.

Van Donk, J. and Mathieu, G., 1969. Oxygen isotope compositions of foraminifera and water samples from the Arctic Ocean. J. Geophys. Res., 74: 3396.

Van Eysinga, F.W.B., 1972. Geological Time Table. Elsevier, Amsterdam (wall chart). Vangengeim, E.A. nnd Sher, A.V., 1970. Siberian equivalents of the Tiraspol faunal

Van der Hammen, T., Wijmstra, T.A. and Zagwijn, W.H., 1971. The floral record of the complex. Palaecgeogr., Palaeoclimatol., Palaeoecol., 8: 197-207.

Late Cenozoic of Europe. In: K.K. Turekian (Editor), The Late Cenozoic Glacial Ages. Yale University Press, New Haven, Conn., pp. 391-424.

Van Hinte, J.E., I!. 69. The nature of biostratigraphic zones. Proc. 1r.t. Conf. Planktonic Microfossils, I s ’ , Geneva, 1967, 2. E.J. Brill, Leiden, pp. 267-276.

Van Montfrans, H %I., 1971a. Palaeomagnetic dating in the North Sea Basin. Thesis, University of Amsterdam, Princo, Rotterdam, 113 pp.

Van Montfrans, H.kM., 1971b. Paleomagnetic dating in the North Sea Basin. Earth Planet. Sci. Lett., 11: 226-235.

Van Montfrans, H.M. and Hospers, J., 1969. A preliminary report on the stratigraphical position of the Matuyama-Brunhes geomagnetic field reversal in the Quaternary sediments of the Netherlands. Geol. Mijnb., 48: 565-572.

the ecology of Foraminifera. Geol. Mijnb., N.S., 12: 26-30.

boundary in the North Sea Basin: revision of its position in the marine beds. Geol. Mijnb., 6: 627-640.

Vekua, A.K., 1972. Kvabebskaya Fauna Akchagyl’skikh Pozvorochnykh. Nauka, Moscow, 351 pp.

Vella, P., 1972. The Plio-Pleistocene boundary in New Zealand. In: M.N. Alekseev, E.A. Vangengeim, K.V. Nikiforova and I.M. Khoreva (Editors), Int. Colloq. Problem “The Boundary Between the Neogene and Quaternary”, 111. Moscow, pp. 165-183.

Vella, P. and Nicol, E.R., 1970. Chlamys delicatula and the Hautawan Stage. N.Z.J. Geol. Geophys., 13: 873-976.

Van Voorthuysen, J.H., 1957. The Plio-Pleistocene boundary in the Netherlands based on

Van Voorthuysen, J.H., Toering, K. and Zagwijn, W.H., 1972. The Plio-Pleistocene

21 3

Vella, P. and Nicol, E.R., 1971. Stratigraphic position of the New Zealand Pliocene- Pleistocene boundary. N. Z. J. Geol. Geophys., 14: 421-422.

Venzo, S., 1965. The Plio-Pleistocene boundary in Italy. Rep. INQUA Congr., 6th, Warsaw, 1961, pp. 367-392.

Venzo, S., 1968. New data on the Pliocene-Pleistocene boundary in northern Italy: Western Emily and Lombardy, Po Valley. In: R.B. Morrison and H.E. Wright Jr. (Edi- tors), Means of Correlation of Quaternary Successions. Proc. INQUA Congr., 7th, Princeton, N.J., 8: 349.

Verdenius, J.G., 1970. Neogene stratigraphy of the western Guadalquivir basin, southern Spain. Utrecht Micropaleontol. Bull., 3: 109 pp.

Vezian, A., 1865. Prodome de GQologie, 3. Pans. Vine, F.J., 1968. Magnetic anomalies associated with mid-ocean ridges. In: R.A. Phinney

(Editor), The History of the Earth’s Crust, A Symposium. Princeton University Press, Princeton, N.J., pp. 61-72.

stratigraphical studies of the East Rudolf Basin, Kenya. 11. Nature, 231 : 245-248.

D.J.W., Pratt, R.M., Schrader, H.-J., Weser, D., Wise Jr., S.W., 1971. Deep Sea Drilling Project, Leg 18. Geotimes, 16(10): 12-14.

Java and India and the Plio-Pleistocene boundary. K. Ned. Akad. Wet., Proc., Ser. B, 59: 204-210.

Von Koenigswald, G.H.R., 1962. Potassium-argon dates for the Upper Tertiary. K. Ned. Akad. Wet., Proc., Ser. B, 65: 32-34.

Von Koenigswald, G.H.R., 1970 . Hipparion as a guide fossil. G. Geol., Ser. 2, 35(1): 197-1 99.

Vondra, C.F., Johnson, G.D., Bowen, B.E. and Behrensmeyer, A.K., 1971. Preliminary

Von Huene, R., Kulm, L.D., Duncan, J.R., Ingle, J.C., Kling, S.A., Musich, L.M., Piper,

Von Koenigswald, G.H.R., 1956. Remarks on the correlation of mammalian faunas of

Warren, A.D., 1972. Luisian and Mohnian biostratigraphy of the Monterey Shale a t New- port Lagoon, Orange County, California. In: E.H. Stinemeyer (Editor), Proc. Pac. Coast Miocene Biostratigr. Symp., Pac. Sect. SOC. Econ. Paleotol. Mineral., Bakers- field, Calif., pp. 27-36.

Watkins, N.D., 1968. Short period geomagnetic polarity events in deep-sea sedimentary core. Earth Planet. Sci. Lett., 4(5): 341-349.

Watkins, N.D., 1972. Review of the development of the geomagnetic polarity time-scale and discussion of prospects for its finer definition. Geol. SOC. Am. Bull., 83: 551-574.

Watkins, N.D. and Goodell, H.G., 1967a. Confirmation of the reality of the Gilsa geomagnetic polarity event. Earth Planet. Sci. Lett., 2: 123-129.

Watkins, N.D. and Goodell, H.G., 1967b. Inconsistencies resulting from a limited study of magnetic polarity variation in submarine sedimentary cores. Trans. Am. Geophys. Union, Abstr., 48: 90.

Watkins, N.D. and Goodell, H.G., 1967c. Geomagnetic polarity change and faunal extinction in the Southern Ocean. Science, 156: 1083-1086.

Watkins, N.D. and Kennett, J.P., 1971. Antarctic bottom water: major change in velocity during the late Cenozoic between Australia and Antarctica. Science, 173: 873.

Watkins, N.D. and Kennett, J.P., 1972. Regional sedimentary disconformities and upper Cenozoic changes in bottom water velocities between Australia and Antarctica. In: D.E. Hayes (Editor), Antarctic Res. Ser. 19, Antarctic Oceanology 11: The Australian- New Zealand Sector, pp. 273-293.

Pleistocene tillites in Iceland. Geol. Rundsch., 54: 364. Wensink, H., 1965. Paleomagnetic stratigraphy of younger basalts and intercalated Plio-

West, R.G., 1968. Pleistocene Geology and Biology. John Wiley, New York, N.Y., 377 QD.

West, RG. and Banham, P.H., 1968. Short field meeting on the North Norfolk Coast.

West, R.G. and Wilson, D.G., 1966. Cromer Forest Bed Series. Nature, 209: 497-498. Report by the Directors. Proc. Geol. Assoc., 79: 493-512.

214

Wilcoxon, J.A., 1969. Tropical planktonic zones and calcareous nannoplankton correla-

Wilson, R.W., 1968. Insectivores, rodents, and intercontinental correlation of the Mio-

Winkler-Hermaden, A., 1913. Das eruptivgebiet von Gleichenberg. Jahrb. Geol. Reichs-

Winkler-Hermaden, A., 1943. Die tertiare Schichtfolge am Alpeno-stabfall und ihre

tions in part of the California Miocene. Nature, 221: 950-951.

cene. Proc. Int. Geol. Congr., 23rd, Prague, 10: 19-25.

anstalt, Wien, 63.

Beziehungen zu jener des pannonischen Beckens (Zur Stratigraphie des Jungtertiars der Donaulander). Reichsanst. Bodenforsch., Zweigst., Wien, Mitt., 6 : 67-92.

schen Beckens (Fortsetzung und Schluss). Naturwiss. Ver. Steiermark, Mitt., 81-82: Winkler-Hermaden, A., 1952. Neue Beobachtungen im Tertiarbereich des mittelsteiri-

145-166. Winkler-Hermaden, A., 1960. Der Vergleich der obermiozan-pliozanen Schichtfolgen im

Mediterranbereich mit jenen in den pannonisch-pontischen Gebieten. Verh. Comm. Neogene Mediterr., 1. Tag. Wien, 1959, pp. 225-243.

Winterer, E.L., Riedel, W.R., Bronnimann, P., Gealy, E.L., Heath, G.R., Kroenke, L., Martini, E., Moberly Jr., R., Resig, J. and Worsley, T., 1971. Initial Reports of the Deep Sea Drilling Project, VII. US. Government Printing Office, Washington, D.C., pp. iv-xix + 3-1757. (Parts 1 + 2.)

Wolfe, J.A. and Hopkins, D.M., 1967. Climatic changes recorded by Tertiary land floras in northwestern North America. In: K. Hatai (Editor), Tertiary Correlations and Climatic Changes in the Pacific. Pac. Sci. Congr., l l t h , Tokyo, 1966, pp. 67-76.

Wolfe, J.A. and Leopold, E.B., 1967. Neogene and early Quaternary vegetation of northwestern North America and northeastern Asia. In: D.M. Hopkins (Editor), The Bering Land Bridge. Stanford University Press, Stanford, Calif., pp. 193-206.

Wollin, G., Ericson, D.B., Ryan, W.B.F. and Foster, J.H., 1971. Magnetism of the earth and climatic changes. Earth Planet. Sci. Lett., 12( 1971): 175-183.

Wong, H.K. and Zarudzki, E.F.K., 1969. Thickness of unconsolidated sediments in the eastern Mediterranean. Geol. SOC. Am. Bull., 80: 2611-2614.

Wood 11, H.E., Chaney, R.W., Clark, J., Colbert, E.H., Jepsen, G.L., Reeside Jr., J.B. and Stock, C., 1941. Nomenclature and correlation of the North American continental Tertiary. Geol. Soc. Am. Bull., 52: 1-48.

Wray, J.L. and Ellis, C.H., 1965. Discoaster extinction in neritic sediments northern Gulf of Mexico. Bull. Am. Assoc. Pet. Geol., 49: 98-99.

Yeats, R.S., 1965. Pliocene seaknoll at South Mountain, Ventura Basin, California. Bull. Am. Assoc. Pet. Geol., 49(5): 526-546.

Yeats, R.S., McLaughlin, W.A. and Edwards, G., 1967. K-Ar mineral age of ash bed in Pic0 Formation, Ventura Basin, California. Bull. Am. Assoc. Pet. Geol., 51( 3 ) : 486.

Zagwijn, W.H., 1960. Aspects of the Pliocene and the Early Pleistocene vegetation in the Netherlands. Meded. Geol. Sticht., Ser. C-111, 5: 1-5.

Zagwijn, W.H., 1963a. Pleistocene stratigraphy in the Netherlands, based on changes in vegetation and climate. Verh. Ned. Geol. Mijnbouwk. Genoot., Geol. Ser., 21( 2):

Zagwijn, W.H., 1963b. Pollen-analytic investigations in the Tiglian of the Netherlands.

Zagwijn, W.H., 1974. A model theory for the Plio-Pleistocene boundary determination.

173-196.

Meded. Geol. Sticht., N. S., 16: 49-71.

In: T. Saito and L. Burckle (Editors), Late Neogene Epoch Boundaries. Micropaleon- tology Press, Am. Mus. Nat. History, New York, N.Y. (in press).

“Cromerian” in the Netherlands: pollen-analysis, palaeomagnetism and sedimentary petrology. Geol. Mijnb., 50: 41-58.

Zeuner, F.E., 1959. The Pleistocene Period; its climate, chronology and faunal succession. Hutch. Sci. Tech., Lond., 447 pp.

Zagwijn, W.H., Van Montfrans, H.M. and Zandstra, J.G., 1971. Subdivision of the

215

ADDITIONAL REFERENCES

Ambrosetti, P., Bonadonna, F.P., Ferrara, G., Fornaseri, M., Tolomeo, L. and Tongiorgi, E., 1968. Segnalazoine di un livello vulcanico delle argille plioceniche della Sabina (Lazio settentrionale). Boll. SOC. Geol. Ital., 87 : 333-340.

Berggren, W.A. and Amdurer, M., 1973. Late Paleogene and Neogene planktonic foraminiferal biostratigraphy of the Atlantic Ocean. Riv. Ital. Paleontol., 79 (3): 337-392.

Jenkins, D.G., 1971. New Zealand Cenozoic planktonic foraminifera. N. Geol. Surv. Paleontol. Bull., 42: 1-278 (Pls. 1-23, Figs. 1-2, Tables 1-58).

Jenkins, D.G. and Orr, W.N., 1972. Planktonic foraminiferal biostratigraphy of the eastern equatorial Pacific - D.S.D.P. Leg. 9. In : J.D. Hays et al., Initial Reports of the Deep Sea Drilling Project, IX. U.S. Government Printing Office, Washington, D.C., pp. 1060- 1193 (PIS. 1-41).

Kennett, J.P., 1973. Middle and Late Cenozoic planktonic foraminiferal biostratigraphy of the Southwest Pacific- D.S.D.P. Leg 21. In : R.E. Burns, J.E. Andrews et al., Initial Reports of the Deep Sea Drilling Project, XXI. U.S. Government Prin€ing Office, Washington, D.C., pp. 575-640 (pls. 1-21).

Kennett, J.P. and Watkins, N.D., 1974. Late Miocene-Early Pliocene paleomagnetic stratigraphy, paleoclimatology and biostratigraphy in New Zealand. Bull. Geol. SOC. Am., 85 (in press).

Koizumi, I., 1968. Tertiary diatom flora of Oga Peninsula, Akita Prefecture, northeast Japan. Tohoku Univ. Sci. Rep., Sec. Ser. (Geology), 40 (3): 171-225 (pls. 32-35, figs. 1-15, Charts 1-2, Tables 1-2).

Koizumi, I., 1973. The Late Cenozoic diatoms of Sites 189-193, Leg 19, Deep Sea Drilling Project. In : J.S. Creager et al., Initial Reports of the Deep Sea Drilling Project, XIX. U.S. Government Printing Office, Washington, D.C., pp. 805-856 (Pls. 1-8, Figs. 1-13, Tables 1-13).

Kurt& B., 1956. The status and affinities of Hyaena sinensis Owen and Hyaena ultima Matsumoto. Am. Mus. Novit., 1764: 48 pp.

Martini, E., 1972. Der stratigraphischen Wert von Silicoflagellaten im Jungtertiar von Kalifornien und des ostlichen Pazifischen Ozeans. Dtsch. Geol. Ges. Nachrichten, 5 : 47-49.

Drilling Project, VIII. U.S. Government Printing Office, Washington, D.C., pp. 727-775 (pls. 1-13, Figs. 1-3).

Moore, T., 1972. Mid-Tertiary evolution of the radiolarian genus Calocycletta. Micropaleontology, 18( 2) : 144-152 (Pls. 1-2).

Opdyke, N.D., Burckle, L.H., and Todd, A., 1974. The extension of the magnetic time-scale in sediments of the Central Pacific Ocean. Earth Planet. Sci. Lett., 22: 300-306.

Page, R.W. and McDougall, I., 1972. Ages of mineralization of gold and porphyry copper deposits in the New Guinea highlands. Econ. Geol., 67: 1034-1048.

Parker, F.L., 1973a. Correlations by planktonic foraminifera of some Tertiary localities in northern Italy. Micropaleontology, 19( 3) 235-238.

Parker, F.L., 197313. Late Cenozoic biostratigraphy (planktonic foraminifera) of tropical Atlantic deep-sea sections. Rev. Micropaleontol., 5( 2): 253-289 ( 3 Pls).

Perconig, E., 1971. Mise au point du stratotype de 1’Andalousien. Congr. Comm. Mediterr. Nkogkne Stratigr., 5th, Lyon (preprint).

Sanfilippo, A., 1971. Neogene radiolarians of the Mediterranean and western Pacific. In : A. Farinacci (Editor), Proceedings of the I1 Planktonic Conference Roma 1970. Edizioni Tecnoscienza, Rome, 2 : 1122-1 127.

toms, siliflagellates and calcareous nannofossils in the Mediterranean Neogene. Micro- paleontology, 19( 2) : 209-234 (Pls. 1-6).

Moore, T., 1971. Radiolaria. In: J. I. Tracey et al., Initial Reports of the Deep Sea

Sanfilippo, A., Burckle, L.H., Martini, E., and Riedel, W.R., 1973. Radiolarians, dia-

21 6

Schrader, H.J., 1973a. Cenozoic diatoms from the northeast Pacific, Leg 18. In: L.D. Kulm, R. Von Huene et al., Initial Reports of the Deep Sea Drilling Project, XVIII. U.S. Government Printing Office, Washington, D.C., pp. 673-798 (Pls. 1-26).

Denticula in Neogene North Pacific sediments. Micropaleontology, 19(4) : 417-439.

italienne. Quaternaria, 6: 391-413.

Epoch 17 - Matuyama. Trans. Am. Geophys. Union, 54 : 255 (abstract).

zones, Brunhes to Polarity Epoch 20. Earth Planet. Sci. Lett., 22: 307-319.

gene time scale. In: J.E. Hazel and E. Kauffman, New Concepts in Biostratigraphy. Paleontol. SOC. Am., Spec. Publ. (in press).

Morphologie der jungtertiaren Entwicklungsgeschichte der Ostalpen. Sitzungsber. Akad. Wien, Math.-Naturw. Kl., 132.

Schrader, H.J., 1973b. Stratigraphic distribution of marine species of the diatom

Selli, R., 1962. Le Quaternaire rnarin du versant Adriatique-Ionien de la pgninsule

Theyer, F. and Hammond, S.R., 1973. Magnetic polarity sequence and radiolarian zones,

Theyer, F. and Hammond, S.R., 1974. Paleomagnetic polarity sequence and radiolarian

Van Cowering, J.A. and Berggren, W.A., 1974. The biostratigraphical basis of the Neo-

Winkler Hermaden, A., 1924. Beziehungen zwischen Sediment und Tektonik und

late neogene (developments in palaeontology and stratigraphy)

Documents